Perfect 10, Inc.
v.
Giganews, Inc.

This case is not covered by Casetext's citator
UNITED STATES DISTRICT COURT CENTRAL DISTRICT OF CALIFORNIAMar 24, 2015
Case No. CV 11-07098-AB (SHx) (C.D. Cal. Mar. 24, 2015)

Cases citing this case

How cited

lock 14 Citing caseskeyboard_arrow_right

Summaries written by judges

Summaries

  • approving "paralegal fees at rates between $240 for a paralegal with five years' experience to $345 for a paralegal with 23 years' experience"

    Summary of this case from Chambers v. Whirlpool Corp.

Case No. CV 11-07098-AB (SHx)

03-24-2015

PERFECT 10, INC. Plaintiff, v. GIGANEWS, INC. et al. Defendants.


[IN CHAMBERS] ORDER GRANTING DEFENDANTS' MOTION FOR ATTORNEYS' FEES AND COSTS IN PART AND AWARDING DEFENDANTS $5,213,117.06 IN ATTORNEYS' FEES AND $424,235.47 IN NON-TAXABLE COSTS (DKT. NO. 644.)

Pending before this Court is Defendants' post-judgment motion for attorneys' fees and non-taxable costs. (Dkt. No. 644.) The Court took the matter under submission on January 29, 2015. (Dkt. No. 665.) After full consideration of the briefs and supporting evidence, and as stated more fully below, the Court finds Defendants are entitled to an award of attorneys' fees and non-taxable costs and GRANTS the motion IN PART. The Court AWARDS Defendants reasonable attorneys' fees in the amount of $5,213,117.06 and non-taxable costs in the amount of $424,235.47. /// /// ///

I. BACKGROUND

The full procedural history of this litigation is familiar to the parties and too voluminous to recite here. This action has involved more than 30 noticed motions, including a motion for change of venue, two motions to dismiss, three Daubert expert witness motions eight motions for summary judgment, and multiple discovery and sanctions motions. The docket in this action includes nearly 700 entries and exceeds 38,000 pages. Instead, the Court briefly summarizes Perfect 10's various claims against Defendants and the resolution of each of those claims in Defendants' favor.

Plaintiff initially brought suit against Defendants for copyright infringement, trademark infringement, trademark dilution, unfair competition, and violation of Perfect 10's right of publicity. (Dkt. No. 1.) On Defendants' first motion to dismiss, the Honorable A. Howard Matz dismissed Perfect 10's trademark claims with leave to amend, concluding Perfect 10 failed to allege how Defendants "used" Perfect 10's marks in a commercial transaction "merely by offering a search function that allows third parties to search for images using Plaintiff's marks as search terms." (Dkt. No. 97, p. 20.) As to Plaintiffs' state law claims for unfair competition and violation of the right of publicity, Judge Matz concluded Perfect 10's claims were preempted by the Communications Decency Act (47 U.S.C. §230(c)(1)), and dismissed those claims with leave to amend. (Dkt. No. 97, pp. 22-24.) In a lengthy discussion, Judge Matz further dismissed all of Perfect 10's copyright claims (for direct, vicarious, and contributory infringement) against Defendant Livewire with leave to amend, and dismissed Perfect 10's claim for direct infringement against Defendant Giganews with leave to amend. (Dkt. No. 97, pp. 5-19.)

On amendment, Perfect 10 declined to repeat its trademark claims or it state-law claims for unfair competition and the right of publicity, but reasserted copyright claims against both Defendants on theories of direct, contributory, and vicarious infringement. (Dkt. No. 105.) Defendants again moved to dismiss. (Dkt. No. 110.) The Honorable Audrey B. Collins denied Defendants' motion to dismiss on the question of direct infringement. (Dkt. No. 129, pp. 3-4.) However, in light of Judge Matz' assessment of the issue of direct infringement (with which Judge Collins agreed), Judge Collins only permitted Perfect 10's claim of direct infringement as to Giganews on Plaintiff's allegation that "Giganews itself uploads infringing content onto USENET." (Id., at pp. 3-4.) As to Livewire, Perfect 10's claim of direct infringement survived solely on the theory that Livewire "sells copies" of Perfect 10's copyrighted images "to its customers." (Dkt. No. 129, p. 4.) With respect to Perfect 10's claims of vicarious and contributory infringement, Judge Collins dismissed with prejudice as to Livewire, but denied the motion as to Giganews. (Id., at pp. 4-6.)

After considerable additional motion practice (including Perfect 10's early motion for partial summary judgment on the adequacy of its takedown notices under the Digital Millennium Copyright Act, which Judge Collins denied in full; Dkt. No. 180), the parties filed a total of seven competing motions for partial summary judgment. (Dkt. Nos. 357, 440, 449, 453, 467, 534, & 535.) On November 14, 2014, the Court granted partial summary judgment in Defendants' favor on Perfect 10's claim for direct copyright infringement (Dkt. No. 619), and in Giganews' favor on Perfect 10's claims for indirect copyright infringement. (Dkt. No. 620.) Because those two orders resolved all of Perfect 10's outstanding claims against Defendants, the Court denied the parties' remaining motions for partial summary judgment as moot (Dkt. No. 621), and entered judgment in Defendants' favor on November 26, 2014. (Dkt. No. 628.) On March 6, 2015, the Court denied Perfect 10's motion for reconsideration of its summary judgment order on indirect copyright liability after substantial briefing and oral argument. (Dkt. No. 682.) ///

Defendants now move for an award of attorneys' fees and costs pursuant to 17 U.S.C. section 505, 15 U.S.C. section 1117, and California Civil Code section 3344(a) for their successful defense of the action. (Dkt. No. 644.) Defendants request a total of $6,104,372.66 in attorneys' fees and $436,634.97 in non-taxable costs. (Id., p. 25.) Perfect 10 opposes. (Dkt. No. 650.) The Court took the matter under submission without hearing on January 29, 2015. (Dkt. No. 663.)

II. LEGAL STANDARD

A. Attorneys' Fees

In every case where a prevailing party seeks its attorneys' fees from an opponent, courts begin with "the general rule in this country that unless Congress [or a state legislature] provides otherwise, parties are to bear their own attorney's fees." Fogerty v. Fantasy, Inc. ("Fogerty I"), 510 U.S. 517, 533 (1994). Here, Defendants rely on three statutory attorneys' fee provisions to overcome the so-called "American Rule."

1. Attorneys' Fees Under the Copyright Act

The bulk of Defendants' argument for an award of attorneys' fees centers on the Copyright Act, which provides that in a copyright action, "the court in its discretion may allow the recovery of full costs by or against any party other than the United States" including "a reasonable attorney's fee to the prevailing party as part of the costs." 17 U.S.C. §505. However, an award of attorneys' fees under the Copyright Act is a matter of discretion, not of right. Fogerty I, 510 U.S. at 533. Under section 505, "[p]revailing plaintiffs and prevailing defendants are to be treated alike, but attorney's fees are to be awarded to prevailing parties only as a matter of the court's discretion." Id. at 534. In determining whether, and how extensively, courts should exercise that discretion, "[d]istrict courts should consider, among other things, [1] the degree of success obtained; [2] frivolousness; [3] motivation; [4] objective unreasonableness (both in the factual and legal arguments in the case); and [5] the need in particular circumstances to advance considerations of compensation and deterrence." Historical Research v. Carbal, 80 F.3d 377, 378 n.1 (9th Cir. 1996). However, each of those factors is merely a guidepost in the exercise of the court's "equitable discretion" and "courts are not limited to considering them." Fantasy, Inc. v. Fogerty ("Fogerty II"), 94 F.3d 553, 559 (9th Cir. 1996). Indeed, "[c]ourts have awarded costs for copyright claims based on a single factor ... ." Robinson v. Lopez, No. CV 03-3732-LGB (PLAx), 2003 WL 23162906, at *2 (C.D. Cal. Nov. 24, 2003)

Even looking to those five articulated factors, however, it is important to emphasize that a court's discretion to award attorneys' fees under the Copyright Act "is not cabined by a requirement of culpability on the part of the losing party." Fogerty II, 94 F.3d at 555. "Within this framework courts are given wide latitude" to award or deny attorneys' fees under the Copyright Act, even when a party acted reasonably or in good faith. Entertainment Research Group, Inc. v. Genesis Creative Group, Inc., 122 F.3d 1211, 1229 (9th Cir. 1997); accord Fogerty II, 94 F.3d at 556 (fee award proper under the Copyright Act despite fact that the "lawsuit was neither frivolous nor prosecuted in bad faith").

Ultimately, "[f]aithfulness to the purposes of the Copyright Act is.the pivotal criterion" under section 505. Fogerty II, 94 F.3d at 558. "The primary objective of the Copyright Act is to encourage the production of original literary, artistic, and musical expression for the good of the public." Fogerty I, 510 U.S. at 524. Although section 505 commonly accomplishes that purpose by incentivizing plaintiffs to bring suit to protect their copyrights, "[t]he primary objective of copyright is not to reward the labor of authors, but '[t]o promote the Progress of Science and useful Arts.'" Feist Publications v. Rural Telephone Svc. Co., 499 U.S. 340, 349-50. Indeed, section 505's provision for attorneys' fees to prevailing defendants in addition to prevailing plaintiffs recognizes "the important role played by copyright defendants" in advancing the objectives of the copyright act. Fogerty I, 510 U.S. at 532 n.18.

"Because copyright law ultimately serves the purpose of enriching the general public through access to creative works, it is peculiarly important that the boundaries of copyright law be demarcated as clearly as possible. To that end, defendants who seek to advance a variety of meritorious copyright defenses should be encouraged to litigate them to the same extent that plaintiffs are encouraged to litigate meritorious claims of infringement... Thus a successful defense of a copyright infringement action may further the policies of the Copyright Act every bit as much as a successful prosecution of an infringement claim by the holder of a copyright."

Id
. at 527.

2. Attorneys' Fees Under the Lanham Act

Under the Lanham Act courts "in exceptional cases may award reasonable attorney fees to the prevailing party." 15 U.S.C. §1117. "An action may be considered exceptional '[w]hen a plaintiff's case is groundless, unreasonable, vexatious, or pursued in bad faith.'" Secalt S.A. v. Wuxi Shenxi Const. Mach. Co., 668 F.3d 677, 687 (9th Cir. 2012), quoting Stephen W. Boney, Inc. v. Boney Servs., 127 F.3d 821, 827 (9th Cir.1997). As the Supreme Court recently held construing identical language under the Patent Act, "an 'exceptional' case is simply one that stands out from others with respect to the substantive strength of a party's litigating position (considering both the governing law and the facts of the case) or the unreasonable manner in which the case was litigated." Octane Fitness, LLC v. ICON Health & Fitness, Inc., ___ U.S. ___, 134 S. Ct. 1749, 1756, 188 L. Ed. 2d 816 (2014). Like in the copyright context, "the mere absence of bad faith on [the losing party's part] does not render [the prevailing party] ineligible for attorneys' fees." Secalt S.A. v. Wuxi Shenxi Const. Mach. Co., 668 F.3d at 687 (quoting opinion below and affirming). Still, attorneys' fees under the Lanham Act require something more than the Copyright Act. Although the Lanham Act may not require subjective bad faith, a defendant seeking attorneys' fees under the Lanham Act must demonstrate, at minimum, that "the plaintiff has no reasonable or legal basis to believe in success on the merits." Id. Though a claim may not find any direct support in existing law, for example, a plaintiff's claim is not objectively unreasonable for the purpose of awarding attorneys' fees under the Lanham Act where it could be considered an attempt to extend existing law." Cairns v. Franklin Mint Co., 115 F. Supp. 2d 1185, 1188 (C.D. Cal. 2000) aff'd, 292 F.3d 1139 (9th Cir. 2002).

3. Attorneys' Fees Under Cal. Civ. Code §3344(a)

California law recognizes both statutory and common law rights to publicity. Downing v. Abercrombie & Fitch, 265 F.3d 994, 1001 (9th Cir. 2001). Under California Civil Code section 3344, "[t]he prevailing party in any action [for violation of California's statutory right to publicity] shall ... be entitled to attorney's fees and costs." Cal. Civ. Code §3344(a). California Civil Code section 3344 "leaves no room for ambiguity" (Kirby v. Sega of Am., Inc., 144 Cal. App. 4th 47, 62, 50 Cal. Rptr. 3d 607, 618 (Cal. Ct. App. 2006)), and the statute "mandates an award of attorney's fees...." Love v. Associate Newspapers, Ltd. ("Love II") 611 F.3d 601, 614 (9th Cir. 2010). While California's corollary common law cause of action does not permit recovery of attorneys' fees, where a plaintiff makes claims under both section 3344 and the common law, courts will not parse a prevailing party's fees between the common law and statutory claims where the common law claims "are clearly inextricably intertwined with claims that do provide an independent basis for fees." Love v. Mail on Sunday ("Love I"), No. CV 05-7798-ABC (PJWx), 2007 WL 2709975, at *6 n.3 (C.D. Cal. Sept. 7, 2007); accord Akins v. Enter. Rent-A-Car Co. of San Francisco, 79 Cal. App. 4th 1127, 1133, 94 Cal. Rptr. 2d 448, 452 (Cal. Ct. App. 2000).

4. Attorneys' Fees Are Not Recoverable Under California's Unfair Competition Law

Defendants concede that "[a]ttorneys' fees are not generally recoverable under California's unfair competition law." (Dkt. No. 644, p. 19.) "[W]here a plaintiff sues solely under the unfair competition law, fees may not be recovered by a prevailing defendant. Walker v. Countrywide Home Loans, Inc., 98 Cal.App.4th 1158, 1180, 121 Cal. Rptr. 2d 79 (Cal. Ct. App. 2002). However, where a plaintiff asserts a claim under California's unfair competition law in addition to other claims that provide for attorneys' fees, under California law, "the trial court has discretion to apportion fees to claims not brought pursuant to that law-as long as those claims authorize attorney fees awards." Id.; accord Love I, 2007 WL 2709975, at *6 n.3. However, [a]pportionment is not required" under California law "when the issues in the fee and nonfee claims are so inextricably intertwined that it would be impractical or impossible to separate the attorney's time into compensable and noncompensable units." Graciano v. Robinson Ford Sales, Inc., 144 Cal.App.4th 140, 159, 50 Cal. Rptr. 3d 273, 289 (Cal. Ct. App.2006). ///

5. Calculating the Proper Amount of Attorneys' Fees

Although district courts enjoy "wide discretion in determining the reasonableness of an attorney's fees," the court begins from the premise that "[a]ttorney's fees are based on the 'lodestar' calculation." Brighton Collectibles, Inc. v. RK Texas Leather Mfg., No. 10-CV-419-GPC WVG, 2014 WL 5438532, at *3 (S.D. Cal. Oct. 24, 2014). "When it sets a fee, the district court must first determine the presumptive lodestar figure by multiplying the number of hours reasonably expended on the litigation by the reasonable hourly rate." Intel Corp. v. Terabyte Int'l, Inc., 6 F.3d 614, 622 (9th Cir. 1993). "This calculation provides an objective basis on which to make an initial estimate of the value of a lawyer's services." Hensley v. Eckerhart, 461 U.S. 424, 433, 103 S. Ct. 1933, 1939, 76 L. Ed. 2d 40 (1983).

The Ninth Circuit "has not mandated the lodestar method or comparison to market rates in calculating attorneys' fees under the Copyright Act ... ." Kourtis v. Cameron, 358 Fed.App'x. 863, 868 (9th Cir. 2009), citing Twentieth Century Fox Film Corp. v. Entertainment Distributing, 429 F.3d 869 (9th Cir.2005); accord Figure Eight Holdings, LLC v. Dr. Jay's, Inc., No. CV 10-7828 R AJWX, 2011 WL 5839655, at *1 (C.D. Cal. Nov. 18, 2011). However, the Court thinks the underlying premise of the lodestar method sound, and generally applies it in this action. But to the extent traditional lodestar analysis imposes more stringent limitations on the Court's discretion, the Court is not bound by those limitations, at least as they concern an award of fees under the Copyright Act.

"[T]o determine whether attorneys for the prevailing party could have reasonably billed the hours they claim to their private clients, the district court should begin with the billing records the prevailing party has submitted." Gonzalez v. City of Maywood, 729 F.3d 1196, 1202 (9th Cir. 2013). However, "a district court should exclude from the lodestar amount hours that are not reasonably expended because they are 'excessive, redundant, or otherwise unnecessary.'" Van Gerwen v. Guarantee Mut. Life Co., 214 F.3d 1041, 1045 (9th Cir. 2000). Normally, courts will conduct an "hour-by-hour analysis of the fee request, and exclude those hours for which it would be unreasonable to compensate the prevailing party." Gonzalez v. City of Maywood, 729 F.3d at 1202. However, "when faced with a massive fee application the district court has the authority to make across-the-board percentage cuts either in the number of hours claimed or in the final lodestar figure as a practical means of trimming the fat from a fee application." Gates v. Deukmejian, 987 F.2d 1392, 1399 (9th Cir.1992) (internal quotes omitted). A court that exercises its discretion to reduce the lodestar amount by an across-the-board percentage reduction typically must explain its reasons for doing so. Gonzalez v. City of Maywood, 729 F.3d at 1202. "Nevertheless, the district court can impose a small reduction, no greater than 10 percent—a 'haircut'—based on its exercise of discretion and without a more specific explanation." Moreno v. City of Sacramento, 534 F.3d 1106, 1112 (9th Cir. 2008).

As to the reasonableness of the rates charged, the "reasonable fees 'are to be calculated according to the prevailing market rates in the relevant community, regardless of whether plaintiff is represented by private or nonprofit counsel.'" Van Skike v. Dir., Office of Workers' Comp. Programs, 557 F.3d 1041, 1046 (9th Cir. 2009), quoting Blum v. Stenson, 465 U.S. 886, 895 (1984). "Generally, the relevant community" for the purpose of determining the prevailing market rate "is the forum in which the district court sits." Barjon v. Dalton, 132 F.3d 496, 500 (9th Cir. 1997). Absent evidence that "local counsel was unavailable, either because they are unwilling or unable to perform because they lack the degree of experience, expertise, or specialization required to handle properly the case," courts rarely look to other legal markets to determine the appropriate prevailing rate. Gates v. Deukmejian, supra, 987 F.2d at 1405. However, courts do not limit their analysis of a prevailing market rate the specific practice area at issue in the litigation. "[T]he proper scope of comparison...extends to all attorneys in the relevant community engaged in equally complex Federal litigation, no matter the subject matter." Prison Legal News v. Schwarzenegger, 608 F.3d 446, 455 (9th Cir. 2010) (internal quotes omitted). That is, courts determine the reasonableness of a rate based upon "the rates prevailing in that district for similar services by lawyers of reasonably comparable skill, experience and reputation," irrespective of practice area. Id., quoting Blum v. Stenson, supra, 465 U.S. at 895 n.11. "Unless counsel is working outside his or her normal area of practice," evidence that a billing rate was the usual rate the attorney charges for his or her services is evidence that the rate is comparable to the market rate. Moore v. James H. Matthews & Co., 682 F.2d 830, 840 (9th Cir. 1982). This is because "[b]illing rates usually reflect, in at least a general way, counsel's reputation and status (i.e., as partner, associate, or law clerk)." Id.; accord Kourtis v. Cameron, 358 Fed.App'x. 863, 868 (9th Cir. 2009) ("The district court's calculation of an attorney's fee award ... based on the actual rates charged by [prevailing party's] attorneys was reasonable under 17 U.S.C. § 505."). However, evidence of an attorney's normal billing rate is not necessarily dispositive. "Attorney affidavits regarding prevailing fees in the community and rate determinations in other cases" may also provide "satisfactory evidence of the prevailing market rate." Lakim Indus., Inc. v. Linzer Products Corp., No. 2:12-CV-04976 ODW, 2013 WL 1767799, at *8 (C.D. Cal. Apr. 24, 2013); accord United Steelworkers of Am. v. Phelps Dodge Corp., 896 F.2d 403, 407 (9th Cir.1990) (same).

Once calculated, "[t]he lodestar amount is presumptively the reasonable fee amount" and may only be adjusted upward or downward by applying a multiplier in "rare" or "exceptional" cases where "the lodestar amount is unreasonably low or unreasonably high." Van Gerwen v. Guarantee Mut. Life Co., supra, 214 F.3d at 1045. "[T]he fee applicant bears the burden of establishing entitlement to an award and documenting the appropriate hours expended and hourly rates." Hensley v. Eckerhart, supra, 461 U.S. at 437. However, a prevailing party "is not required to record in great detail how each minute of his time was expended." Id., at 437 n.12. The prevailing party seeking attorneys' fees need only "identify the general subject matter of his time expenditures" to meet its burden of establishing its fee request is reasonable. Id. This limited obligation reflects the broader policy that a "request for attorney's fees should not result in a second major litigation." Id., at 437.

Neither party urges the Court to apply an upward or downward multiplier, and the Court does not address the propriety of a multiplier in this case under the factors articulated in Kerr v. Screen Guild Extras, Inc., 526 F.2d 67, 70 (9th Cir.1975) except to the extent some of those factors have been subsumed into the lodestar calculation itself. See Van Gerwen v. Guarantee Mut. Life Co., 214 F.3d at 1047 n.2; Morales v. City of San Rafael, 96 F.3d 359, 364 (9th Cir. 1996).

B. Non-Taxable Costs

The Copyright Act also affords courts discretion to award a prevailing party "recovery of full costs." 17 U.S.C. §505. The statute's use of the term "full costs" means that "district courts may award otherwise non-taxable costs, including those that lie outside the scope of [28 U.S.C.] § 1920, under § 505." Twentieth Century Fox Film Corp. v. Entertainment Distributing, 429 F.3d 869, 885 (9th Cir. 2005). In addition to regular taxable costs, allowable costs under section 505 include costs for service of process, depositions expenses, copying, computer assisted legal research, expert witness fees, and travel costs. See, e.g., Kourtis v. Cameron, supra, 358 Fed.App'x. at 868 (expert witness fees allowable as non-taxable costs under §505); ExperExchange, Inc. v. Doculex, Inc., No. C-08-03875 JCS, 2010 WL 1881484, at *12 (N.D. Cal. May 10, 2010) (expert fees and travel costs allowable as costs under §505); Yue v. Storage Tech. Corp., No. C07-05850 JW, 2008 WL 4185835, at *6 (N.D. Cal. Sept. 5, 2008) (allowing "legal research and courier fees" as costs under §505); Atl. Recording Corp. v. Andersen, No. CV 05-933 AC, 2008 WL 2536834, at *19 (D. Or. June 24, 2008) ("In this circuit, charges for computer-assisted legal research are recoverable as costs if they are ordinarily billed to clients separately."), citing Trustees of Const. Industry and Laborers Health and Welfare Trust v. Redland Ins. Co., 460 F.3d 1253, 1258-1259 (9th Cir.2006). However, general costs of doing business "that should be subsumed in a form's overhead," such as staff overtime, are generally disallowed under section 505. See Berry v. Hawaiian Express Service, Inc., No. 03-00385 SOM-LEK, 2006 WL 4102120, at *16.

III. DISCUSSION

The Court examines Defendants' request for attorneys' fees and costs separately.

A. Defendants' Are Entitled to an Award of Attorneys' Fees Under the Copyright Act and Cal. Code Civ. Proc. 3344(a)

1. Attorneys' Fees Under the Copyright Act

The bulk of both parties' arguments centers on Defendants' request for an award of attorneys' fees under the Copyright Act. This is unsurprising given that Perfect 10 has only pursued copyright claims since March 8, 2013 when Judge Matz dismissed Perfect 10's trademark and state-law claims, and the vast majority of litigation in this action occurred solely on Perfect 10's copyright causes of action. Perfect 10 does not dispute Defendants were "prevailing parties" under 17 U.S.C. §505. Instead, Perfect 10 asserts it would be inappropriate to award Defendants attorneys' fees under the Copyright Act in light of the Fogerty factors outlined above. Though not an exclusive list of factors for the Court to consider, the Court addresses each of the Fogerty factors in turn. /// /// /// ///

a. Degree of Success Obtained

This factor weighs heavily in support of an award of attorneys' fees under the Copyright Act. On each of Perfect 10's three theories of copyright infringement, Defendants won unqualified victories: Livewire defeated Perfect 10's claims of secondary infringement at the pleading stage without leave to amend, and Defendants won on each of Perfect 10's remaining claims on summary judgment. This sort of complete victory on the merits is significant. "A successful defense furthers the purposes of the Copyright Act just as much as a successful infringement suit does." Inhale, Inc. v. Starbuzz Tobacco, Inc., 755 F.3d 1038, 1043 (9th Cir. 2014). This is particularly so where, as here, Defendants' successful defense of the action helped "demarcate" the "boundaries of copyright law." Fogerty I, 510 U.S. at 527. Moreover, the Usenet services Defendants offer "serv[e] to foster an exchange of information or opinion on a particular topic running the gamut from, say, the music of Wagner to Balkan politics to AIDS prevention to the Chicago Bulls." Reno v. ACLU, 521 U.S. 844, 851 (1997). Perfect 10's unmeritorious claims against the leading Usenet service provider in the country "posed a serious threat to the public's access to free and competitive expression." Mattel, Inc. v. MGA Entm't, Inc. ("Mattel I"), No. CV 04-9049 DOC RNBX, 2011 WL 3420603, at *4 (C.D. Cal. Aug. 4, 2011) aff'd sub nom. Mattel, Inc v. MGA Entm't, Inc. ("Mattel II"), 705 F.3d 1108 (9th Cir. 2013) That sort of "successful defense of a copyright infringement action" advances the purpose of the Copyright Act and Defendants' should be "encouraged to litigate" their "meritorious copyright defenses." Fogerty I, 510 U.S. at 527; see also Mattel I, 2011 WL 3420603, at *4 (even in absence of frivolous or unreasonable claims, successful defense furthered the interests of the copyright act by defeating an "over-aggressive copyright prosecution" exemplified by "pursuit of grossly overbroad monetary and injunctive relief.")

The extent to which this factor weighs in Defendants' favor would alone support an award of attorneys' fees. Robinson v. Lopez, supra, 2003 WL 23162906, at *2; see also Los Angeles News Service v. Reuters Television Int'l, Ltd., 149 F.3d 987, 997 ("fees are warranted inasmuch as [defense of the action] served the purposes of the Copyright Act for [defendant] to establish its right to extraterritorial damages and to defend its favorable ruling below on fair use" under the Copyright Act). However, the Court need not determine whether it would award attorneys' fees in this case on the degree of success alone because a number of other factors weigh in favor of an award of attorneys' fees under the Copyright Act.

b. Frivolousness

"A claim is not frivolous" under the Copyright Act "merely because it is unsuccessful." Bisson-Dath v. Sony Computer Entm't Am. Inc., No. CV-08-1235 SC, 2012 WL 3025402, at *2 (N.D. Cal. July 24, 2012). Rather, a frivolous claim under the Copyright Act is one that, in either the factual or legal assertions (Perfect 10, Inc. v. CCBill LLC ("CCBill"), 488 F.3d 1102, 1120 (9th Cir. 2007)), is "clearly baseless," involving "fantastic or delusional scenarios." Perfect 10, Inc. v. Visa Int'l Serv. Ass'n, No. C 04-00371 JW, 2005 WL 2007932, at *4 (N.D. Cal. Aug. 12, 2005), quoting Neitzke v. Williams, 490 U.S. 324, 325-28 (1989). Put another way, "[a] case is deemed frivolous only when the "result is obvious or the arguments are wholly without merit." Glass v. Sue, No. CV 09-8570-RGK SHX, 2011 WL 561028, at *3 (C.D. Cal. Feb. 8, 2011), citing Karam v. City of Burbank, 352 F.3d 1188, 1195 (9th Cir. 2003).

The Court does not find Perfect 10's claims of vicarious and contributory copyright infringement against Defendants were so obviously without merit as to be frivolous. Although the Court ultimately found those claims without merit, it only did so after two rounds of motions to dismiss (as to Livewire) and two rounds of summary judgment (as to Giganews), involving difficult and somewhat novel questions of law. See Int'l Bhd. of Teamsters, Chauffeurs, Warehousemen & Helpers of Am., AFL-CIO, Local 631 v. Silver State Disposal Serv., Inc., 109 F.3d 1409, 1412 (9th Cir. 1997) (noting an appeal "is less likely to be considered frivolous" under Fed. R. App. P. 38 "when there is very little case law directly apposite.") (internal quotes omitted); see also Smith v. Jackson, 84 F.3d 1213, 1221 (9th Cir. 1996) (holding district court properly exercised discretion in finding copyright claims non-frivolous where "the claims had a legal basis sufficient to survive summary judgment and a factual basis supported by expert testimony").

Although a somewhat closer call, the Court does not find Perfect 10's claims of direct copyright infringement were frivolous, either, at least at the outset. It is true that Perfect 10's claims of direct infringement as to Giganews survived the pleadings stage solely on a theory that Giganews' employees personally uploaded infringing content onto Giganews' servers. (Dkt. No. 129, pp. 3-4.) But that was only after the Court's lengthy legal analysis on an admittedly unsettled (at the time) question of law - whether direct infringement requires volitional conduct. (Dkt. No. 97, pp. 7-13; 129, pp. 3-4.) As to Livewire, Perfect 10's claim of direct infringement survived on the theory that Livewire "sells copies" of Perfect 10's copyrighted images "to its customers." (Dkt. No. 129, p. 4.) Although Perfect 10 failed to substantiate this theory in discovery, Defendants do not attempt to argue that the factual allegation was "fantastic or delusional" at the outset. As discussed more fully below, the Court does conclude that Plaintiff's insistence on pursuing its direct infringement claims after the close of discovery indicates its motivation for bringing and maintaining suit, but it does not find Perfect 10's claims were frivolous or wholly without merit when Perfect 10 brought suit.

However, the fact that Perfect 10's copyright claims were not frivolous when it filed them in 2011 does not undermine the Court's ultimate conclusion that attorneys' fees would promote the purpose of the Copyright Act and are appropriate in this case. "[A] finding of frivolity is not essential to award attorneys' fees...." DuckHole Inc. v. NBCUniversal Media LLC, No. CV-12-10077-BRO, 2013 WL 5797204, at *2 (C.D. Cal. Oct. 25, 2013) (declining to address the "frivolousness" factor at all because such a finding was unnecessary for the decision); accord Mattel II, 705 F.3d at 1111 (denying the plaintiff's attempt to "resurrect the long-rejected requirements of frivolousness and bad faith" for an award of fees under the Copyright Act).

c. Motivation

This third Fogel factor weighs in favor of an award of attorneys' fees. As the Court observed in both its order granting Giganews' summary judgment motion on indirect copyright infringement and in its order denying Perfect 10's motion, Perfect 10's undisputed conduct in this action has been inconsistent with a party interested in protecting its copyrights. (Dkt. Nos. 620, p. 13; 682, pp. 21-22.) All of the evidence before the Court demonstrates that Perfect 10 is in the business of litigation, not protecting its copyrights or "stimulat[ing] artistic creativity for the general public good." Fogerty I, 510 U.S. at 527.

Perfect 10 has never been a self-sustaining business, and to date, has lost more than $50 million dollars, if not more. (Dkt. No. 644-15, pp. 159:7-21, 161:16-23.) However, this loss appears to be largely intended by Perfect 10's President and CEO Norman Zada, who described Perfect 10 to non-party Rebekah Chaney as a "tax write-off." (Dkt. No. 644-18, pp. 49:25-50:1.) Because Perfect 10 is a closely held corporation, Zada testified he is able to deduct Perfect 10's significant losses on his personal income taxes. (Dkt. No. 644-17, p. 248:1-11.) In fact, Zada told Ms. Chaney that he "needed [Perfect 10] to offset money he made in the market" and "needed the loss to represent how small businesses couldn't make money because of piracy on the Internet." (Dkt. No. 644-18, p. 51:4-8.)

Rather than bringing suit for the purpose of protecting its copyrights and stimulating artistic creativity, the evidence reveals that Zada's interest in the copyrights held by his "tax write-off" is solely in litigation. In deposition, for example, Perfect 10's President and CEO Norman Zada testified that, to date, it has filed between 20 and 30 copyright infringement lawsuits. (Dkt. No. 644-15, pp. 10:19-11:7.) In the life of the company, more than half of Perfect 10's revenues have been generated by litigation. (Id., 161:24-162:5.) However, all of those revenues were generated by settlements and defaults -Perfect 10 has never obtained a judgment in a contested proceeding in any of its roughly two dozen copyright lawsuits. (Id., at p. 166:4-17.) Similarly, litigation expenses make up the largest share of Perfect 10's expenses, which are on par with, if not greater than, Perfect 10's personnel expenses. (Id., at p. 162:8-23.) In his capacity as President and CEO, Zada spends "eight hours a day," 365 days a year on litigation (Dkt. No. 644-16, p. 541:1-2), "working on various court cases that [Perfect 10] ha[s] going on." (Dkt. No. 644-15, p. 84:1-4.) Indeed, Zada admitted that, in the past, Perfect 10 has expressly purchased copyrights from other copyright holders "because [Perfect 10] thought they would be helpful in [its] litigation efforts ... ." (Id., at pp. 39:25-40:1.)

The evidence before the Court also demonstrates Perfect 10 continued that pattern in this litigation, which, as the Court previously noted, has been inconsistent with that of a plaintiff interested in actually protecting its copyrights from unauthorized use. (See Dkt. Nos. 620, p. 13; 682, pp. 21-22.) Perfect 10 has a long, documented history of sending service providers inadequate takedown notices under the DMCA that fail to identify specific infringing material, and then bringing suit for the service providers' failure to respond to deficient DMCA takedown notices. See, e.g., CCBill, 488 F.3d at 1111-13 (concluding Perfect 10's takedown notices were deficient under DMCA); Dkt. No. 180 (Judge Collins holding Perfect 10's takedown notices in this action inadequate under DMCA); Perfect 10 v. Google, Inc., No. CV 04-09484-AHM (SHx), Dkt. No. 937, pp. 12-25 (C.D. Cal. July 26, 2010) (holding the bulk of Perfect 10's takedown notices were deficient under the DMCA for failure to adequately identify infringing material). Despite numerous admonitions that its inadequate notices "unduly burden[] service providers" and reminders that "the burden of policing copyright infringement" falls "squarely on" Perfect 10 (CCBill, 488 F.3d at 113), Perfect 10 never attempted to submit a takedown notice in this action that Giganews could actually use.

In his declaration in opposition to the motion for attorneys' fees, Mr. Zada declares "Perfect 10 also obtained partial summary judgment...as to the sufficiency of its DMCA notices in Perfect 10, Inc. v. Yandex, N.V., No. C. 12-01521 WHA, 2013 WL 1899851, at *4-5. (Dkt. No. 650-8.) Repeating a pattern this Court already discussed in its order denying Perfect 10's motion for reconsideration (see Dkt. No. 682), Perfect 10 fails to cite the Yandex case anywhere in its opposition brief. (See Dkt. No. 650, pp. iii-vi (Perfect 10's table of authorities). In any event, the Yandex decision does not help Perfect 10's assertion that its DMCA notices in this action were adequate. The only issue before the district court in Yandex was whether a single "screenshot" takedown notice "standing alone was compliant with the" DMCA. The district court expressly noted that its order "does not rule on [the] issue" of "whether, in the aggregate, [Perfect 10's] notices complied with the DMCA...." Id., at *4; see also Id, at *5 (noting the only issue on that motion was the sufficiency of a single notice that "was only 11 pages" and included "16 images on a grid," that "required only a few seconds of effort" to review). It is true that, as here, the district court in Yandex rejected the defendant's argument that processing Perfect 10's DMCA notices "required substantial effort" and that Perfect 10 could have made the takedown process much more feasible by sending its DMCA notices as a "simple list of URLs." Perfect 10 v. Yandex N.V., 2013 WL 1899851, at *5. However, unlike here where the evidence on summary judgment showed Giganews stated both publicly on its website and privately many times to Perfect 10 that Message-IDs were essential to a takedown notice, the defendant in Yandex never "communicated [its] preference to Perfect 10." Id. Indeed, unlike the ample evidence in this action, nothing in the Yandex decision suggests the defendant publicized a reasonable set of instructions for preparing a DMCA takedown notice. Where a service provider attempts to meet its obligations under the DMCA by establishing a reasonable internal process for handling notices under the DMCA and widely publicizes the information it needs to promptly process a takedown notice, a copyright holder may not circumvent that reasonable procedure and unilaterally determine what information is sufficient to "find and address the allegedly infringing material expeditiously." Viacom Int'l, Inc. v. YouTube, Inc., 940 F.Supp.2d 110, 115 (S.D.N.Y. 2013); accord Perfect 10, Inc. v. CCBill LLC, 488 F.3d at 1113 (finding an earlier iteration of Perfect 10's takedown notices inadequate under the DMCA and "declin[ing] to shift a substantial burden from the copyright owner to the provider" by requiring the service provider to process the deficient notices); see also Luvdarts, LLC v. AT & T Mobility, LLC, 710 F.3d 1068, 1071 (9th Cir. 2013) (holding in the context of vicarious infringement that "right and ability to supervise should be evaluated in the context of a system's current architecture" rather than the architecture a copyright holder thinks the service provider should have).

Indeed, even after Perfect 10 admittedly learned of a method to produce a takedown notice in "15 minutes" that would result in almost immediate removal of "90 percent of the Perfect 10 content on Giganews's servers," Perfect 10 refused to do so. (Dkt. No. 644-14, pp. 207:15-208:13; see also Dkt. Nos. 441, ¶46 & 537, ¶46 (admitting Perfect 10 "is aware of and has used software that allows it to extract thousands of Message-IDs for messages it believes to be infringing in about 10 seconds"); Dkt. No. 325, ¶18 (same); Dkt. No. 443-2, pp. 104:18-23, 132:1-6 (admitting that it would be "very easy for Perfect 10 to collect message IDs to put into a DMCA notice" and had done so "for approximately 54,000 Perfect 10 messages"). When pressed as to why Perfect 10 continued to refuse to supply the sort of DMCA notice that Judge Collins had already concluded was necessary in this action (i.e., one that included machine-readable Message-IDs), Zada stated "that is simply helping [Defendants] actually remove the material." (Dkt. No. 443-2, p. 181:9-11.) Perfect 10's lack of interest in "helping ... actually remove the material" from Giganews' servers is all the more salient in light of Perfect 10's admission that it has never taken any substantial steps to avoid copyright piracy from its own website. Perfect 10 has never taken any steps to prevent the images it presents on its website from unauthorized use or instituted any security measures (such as a digital watermark) to identify or track individuals who download images from Perfect 10's website and subsequently upload those photos onto the internet. (Dkt. No. 644-15, pp. 76:18-79:21.)

In opposition, Perfect 10 merely identifies one prior case where the trial court determined it could not "determine with any certainty the motivations involved" in Perfect 10 bringing that earlier action. Perfect 10, Inc. v. CWIE, LLC, No. CV 02-7624-LGB (SHx), 2005 WL 5957973, at *2 (C.D. Cal. Feb. 10, 2005). In that case, the Court noted evidence that Perfect 10's "primary source of income comes from litigation," but also pointed to a declaration from Mr. Zada that "Perfect 10 loses money on copyright litigation and that Perfect 10's motivation is to stop infringement of its images." Id. Notably, however, there was no evidence before the court in that case that losses from litigation actually inure to Zada's benefit. Unfortunately for Perfect 10, the same cannot be said here, where there is direct evidence before the Court that Zada intends for Perfect 10 to operate at a loss for tax purposes. More importantly, however, the Court's primary evidence of motivation in this case is Perfect 10's own conduct in this litigation. Unlike CWIE, LLC, there is ample evidence before the Court that Perfect 10 pursued this litigation for reasons inconsistent with the purpose of the Copyright Act, and this factor weighs in favor of an award of attorneys' fees.

In addition to Perfect 10's refusal to take any of the numerous steps available to it to protect its copyrights, Perfect 10 litigated this action in an unnecessarily litigious manner that was guaranteed (if not designed) to drive up the costs of litigation. One obvious example is Perfect 10's insistence on pursuing its direct liability claims after the close of discovery despite the fact that it was unable to produce a single item of evidence on the only theories of direct liability remaining in the case. Instead, Perfect 10 continued to advance those claims on theories already rejected by the Court and for which Perfect 10 never sought reconsideration. (See Dkt. No. 619, pp. 11-16.) Rather than test the arguments it thought viable on appeal, Perfect 10 forced dueling summary judgment motions on theories of direct liability that were unsupported by any evidence at the close of discovery. The record it replete with similar unnecessary motion practice and obstinate behavior in discovery, that would be too lengthy to summarize here. The Court directs the reader's attention to Defendants' more complete summary in their motion. (Dkt. No. 644, pp. 3-10.)

d. Objective Unreasonableness

Courts assess the objective reasonableness of a copyright litigant's claims "both in the factual and in the legal components of the case... ." Fogerty I, 510 U.S. at 534 n.19 (internal quotes omitted). However, courts have not articulated precisely how this factor differs from the question of whether a claim was "frivolous." See, e.g., CCBill, 448 F.3d at 1120 (discussing issues of objective unreasonableness and frivolity without differentiating between the two); see also Dkt. No. 644, pp. 12-14 (Defendants discussing frivolousness and objective unreasonableness largely interchangeably; Dkt. No. 650, at pp. 7-12 (same as to Perfect 10). As with frivolity, "the mere fact that [a party] lost cannot establish his objective unreasonability." Seltzer v. Green Day, Inc., 725 F.3d 1170, 1181 (9th Cir. 2013). Still, the standard for objective unreasonableness is somewhat lower than that for complete frivolity. While frivolity requires evidence that a party's claims were "clearly baseless" involving "fantastic or delusional scenarios" (Perfect 10, Inc. v. Visa Int'l Serv. Ass'n, supra, 2005 WL 2007932, at *4), a claim is objectively unreasonable where the party advancing it "should have known from the outset that its chances of success in this case were slim to none." SOFA Entertainment, Inc. v. Dodger Prods., Inc., 709 F.3d 1273, 1280 (9th Cir. 2013); compare Glass v. Sue, supra, 2011WL 561028, at *3 (describing "frivolous" claim as one where "the result is obvious or the arguments are wholly without merit).

Although the standard is slightly lower for objective unreasonableness, the Court is not persuaded that Perfect 10's decision to advance its claims for contributory and vicarious infringement were so beyond the pale as to be objectively unreasonable. And while Perfect 10's conduct in advancing those claims may have been questionable at times, that conduct goes more to the factor of motivation than the objective unreasonableness of Perfect 10's factual and legal claims. (See, e.g., section III.A.1.c and footnote 3, above.) Nor does the Court find Perfect 10's decision to bring its direct infringement claims was objectively unreasonable at the outset of the litigation. Again the issue of "volitional conduct" was at least open to reasonable dispute at the time Perfect 10 brought suit. However, once that Judge Matz conclusively answered that question (Dkt. No. 97), all that remained of Perfect 10's direct infringement claims were two factual assertions that were questionable at best: that Giganews' employees directly uploaded infringing Perfect 10 content to the Usenet and that Livewire directly sold access to Perfect 10 images. (Dkt. No. 129, pp. 3-4.) Despite years of discovery, Perfect 10 failed to identify a single item of evidence that would even tangentially support either allegation, and it was objectively unreasonable for Perfect 10 to pursue its narrowed claim of direct infringement as to either Giganews or Livewire beyond that point. Entertainment Research Group, Inc. v. Genesis Creative Group, 122 F.3d 1211, 1229 (9th Cir. 1997) (party acted objectively unreasonably by pursuing claims for which it produced "no evidence at all...even after three years and numerous depositions"). Though unnecessary to an award of attorneys' fees in light of the Court's other findings, this factor weighs slightly in Defendants' favor.

e. Considerations of Compensation and Deterrence

The last Fogerty factor weighs in favor of an award of attorneys' fees under section 505. This aspect of the Fogerty analysis recognizes that "[d]eterring non-meritorious lawsuits against defendants seen as having 'deep pockets' and compensating parties that must defend themselves against meritless claims are both laudible ends." Scott v. Meyer, No. CV 09-6076 ODW(RZX), 2010 WL 2569286, at *3 (C.D. Cal. June 21, 2010). When looking at this factor in the context of a successful defendant, courts must be careful not to "discourage 'starving artists' from defending copyrights in original works due to the threat of attorney's fees." Brod v. Gen. Pub. Grp., Inc., 32 F. App'x 231, 236 (9th Cir. 2002); accord Reinicke v. Creative Empire LLC, No. 12CV1405-GPC KSC, 2014 WL 5390176, at *6 (S.D. Cal. Oct. 22, 2014) (declining to exercise discretion to award attorneys' fees to a defendant under section 505 because court did "not want to discourage individual plaintiff language developers from asserting their copyright ownership rights"). Perfect 10 is not a "starving artist" in need of protection from an award of attorneys' fees - it is a serial plaintiff operating on a self-proclaimed business model of litigation. Although Mr. Zada admitted to spending eight hours per day, 365 days per year on litigation for Perfect 10, he spends only 40 to 50 hours per year attempting to create new artistic content. (Dkt. No. 644-16, p. 540:11-541:5.) Other Perfect 10 employees similarly testified that they spend the vast majority of their time working exclusively on litigation efforts. (Dkt. Nos. 655-15, pp. 14:8-16:2; 655-17, pp. 22:15-23:10, 79:10-80:10, 157:5-158:1; 655-18, pp. 106:19-107:22.)

This effort to create new content is contrasted with Perfect 10's efforts to purchase copyrights to pre-existing images, which Perfect 10 admits it has done, at least in part, to advance its litigation interests in specific litigation. (Dkt. No. 644-15, pp. 38:19-42:17.) But even where Perfect 10 has purchased the copyrights to photographs outside the context of specific litigation, Zada admitted in deposition that Perfect 10 generally purchases copyrights only after it determines that the image has already been the subject of copyright infringement on the internet. (Id.) Moreover, Zada stated that such purchases composed less than four percent of Perfect 10's total lifetime expenses.

In light of Perfect 10's well-documented improper motive in bringing suit (see section III.A.1.c, above), the Court has little concern that an award of attorneys' fees in this action will discourage "starving artists" from protecting their copyrights. If anything, it will discourage serial litigants from bringing unmeritorious suits and then unnecessarily driving up litigation costs in order to drive a settlement. Such a result is entirely consistent with the purpose of the Copyright Act, and this factor weighs in favor of an award of attorneys' fees.

f. Financial Strength

Although not expressly identified as one of the original Fogerty factors, courts in this circuit also consider "whether the chilling effect of attorney's fees may be too great or impose an inequitable burden on an impecunious plaintiff." Ets-Hokin v. Skyy Spirits, Inc., 323 F.3d 763, 766 (9th Cir. 2003); accord Perfect 10, Inc. v. CWIE, LLC, supra, 2005 WL 5957973, at *3 ("the relative financial strength of the parties, and whether the amount requested is excessive in light of the losing party's resources" are factors to consider). Mr. Zada declares "Perfect 10 currently is in substantial debt," that it "has annual revenues of approximately $40,000 per year," and "will be forced to shut down" and "declare bankruptcy" if the Court exercises its discretion to award Defendants attorneys' fees.

The authorities Perfect 10 cites, however, do not suggest it is improper to award attorneys' fees simply because the losing party is impecunious. Rather, they make clear the court should consider whether, in light of the losing party's financial condition, it would be "inequitable" to do so. Ets-Hokin v. Skyy Spirits, Inc., 323 F.3d at 766 (emphasis added). Under the circumstances, the Court finds nothing inequitable in awarding Defendants attorneys' fees. See also Mattel, Inc. v. Walking Mountain Prods., supra, 353 F.3d 792, 816 ("Under the Copyright Act, the question is whether a successful defense of the action furthered the purposes of the Act, not whether a fee award would do so.") Perfect 10 admits that it has likely "never been solvent" in more than 15 years of operation. (Dkt. No. 644-15, p. 159:19-21.) Indeed, Perfect 10 has repeatedly reported that it was on the verge of bankruptcy. See, e.g., Perfect 10, Inc. v. Google, Inc., 653 F.3d 976, 981 (9th Cir. 2011) (noting Perfect 10's argument the same year this action was filed that Perfect 10 was "very close to bankruptcy"). That is, despite the fact that Perfect 10's primary business is copyright litigation, Perfect 10 effectively argues that it could never be subject to any attorneys' fee award under the Copyright Act because it is perpetually in debt and on the verge of bankruptcy. The Court is not persuaded, particularly where, as here, the evidence suggests Perfect 10's impecunity is intentional. (See Dkt. No. 644-18, p. 51:4-8.)

The Court's decision to award attorneys' fees under the Copyright Act in this action is driven in part by Perfect 10's own conduct (as is the size of the fee award), and the Court finds nothing inequitable about imposing it. That Perfect 10 has heretofore avoided this consequence is also immaterial to the question at hand.

g. The Court Exercises its Discretion to Award Defendants Attorneys' Fees Under the Copyright Act

Having weighed all of the various Fogerty factors, as well as the other factors articulated by the Ninth Circuit and the totality of the other circumstances in this action, the Court finds an award of attorneys' fees in this action would promote the purposes of the Copyright Act. The Court exercises its discretion to award Defendants (as the prevailing parties in this action) "a reasonable attorney's fee" for the costs of reasonably defending against Perfect 10's copyright claims. 17 U.S.C. §505.

2. Attorneys' Fees Under the Lanham Act

Defendants also seek an award of attorneys' fees under the Lanham Act, which Defendants estimate constituted 35 percent of their work on this case up to the point when Judge Matz dismissed the trademark claim on March 8, 2013. (Dkt. No. 644-1, ¶56.) To award attorneys' fees under the Lanham Act, the Court must find Perfect 10's trademark claims presented an "exceptional case" (15 U.S.C. §1117), such that Perfect 10's trademark claims were "groundless, unreasonable, vexatious, or pursued in bad faith." Secalt S.A. v. Wuxi Shenxi Const. Mach. Co., supra, 668 F.3d at 687. That is, Perfect 10's trademark claim must have "st[ood] out from others with respect to the substantive strength of a [Perfect 10's] litigating position (considering both the governing law and the facts of the case) or the unreasonable manner in which the case was litigated." Octane Fitness, LLC v. ICON Health & Fitness, Inc., supra, 134 S. Ct. at 1756.

Perfect 10 does not dispute Defendants' allocation of 35 percent of Defendants' pre-March-2013 fees to the trademark claim. Instead, Perfect 10 argues that the total number of hours Defendants claim is excessive. The Court accepts the undisputed 35 percent allocation, and separately addresses the reasonableness of defense counsels' overall claimed hours below. (See Section III.A.2.b, ante.)

Admittedly, Perfect 10's claims for trademark infringement under the Lanham Act were particularly strong. As Judge Matz noted, Perfect 10's Lanham Act claim for direct infringement in the original complaint was premised on three allegations:

(1) Defendants "copied and sold" images on its servers bearing Perfect 10's marks in the title of the digital file
(2) Defendants "allow[ed] their users to use Perfect 10 marks to conduct searches, and;
(3) "Defendants provide search results containing Perfect 10's marks in the title" when a user searched for one of Perfect 10's marks.

(Dkt. No. 97, pp. 19-20 (citing Dkt. No. 1, ¶¶24, 29, 54).) Judge Matz rejected Plaintiff's claim for direct trademark infringement, concluding allegations Defendants "merely offer[ed] a search function that allows third parties to search for images using Plaintiff's marks as search terms" did not constitute a commercial use of Perfect 10's marks under the Lanham Act. (Id., at p. 20.) Nor, Judge Matz held, did such allegations suggest any third party had "used" Perfect 10's marks in a commercial transaction simply by uploading the images for others to view for free. (Id., at 20-21.) Finally, Judge Matz held, Perfect 10 failed to allege any trademark dilution because Perfect 10 failed to allege Defendants "'used' Plaintiff's protected marks." (Id., at pp. 21-22.)

Despite those deficiencies in Perfect 10's allegations of fact, however, Judge Matz also concluded Perfect 10's Lanham Act claims were not so far-afield as to deny leave to amend. (Dkt. No. 97, p. 24.) On amendment, Perfect 10 reasonably elected to drop its unmeritorious Lanham Act claim rather than pursue it further. (See Dkt. No. 105 (Perfect 10's First Amended Complaint).) It may be that Perfect 10's trademark claims were "outside the limits of any previous ... trademark decisions," but that fact alone does make Perfect 10's trademark claims an "exceptional case." Perfect 10, Inc. v. Visa Int'l Serv. Ass'n, supra, 2005 WL 2007932, at *6. A trademark claim is not "unreasonable" or "vexatious" simply because it is novel. Id. ("complex and novel legal issues related to the Internet cannot be considered as a groundless or unreasonable only because they do not meet the current legal standards"). That Perfect 10 declined to pursue its Lanham Act claims after Judge Matz determined Perfect 10's they were unavailing only supports the conclusion that Perfect 10 acted reasonably in pursuing its trademark claims. There is nothing about Perfect 10's conduct with respect to its trademark claims that make this action "exceptional," and it would be inappropriate to award Defendants' attorneys' fees under 15 U.S.C. section 1117.

3. Attorneys' Fees Under Perfect 10's State Law Claims

Defendants further seek an award of attorneys' fees under California Civil Code section 3344(a) for expenses incurred defending against Perfect 10's statutory right of publicity claim. Perfect 10 does not oppose an award of attorneys' fees on this claim, and such an award is mandatory under California law, not discretionary. Love II, supra, 611 F.3d at 614. Nor does Perfect 10 oppose Defendants contention (see Dkt. No. 644, pp. 18-19) that Perfect 10's statutory and common law right of publicity claims were so "inextricably intertwined" that it would be impossible to differentiate fees associated solely with the statutory claim (for which fees are mandatory) from the common law claim (for which fees are unavailable). Love I, 2007 WL 2709975, at *6 n.3. Moreover, Perfect 10 does not dispute Defendants' allocation of 15 percent of Defendants pre-March-2013 fees to the publicity claims.

As with the trademark claim (see note 6, above) Perfect 10 merely argues that Defendants spent an unreasonable amount of time litigating all of the initial motion practice in this case. (See Dkt. No. 650, p. 25.) The Court addresses this argument separately in section III.A.2.b, below.

Defendants concede that they are not entitled to an award of attorneys' fees for defending against Perfect 10's claim under California's unfair competition law. (Dkt. No. 644, p. 19.) However, Defendants urge that at least some of the time spent defending Perfect 10's unfair competition claim was inextricably intertwined with Perfect 10's defense of other claims for which fees are available. (See Dkt. No. 1, ¶65 (Perfect 10 alleging Defendants engaged in unlawful conduct for the purposes of unfair competition by violating Perfect 10's publicity rights and "selling without authorization" Perfect 10's copyright materials). To account for the portions of Perfect 10's unfair competition that were not inextricably intertwined with the claims for which fees are appropriate, Defendants reduce their fee request by five percent of the fees incurred up to March 8, 2013 when Judge Matz dismissed all of Perfect 10's non-copyright claims. (Dkt. No. 644, p. 20.) In opposition, Perfect 10 does not quarrel with Defendants' assertion that the unfair competition claim was inextricably intertwined with some claims for which fees may be appropriate or the size of Defendants' proposed reduction.

This five percent reduction is already factored into Defendants $6,104,372.66 fee request. Perfect 10 does not oppose the five percent allocation, and the Court does not address it further.

B. The Amount of Attorneys' Fees

To determine the appropriate lodestar amount, the Court must consider the reasonableness of both the rates Defendants' attorneys charged and the amount of time Defendants' attorneys spent working at those rates. Intel Corp. v. Terabyte Int'l, Inc., supra, 6 F.3d at 622. ///

1. The Rates Defendants Request are Consistent with Prevailing Market Rates

An attorney's rate is "reasonable" for the purpose of a fee award if it comports with "prevailing market rates in the relevant community, regardless of whether plaintiff is represented by private or nonprofit counsel." Van Skike v. Dir., Office of Workers' Comp. Programs, supra, 557 F.3d at 1046. To determine whether a rate falls within the scope of "prevailing market rates," courts consider rates charged by other "attorneys in the relevant community engaged in equally complex Federal litigation, no matter the subject matter" for "for similar services by lawyers of reasonably comparable skill, experience and reputation." Prison Legal News v. Schwarzenegger, 608 F.3d at 455 (internal quotes omitted).

Defendants request reimbursement for attorney, paralegal, and support staff time from its legal team at the law firms of Fenwick & West and Winston & Strawn. Defendants' lead trial counsel, Andrew Bridges, is a senior partner at Fenwick & West with 29 years of experience. (Dkt. No. 644-1, ¶17.) Mr. Bridges billed his time from $825 per hour in 2011 up to $930 per hour in 2014. (Id.) Other, more junior partners on Defendants' legal team billed between $610 per hour in 2011 and $750 per hour in 2014. (Id, at. ¶¶18, 39.) For associate time, Defendants request between $350 per hour for the most junior associate in 2011 up to $690 per hour for an associate with 11 years' experience in 2014. (Id., at ¶¶21, 26.) Defendants' legal team also billed $200 for a staff attorney, $100 for a contract attorney, and between $240 and $345 for paralegal time. (Id., at ¶¶31-33, 35, 36.) The Court more fully summarizes each individual's experience, rates, and hours billed in Appendix A to this order.

Lead Defense counsel Andrew Bridges and a handful of other attorneys on this case worked at the law firm of Winston & Strawn when the litigation began in 2011. (See Dkt. No. 664-1, ¶17.) However, Mr. Bridges and a few other attorney apparently left Winston & Strawn for Fenwick & West in 2011, taking this litigation with them.

Robert Winat, a paralegal at Fenwick & West with more than 23 years of experience is the most expensive paralegal. (Dkt. No. 644-1, ¶35.) The next most expensive paralegal is Carol McCroy, who has 16 years of experience and was billed at $295 per hour in 2014. (Id., at ¶34.)

a. Attorney Rates

Perfect 10 argues Defendants' attorneys charged unreasonably high rates for its partners and associates. (Dkt. No. 650, pp. 17-21.) Perfect 10 first notes a number of intellectual property cases decided in this district in the past four years where attorneys billed at rates between $250 per hour and $530 per hour to suggest that Defendants rates are unreasonably high. (Dkt. No. 650, pp. 18-19.) The Court begins by noting that many of the billing rates at issue in this case are well within this limited range Perfect 10 suggests as reasonable in this district. See generally Appendix A, ante. Perfect 10 also argues one of its former attorneys with impressive credentials (Jeffrey Mausner) never charged Perfect 10 more than $320 per hour. (Id., p. 19; Dkt. No. 650-6, ¶3.) Perfect 10 further asserts the so-called "Laffey Matrix" suggests rates between $261.12 per hour for first year associates and $532.48 per hour for senior partners in Los Angeles. (Dkt. No. 650, p. 20; Dkt. No. 650-7, Exh. H.) Additionally, Perfect 10 argues, the American Intellectual Property Law Association ("AIPLA") Report of Economic Survey identified the median intellectual property partner billing rate in Los Angeles as $575 per hour in 2012, and the median Los Angeles associate billing rate as $395. (Dkt. No. 650-2, Exh. 7.) Perfect 10's expert John D. O'Connor also opines that this case was not sufficiently complex to warrant Defendants' use of top-tier attorneys at large law firms like Fenwick & West and Winston & Strawn, and that litigators at "smaller firms and spin-offs from larger firms" could have handled this action at significantly lower billing rates. (Dkt. No. 650-7, ¶¶48-54.)

Perfect 10 does not dispute that rates of $200 per hour and $100 per hour billed for staff attorneys and a contract attorney, respectively, were reasonable. The Court agrees that those rates, which are well below even the lowest suggested attorney rates in this action (and below even a reasonable paralegal rate) were reasonable.

i. Perfect 10's Attorney Rates

Although the rates charged by opposing counsel may sometimes be "useful", "the district court has the discretion not to rely on them." Gonzalez v. City of Maywood, supra, 729 F.3d at 1202. Here, the Court finds evidence regarding the rates Perfect 10's attorneys charge to be of limited use. To begin with, Perfect 10's own expert opines that "top partner rates" in the "Standard metropolitan intellectual property litigation" in California range "from $450 an hour to $600 an hour," significantly higher than the $320 an hour Mr. Mausner charged Perfect 10. Indeed, Mr. Mausner's billing rate is at the low end of Perfect 10's own proposed Laffey Matrix, and well below the bottom range Perfect 10 suggests from the AIPLA study.

ii. The Laffey Matrix

Nor is the Court persuaded that the Laffey Matrix is an appropriate barometer of "prevailing market rates" in the Central District. It is true that courts in this circuit are divided over the relevance of the Laffey Matrix in determining prevailing market rates. Compare, e.g., Crux ex rel. Cruz v. Alhambra School Dist., 601 F.Supp.2d 1183, 1200 n.15 (C.D. Cal. 2009) (finding Laffey Matrix "inconsistent with the standards" for calculating a reasonable fee); and Housing Rights Ctr. v. Sterling, No. CV 03-859 DSF, 2005 WL 3320738, at *2 (C.D. Cal. Nov. 1, 2005) (declining to apply Laffey Matrix); with Young v. Polo Retail, LLC, No. C-02-4546 VRW, 2007 WL 951821, *7 (N.D.Cal. Mar. 28, 2007) (applying Laffey Matrix); and In re HPL Technologies, Inc. Securities Litig., 366 F.Supp.2d 912, 921-22 (same). However, review of the Laffey Matrix's history, underlying purpose, and methodology show that reliance on that matrix to determine the "prevailing market rates" for attorneys in this district is inconsistent with the requirement that courts look to prevailing rates within the district.

In Laffey v. Northwest Airlines, Inc., 572 F.Supp. 354 (D.D.C. 1983), the district court for the District of Columbia looked to evidence produced by the parties in that action regarding prevailing legal rates in the District of Columbia, and set out a number of reasonable rates for an attorney fee award using ranges of years of attorney experience. Id., at pp. 371-75. Since that time, the "Laffey matrix has been regularly prepared and updated by the Civil Division of the United States Attorney's Office for the District of Columbia and used in fee shifting cases, among others." In re Chiron Corp. Sec. Litig., No. C-04-4293 VRW, 2007 WL 4249902, at *6 (N.D. Cal. Nov. 30, 2007). As explained in the Laffey Matrix itself, "[t]he matrix begins" with the "hourly rates approved in Laffey for work done principally in 1981-82" in the District of Columbia. (Dkt. No. 650-7, Exh. D, Explanatory Note 3.) The D.C. U.S. Attorney's Office then annually adjusts those rates "by adding the change in the cost of living for the Washington D.C. area to the applicable rate for the prior year, and then rounding up to the nearest multiple of $5 ... ." (Id.) This methodology is expressly designed to "ensure that the relationship between the highest rate and the lowest rate remains reasonably constant" to what it was in the District of Columbia in 1981-82. (Id.) "Changes in the cost of living are measured by the Consumer Price Index for All Urban Consumers ... for Washington-Baltimore, DC-MD-VA-WV, as announced by the Bureau of Labor Statistics for May of each year." (Id.)

Most problematic for Perfect 10's suggestion that the Court rely on the Laffey Matrix is the fact that it was and is expressly designed to reflect market rates in the District of Columbia. Here, however, the Court must look to market rates in the Central District of California, not the District of Columbia. Barjon v. Dalton, 132 F.3d 496, 500 (9th Cir. 1997). And even in the District of Columbia, the Laffey Matrix's underlying data "begins" with rates posited by parties in a single lawsuit, not a comprehensive study of prevailing market rates. Furthermore, the Laffey Matrix's methodology is premised on the notion that "prevailing market rates" for attorneys have risen at the same pace as the cost of living every year since the early 1980s and that the proportional difference in rates charged for a junior associate and a senior partner have remained constant for more than 30 years - assumptions that any attorney practicing in the Central District for the past 30 years would immediately debunk. (See Dkt. No. 655-34, ¶4 (Defendants' expert opining that "demand for legal services has [] grown dramatically" since the early 1980s and that "merely adjusting 1983 rates for general inflation materially underestimates the effect that demand has had on the rates charged for legal services generally and the IP litigation market specifically"). Indeed, even the cost of living increases built into the Laffey Matrix are those for the Washington, D.C. area, which can differ significantly from the Central District year-to-year.

Perfect 10's locality-adjusted Laffey Matrix, for example, provides that the reasonable difference in rates between the most junior associate with 1-3 years of experience ($261.12) and the most senior partner with more than 20 years of experience ($532.48) is $271.36. (Dkt. No. 650-7, Exh. F.) Admittedly, either rate is significantly higher than what most people in the greater Los Angeles area earn per hour, and even the Laffey rates warrant some sticker shock for the layman. But under the controlling law, the Court must compare the rates at hand with the market rates charged by private law firms in the Central District. For better or for worse, the price of litigation in the Central District can be very high, and the suggestion that the range of market rates in the greater Los Angeles area is as narrow as the Laffey Matrix provides "simply does not conform with the reality of Los Angeles firm billing practices." Housing Rights Ctr. v. Sterling, supra, 2005 WL 3320738, at *2. (declining to apply Laffey Matrix as inconsistent with the Central District's legal market).

According to the Bureau of Labor Statistics, for example, the overall cost of living (CPI-U) in the greater Los Angeles area increased at an annualized rate of 1.7 percent from May 2013 to May 2014, while the cost of living in the District of Columbia metropolitan area increased at a rate of 2.2 percent over the same time period. Bureau of Labor Statistics, Crawford & Church, eds., CPI Detailed Report: Data for May 2014, p. 40, http://www.bls.gov/cpi/cpid1405.pdf (accessed March 12, 2015). By contrast, the cost of living in the greater Los Angeles area increased at a rate of 1.1 percent from May 2013 to May 2014, while the cost of living in the greater Washington, D.C. metro area increased at a rate of only 0.5 percent. Bureau of Labor Statistics, Crawford, Church, & Rippy, eds., CPI Detailed Report: Data for May 2013, p. 50, http://www.bls.gov/cpi/cpid1305.pdf (accessed March 12, 2015). Over the course of 30 years, those differences add up.

Nor does Perfect 10's attempt to utilize the United States Government's locality pay adjustment tables (see Dkt. No. 650, p. 20; Dkt. No. 650-7, ¶¶58-61, Exh. F) make the Laffey Matrix any more useful in the Central District of California. Even assuming for the sake of argument that the additional "cost of living" increase federal employees in the Central District earn above federal employees in Washington D.C. incorporates more than 30-years of differences in annual cost of living changes between the two areas (a dubious proposition), reliance on the federal pay adjustment standards still rests on the assumption that prevailing attorney rates have increased at the same rate as the cost of living for three decades. Additionally, the federal pay adjustment Perfect 10 relies upon identifies a single "locality payment" for all of the Los Angeles, Long Beach, and Riverside areas. Perfect 10's logic presumes the same cost of living in Temecula and Beverly Hills. That assumption may be reasonable enough for the United States Government attempting to determine how much to pay every form of public employee, but it is not a realistic assessment of the legal market, where rates vary widely based on experience, reputation, academic credentials, and any other number of factors. See also Trevino v. Gates, 99 F.3d 991, 925 (9th Cir. 1996) (district court erred by relying on rates paid by government entity because "the proper reference point in determining an appropriate fee award is the rates charged by private attorneys in the same legal market").

iii. The AIPLA Economic Survey

Perfect 10's reliance on the AIPLA Economic Survey comes slightly closer to the mark in that it indicates average and median rates for intellectual property lawyers practicing in the Los Angeles area. (See Dkt. No. 650-2, Exh. 7.) However, in Los Angeles, the AIPLA survey only differentiates between partners and associates, without any differentiation based on years of experience. (Id.) And even in those two categories, the survey's results are based on the reported rates of only 13 partners and 13 associates, hardly a representative sample of "prevailing market rates" in the Central District. (Id.)

iv. Other Evidence of Prevailing Market Rates

Because Mr. Mausner's rates, the Laffey Matrix, and the AIPLA survey offer little to no insight on the actual prevailing market rates in the Central district of California for attorneys of equal experience, skill, and reputation, the Court looks to other evidence and concludes Defendants rates are reasonable. Though certainly not dispositive, the rates counsel actually charged Defendants is at least one factor the Court may consider in determining whether the rates sought are reasonable. Moore v. James H. Matthews & Co., supra, 682 F.2d at 840; accord Maldonado v. Lehman, 811 F.2d 1341, 1342 (9th Cir. 1987) ("evidence of counsel's customary hourly rate may be considered by the District Court" as one factor in determining the reasonable market rate). At minimum the rate an attorney actually charges its client is a "good starting to point" because "the actual rate that [the attorney] can command in the market is itself highly relevant proof of the prevailing community rate." Elser v. I.A.M. Nat. Pension Fund, 579 F. Supp. 1375, 1379 (C.D. Cal. 1984) (internal quotes omitted). From that starting point, the Court is not persuaded by Perfect 10's expert, Mr. O'Connor that rates in excess of $600 per hour are excessive in this case.

Defendants' billing records reflect that Defendants actually paid the rates charged at regular intervals. (See generally, Dkt. Nos. 644-7, 644-8, & 644-9; see also Dkt. No. 644-1, ¶15.) Defendants may have been confident of the merits of their defense, but an award of attorneys' fees for the bulk of their claims is discretionary, and there was no guarantee that Defendants would secure a fee award even if they ultimately prevailed on the merits. Although the rates an attorney claims to bill in a fee motion may be of less evidentiary value in the context of an attorney who takes the case on contingency or at a "blended" rate that is artificially low, evidence that an institutional client in a competitive legal market was willing to pay the rates charged without any guarantee of reimbursement is important evidence that the rate was reasonable.

To begin with, Defendants correctly note that when the tables were turned and Perfect 10 sought an award of attorneys' fees in a similar copyright action more than a decade ago, Perfect 10 introduced evidence that "Los Angeles firms bill partners at [] rates between $325 and $850." (Dkt. No. 655-25, p. 11:3-4 (order awarding Perfect 10 attorneys' fees in Perfect 10 v. Cybernet Ventures, Inc., et al., No. CV 01-02595-LGB (SHx) (C.D. Cal. Feb. 17, 2004), at Dkt. No. 416). Ten years later, that evidence from Perfect 10 is consistent with the Mr. Zeughauser's expert opinion that partner rates in the Los Angeles Area can exceed $1,400 per hour on the "super-premium" end. (Dkt. No. 655-34, ¶3.) As Defendants' expert Mr. Zeughauser opines, the rates charged by Fenwick & West and Winston & Strawn were "materially lower ... than those of other attorneys of similar experience and expertise practicing at other preeminent law firms in the Los Angeles market." (Dkt. No. 644-19, ¶11.)

Perfect 10's expert Mr. O'Connor concedes that Fenwick & West and Winston & Strawn are "pre-eminent" law firms with "competent" attorneys, and he acknowledges that fees as high as those at issue here are reasonable in "complex" cases that are "large and burdensome in scope" and require "large numbers of qualified lawyers ... ." (Dkt. No. 650-7, ¶48-50.) That is, even Mr. O'Connor does not dispute that attorneys of comparable "skill, experience, and reputation" in this district regularly charge similar rates - Mr. O'Conner simply opines that this case was "relatively standard DMCA case" and did not require the caliber of lawyering Defendants sought. (Id., at ¶51.) As discussed above, Mr. O'Connor's opinion of this case as a straightforward copyright case is not supported by the record. With more than 30 motions (including eight summary judgment motions), voluminous discovery (and seemingly endless discovery disputes), nearly 700 docket entries, and more than 38,000 pages in the Court's docket, there is little doubt that this case was precisely the sort of complex, large, and burdensome case that justified top-tier litigators. The value of Defendants' choice to retain "pre-eminent" law firms showed in both the quality of defense counsel's lawyering and the decisive results they obtained for Defendants, including a number of victories on previously unsettled questions of law.

The attorney rates outlined in Appendix A are also consistent with rates allowed in other intellectual property cases in this district of similar scale. In Masimo Corp. v. Tyco Health Care Group, L.P., No. CV 0204770-MRP (AJWx), 2007 WL 5279897 (C.D. Cal. Nov. 5, 2007) a court in this district held that attorney rates ranging from $390 per hour to $1,002.96 per hour were reasonable for the Los Angeles area nearly four years before this action was even filed. Id., at *7. Although the court acknowledged the rates in that case were "at the upper end for attorneys in the community" at that time, the court also observed that lead counsel in that case "offer[ed] clients abilities and a skill set that are largely unique and particularly valuable in a case of this complexity." Id. Similarly, in Love I, supra, a court of this district concluded that junior associate rates of $305 per hour to partner rates of $690 per hour were reasonable more than seven years ago as "consistent with the rates typically charged by other highly-regarded southern California law firms for similar work by attorneys of similar experience. Love I, supra, 2007 WL 2709975, at *8.

Importantly, the court in Love I gave "no weight to the Plaintiff's contention that it was unreasonable for Defendants to hire a nationally-renowned law firm - with relatively high hourly rates - given the lack of complexity of this case." Id. There, as here, the court observed the "case was far from routine," involving multiple claims against multiple parties, novel issues of law, and millions of dollars in potential damages. Id. "Far from being an overreaction to a simple case, Defendants' retention of reputable and skilled counsel to engage the multifarious allegations against them was eminently reasonable." Id. Defendants' election to engage Winston & Strawn, and ultimately Fenwick & West when Mr. Bridges changed firms, was all the more reasonable given that Mr. Bridges and Jennifer Golinveaux (another partner on this case while it was still with Winston & Strawn) had already successfully represented Google and Microsoft in similar actions by Perfect 10, obtaining significant published decisions from the Ninth Circuit in Perfect 10 v. Amazon, Inc., 508 F.3d 1146 (9th Cir. 2007) and Perfect 10 v. Google, Inc., supra, 653 F.3d 976. As Defendants correctly observe, Perfect 10's citation to cases in this district involving lower fee awards (Dkt. No. 650, pp. 18-19) all involved issues of significantly less complexity with substantially lower amounts at stake and substantially less evidence of the attorneys' reputation, skill, and experience. (See 655, p. 18 n.12.)

From the outset, Perfect 10 asserted the damages in this case were potentially in the multiple millions. (See Dkt. No. 1, pp. 17-18.) And as recently as July 2014, Mr. Zada estimated Perfect 10's damages in this action were a minimum of $14,634,386 and potentially as high as $51,103,897. (Dkt. No. 644-11, pp. 11-16.)

Finally, the court does not agree with Perfect 10's assessment that "Defendants have failed to adduce evidence that associate attorney and support personnel have extraordinary qualifications which warrant the requested rates...." (Dkt. No. 650, p. 21.) For each attorney for whom Defendants seek fees, Defendants lay out in detail the attorney's academic background, years in practice, work history, specialties, and cases of note. (Dkt. No. 644-1, ¶¶17-42.) Perfect 10 cites no authority to suggest Defendants were required to offer more, and the Court declines to transform this motion into a "second major litigation" on each attorney's curriculum vitae. Under the all the circumstances, the Court finds that the rates billed by each of the attorneys identified in Appendix A was reasonable and was comparable to prevailing market rates for attorneys of similar experience, skill, and reputation in similarly complex litigation.

b. Support Staff and Paralegal Rates

Defendants also seek fees for work performed by several paralegals and discovery support staff. "Work performed by paralegals may be compensated as part of an attorney's fee award." Hirsch v. Compton Unified Sch. Dist., No. CV 12-01269 RSWL MRW, 2013 WL 1898553, at *3 (C.D. Cal. May 3, 2013), citing Missouri v. Jenkins, 491 U.S. 274, 286-88, 109 S.Ct. 2463, 105 L.Ed.2d 229 (1989). Similarly, work performed by litigation support staff that directly support the substantive litigation (as opposed to routine clerical work) is compensable as part of an attorneys' fee award. See Armstrong v. Brown, 805 F. Supp. 2d 918, 922 (N.D. Cal. 2011) (awarding fees for non-paralegal "support staff"); accord Spence v. Wells Fargo Bank, N.A., No. 1:10-CV-2057 AWI GSA, 2012 WL 844713, at *5 (E.D. Cal. Mar. 12, 2012); (awarding fees for "support staff"); Intamin, Ltd. v. Magnetar Technologies Corp., No. CV 04-0511 GAF JWJX, 2009 WL 5215393, at *3 (C.D. Cal. Dec. 28, 2009) (same). Perfect 10 does not dispute that a fee award may properly include work done by paralegals and discovery support staff, but contends Defendants have failed to meet their burden to "substantiate[] the requested paralegal and support staff rates." (Dkt. No. 650, p. 21.)

For their two discovery support staff, Defendants seek rates of $290 for Brad Bonnington and $240 for David Tran. (Dkt. No. 644-1, ¶38.) Mr. Bonnington has managed Defendants' Practice Support Group for 13 years and has "handled electronic information management for more than 500 matters. (Id.) Mr. Tran has been a specialist in that Support Group for 10 years. (Id.) Together, Mr. Bonnington and Mr. Tran were responsible for "coordinating the production and receipt of hundreds of thousands of pages that the parties reviewed and produced during the course of this litigation." (Id.) Although Defendants identify little else in support of their contention that those rates are market rates for discovery support staff, the Court again begins with the evidence of the rates they actually charged. Maldonado v. Lehman, supra, 811 F.2d at 1342. It is also worth noting that Perfect 10 offers an equally limited rebuttal to the evidence of the support staff's actual rates. Perfect 10 merely argues that its adjusted Laffey Matrix indicates rates of roughly $153 for paralegals. (Dkt. No. 650, p. 21; Dkt. No. 650-7, ¶61.) But as the Court already discussed, it is not persuaded that the Laffey Matrix offers relevant evidence regarding the legal market in this district.

However, review of other actions reveals that Defendants' requested support staff rates are at least in the ballpark. In Browne v. Am. Honda Motor Co., No. CV 09-06750 MMM DTBX, 2010 WL 9499073 (C.D. Cal. Oct. 5, 2010), for example, a court in this district concluded that, as of four years ago, rates as high $275 were reasonable for the Central District. Id., at *6 n.31. Several years later, it is reasonable to conclude that market rates have increased from those held reasonable four years ago - particularly in light of the fact that the economy in general (and particularly the legal market) in the greater Los Angeles area has recovered significantly since that time. And while Defendants rates may be near the higher end of the reasonable range of market rates, those rates appear to come with at least a decade each of litigation support experience in hundreds of cases. Moreover, the scope of discovery in this action was both extensive and highly contentious, and it was perfectly reasonable for Defendants to utilize the most skilled discovery support staff available. Though likely near the top of the market, Fenwick & West is able to command the rates requested in the market, and it appears other firms do so as well, when necessary. The Court finds the rates for Mr. Bonnington and Mr. Tran reasonable.

The Court also finds Defendants' paralegal rates are reasonable. Defendants seek paralegal fees at rates between $240 for a paralegal with five years' experience to $345 for a paralegal with 23 years' experience. See Appendix A, ante. Again, Perfect 10 notes the Laffey Matrix suggests $153 for paralegal work, which the Court finds unpersuasive for reasons the Court has already discussed. Instead, recent authorities in the Central District suggest Defendants' requested rates are reasonable. As long ago as 2008, for example, courts of this district repeatedly found paralegal rates regularly reached $250 in the Central District. See Hirsch v. Compton Unified Sch. Dist., No. CV 12-01269 RSWL MRW, 2013 WL 1898553, at *3 (C.D. Cal. May 3, 2013) (collecting cases). And while the, the rate requested of $345 for Robert Winat is admittedly on the higher end, Mr. Winat has 23 years of experience as a paralegal. Compare Samson v. Nama Holdings, LLC, CV 09-01433 MMM, 2009 U.S. Dist. LEXIS 114494 at *13, *18 (C.D.Cal. Aug. 10, 2009) (awarding paralegal rate of $270 nearly six years ago for services of a paralegal with 18 years of experience). The Court finds the rates for Defendants' paralegals identified in Appendix A are reasonable and consistent with comparable market rates.

2. The Hours Billed Are Largely Reasonable

In support of their motion, Defendants submit roughly 800 pages of defense counsel's billing records, and request by the court's count 11,125.5 hours in billed time. (Dkt. No. 644-1, ¶¶17-42; Dkt. Nos. 644-7, 644-8, 644-9.) With the exception of hours billed for non-compensable claims, the Court finds Defendants' billed hours to be largely reasonable and awards Defendants 90% of the hours requested on their compensable claims.

Defendants do not sum the total number of hours they bill. The Court draws its count of total hours from Defendants' summary of the billed hours sought for each of the 27 time keepers for whom defendants actually seek reimbursement. (Dkt. No. 644-1,¶¶17-42.) The Court's calculation differs from Perfect 10's calculation that Defendants seek reimbursement for 11,032.3 hours by 52 time keepers. However, Perfect 10's calculation includes a number of time entries from individuals who performed relatively small amounts of work and for whom Defendants do not seek reimbursement. (Compare Dkt. No. 650-2, Exh. 1; with Dkt. No. 644-1, ¶¶17-42, 47.) Perfect 10's calculation also differs from the number of hours Defendants request for a number of the timekeepers for which they do seek reimbursement. (Compare Dkt. No. 650-2, Exh. 1 (calculating 1,292.2 hours for Mr. Bridges); with Dkt. No. 644-1, ¶17 (requesting a total of 1,209.4 hours for Mr. Bridges). In light of the inconsistencies between Perfect 10's estimate and Defendants' actual request, as well as Defendants access to electronic records and the reduced likelihood of an arithmetic error, the Court uses Defendants summary of the hours requested in Mr. Bridges' declaration, with a total request of 11,125.5 hours. (Dkt. No. 644-1, ¶¶17-42.)

a. The Hours Perfect 10 Spent Litigating this Case Are Not a Good Barometer of Whether Defendants' Billed Hours Are Reasonable; to the Extent Perfect 10's Time Is Relevant, it Supports' Defendants' Request

Perfect 10 asserts that total number of hours billed is "outrageous." Perfect 10 argues Defendants' request for 9,397.2 attorney hours is wildly excessive in comparison to Perfect 10's 3,912.8 attorney hours. (Dkt. No. 650, p. 21; Dkt. No. 6502-, ¶3.) It is, of course true, that "if the time claimed by the prevailing party is of a substantially greater magnitude than what the other side spent, that often indicates too much time is claimed." Democratic Party of Wash. State v. Reed, 388 F.3d 1281, 1287 (9th Cir. 2004). However, that is a rule of thumb, not of law. "Although opposing counsel's billing records may be relevant to determining whether the prevailing party spent a reasonable number of hours on the case, those records are not dispositive... and the district court has the discretion not to rely on them." Gonzalez v. City of Maywood, supra, 729 F.3d at 1202. "[O]pposing parties do not always have the same responsibilities under the applicable rules, nor are they necessarily similarly situated with respect to their access to necessary facts, the need to do original legal research to make out their case, and so on." Ferland v. Conrad Credit Corp., 244 F.3d 1145, 1151 (9th Cir. 2001). As a serial litigant on similar issues, Perfect 10 was able to take advantage of numerous efficiencies not available to Defendants. Perfect 10's paralegal Melanie Poblete, for example, admitted in deposition that she reused a number of discovery responses in this case that Perfect 10 had already produced in other cases, sometimes even preemptively, without doing any new work to expand or reduce the former response. (Dkt. No. 655-15, pp. 186:24-187:10.)

Perfect 10 calculates 9,389.3 attorney hours. (Dkt. No. 650, p. 21; Dkt. No. 650-2, Exh. 1.) As discussed in note 16, above, the Court relies on its own calculation which, in any event, only differs from Perfect 10's calculation by 7.9 hours or 0.008 percent.

Perfect 10's asserted 3,912.8 attorney hours also vastly underestimates the amount of legal work Perfect 10 actually put into this case. Mr. Zada testified in deposition, for example that he spends eight hours per day, 365 days a year working on litigation (Dkt. No. 644-16, p. 541:1-2), and testified that, as of April 2014, he had spent in excess of 2,000 hours working on this case alone. Mr. Zada's work on the case, however, is not reflected in Perfect 10's calculation of billed hours. (Dkt. No. 650-2, ¶3.) This omission is significant, as Mr. Zada's work went well beyond that of a typical corporate representative. Mr. Zada testified that he frequently wrote "the first draft of a brief or [his] declaration, put the exhibits together, and then [passed] that off to the attorneys to make sure everything [was] ... professional and legal." (Dkt. No. 655-7, p. 84:5-10.) At depositions, Mr. Zada frequently interjected and instructed counsel to ask specific questions. (See Dkt. Nos. 655-12; 655-13; 655-14.) Additionally, Mr. Zada frequently interacted directly with expert witnesses in a manner usually reserved for counsel, drafting or editing expert declarations. (See Dkt. No. 655-10 (email exchange between Zada and expert witness re Zada's edits to expert's declaration); Dkt. No. 660 (email exchange in which Zada discusses drafting entire expert declaration); Dkt. No. 661 (same).) In fact, at a hearing before the Magistrate Judge on a discovery motion, Mr. Zada argued a significant portion of the motion before the Magistrate realized Mr. Zada was not an attorney and indicated that he only wanted to hear from counsel. (Dkt. No. 655-11, p. 18:10-19:5.)

Mr. Zada gave that testimony before either party had filed any of the seven second-round summary judgment motions and other substantial motion practice. By way of example, 424 of the 682 docket entries at the time of writing were entered in this action after April 2014 when Mr. Zada testified he had already spent 2,000 hours litigating this case.

Similarly, Perfect 10's paralegal admitted in deposition that she performed countless hours of work on behalf of Perfect 10, none of which are reflected in Perfect 10's claim of under 4,000 hours. Ms. Poblete testified that in the course of her full-time employment with Perfect 10, she spent roughly 80% of her time working on litigation and "the bulk" or "90 percent" of her litigation time working on this specific case. (Dkt. No. 655-15, pp. 94:20-95:14, 112:5-13.) That is, Ms. Poblete, who has worked as a paralegal for Perfect 10 on this action "since its inception" (Dkt. No. 650-2, ¶1), testified that she spends approximately 72 percent (0.8 x 0.9 = 0.72) of her full time employment working on this case. Even conservatively excluding any efforts Ms. Poblete may have undertook prior to filing or after the Court granted summary judgment, Ms. Poblete's own estimate would have her spending 5,385.6 on this case in the 187 weeks between filing and summary judgment. However, Ms. Poblete did not keep time records for her work in this litigation, and it is not reflected in Perfect 10's estimate of hours spent litigating the case. (Dkt. No. 650-2, ¶3; Dkt. No. 655-15, p. 236:12-14.) Moreover, at least some of the work Ms. Poblete performed for Perfect 10 appears to be work traditionally performed by an attorney. (See, e.g., Dkt. No. 661 (email chain between Zada, Poblete and expert in which Poblete proposes substantive edits to expert declaration).

Adding those two estimates alone, Perfect 10's estimate of time spent litigating the action grows from 3912.8 to 11,298.4 hours, in excess of Defendants billed hours. But Mr. Zada and Ms. Poblete are not the only two individuals who performed unreported in-house legal services for Perfect 10. Perfect 10's contractor Gwendalyn Augustine, for example, testified that she "help[ed] prepare filings for the court, [took] documents to the court occasionally, ...[and] help[ed] prepare things that may end up being used in the case." (Dkt. No. 655-17, p. 79:10-15.) Although Ms. Augustine testified that her workload for Perfect 10 had declined recently, as of June 2014 she continued to help Perfect 10 prepare discovery responses, assist Ms. Poblete with Bluebooking, and take filings to court. (Id., at pp. 79:16-80:10, 157:14-158:1.) Likewise, Sheena Chou testified that she performed legal assistant services for Perfect 10 for roughly 20 hours per week when she worked for Perfect 10. (Dkt. No. 655-18, p. 106:19-22.) While it is unclear when Ms. Chou stopped working for Perfect 10, it is undisputed that she did at least some work for them on Perfect 10's motion for preliminary injunction and Perfect 10's opposition to Defendants' venue motion. (See Dkt. No. 18 (Chou's declaration in support of preliminary injunction motion); Dkt. No. 32 -1 (Chou's declaration in support of Perfect 10's opposition to a change of venue).)

Equally problematic for the credibility of Perfect 10's 3,912.8 hour estimate of its work on this case is the fact that its own counsel's sample billing statement shows that Perfect 10's estimate omits other attorneys who performed work in this case. Though not identified in Perfect 10's summary of its legal work (Dkt. No. 650-2, ¶3), Perfect 10's own evidence showed that Perfect 10's former attorney Jeffrey Mausner performed work in this case. (See Dkt. No. 650-5, Exh. 1, p. 1, (billing statements from Benink for "T/C with N[orman] Z[ada] and J[effrey] M[ausner]; p. 4 (Benink billing entries for "T/C with JM and NZ" and "Revise brief with JM edits").) Faced with that inconsistency, Perfect 10 belatedly submitted a declaration from Mr. Mausner acknowledging that he performed work for Perfect 10 on this action, but stated his work was limited to 27.1 hours. (Dkt. No. 662-2, ¶3.) Similarly, Mr. Benink's billing statements demonstrate that Steven Fabrizio, formerly a partner at Jenner & Block (Dkt. No. 655-1, ¶19) performed legal services for Perfect 10, editing a brief in June of 2011. (Dkt. No. 650-5, Exh. 1, p. 4 (Benink billing statements for "Revis[ing] brief after Fabrizio comments" and "Research[ing] further issues from Fabrizio's comments").) In an evidentiary objection, Perfect 10 argues that "Fabrizio has never represented Perfect 10." (Dkt. No. 662, p. 8.) However, Perfect 10 notably does not submit any sworn testimony to that effect, and fails to account for the clear billing entries stating that Mr. Fabrizio participated in drafting various briefs in this case.

While the extent of Perfect 10's reliance on attorneys not accounted for in Perfect 10's 3,912.8 hour estimate is unclear, Mr. Zada's own testimony offers some insight. Perfect 10 estimates that, as of January 12, 2014, its legal expenses amounted to a total of $492,640. However, on April 24, 2014, Mr. Zada estimated Perfect 10's legal expenses in this case were "between [$]4- to [$]650,000" at that time. (Dkt. No. 655-7, p. 138:22-23.) A significant proportion (if not the majority) of litigation in this action occurred after Mr. Zada's April 24, 2014 estimate. (See Footnote 18, herein.) Indeed, if one includes the instant motion, 22 of the 35 motions filed in this case were filed after Mr. Zada estimated legal fees between $400,000 and $650,000. By Mr. Zada's own testimony, it appears clear that Perfect 10 incurred significant legal expenses not reflected in its current estimate. Instead, it seems much more likely Perfect 10's actual legal fees in this action more closely resemble those it incurred in the Google litigation, where Mr. Zada testified Perfect 10 incurred roughly $7,000,000 in outside legal fees alone, without accounting for salaries for in-house counsel, Mr. Zada, or in-house legal support staff. (Dkt. No. 655-8, pp. 546:20-547:1.)

Although the Google action appears to have been even more heavily litigated than this case, it seems a fair comparison for four important reasons: (1) the Google action was filed in 2002 when market rates would have been lower; (2) Perfect 10's legal bills reached $7M despite its evidence in this case that its attorneys charge rates at the lower end of or below market rates; (3) Perfect 10's $7 million legal bill excluded in-house legal support, which (as evidenced above) was likely substantial, and; (4) even considering all of those things, Defendants' claimed legal expenses in this action are roughly $1 million lower than Perfect 10's expenses in the Google action.

Perfect 10 was, of course, well within its rights to do a significant amount of its legal work in-house to save on litigation costs, but Giganews was under no obligation to follow suit. "Plaintiff's counsel may have billed fewer hours, but Plaintiff also lost. Thus, in this case, the difference in hours that Plaintiff and Defendants spent has no bearing on the reasonableness of Defendants' work." Love I, supra, 2007 WL 2709975, at *10. In any event, given Perfect 10's vastly underinflated estimate of the time it spend litigating this case, it actually appears that Defendants achieved those results with less effort than Perfect 10. So to the extent the time Perfect 10 spent is a "particularly good indicator of how much time [was] necessary" to litigate the case (Democratic Party of Washington State v. Reed, supra, 388 F.3d at 1287), it only suggests that Defendants billed hours are reasonable.

b. Defendants' Records Do Not Reflect a "Billing Frenzy"

Perfect 10 next contends that Defendants billing records reflect a "billing frenzy" because defense counsel billed 1,577.1 hours in the first 18 months of litigation compared to 230 hours by Perfect 10's attorneys. (Dkt. No. 650, p. 22.) As discussed above, Perfect 10's attempt to compare its underinflated account of its own time is unpersuasive. However, Perfect 10 does identify a number of individual motions for which defense counsel appears to have billed an unusual number of hours. Perfect 10 calculates, for example, that Defendants request 578 hours for its original motion to dismiss and to change venue. (Dkt. No. 650-2, Exh. 2.) Defendants also billed a similar amount of time opposing Perfect 10's motion for preliminary injunction and billed $346.4 hours to prepare motions to dismiss and strike Perfect 10's First Amended Complaint, which revisited a number of issues addressed in the original motion to dismiss. (Id.) Defendants do not dispute Perfect 10's calculations. While the Court is not persuaded that such billing statements reflect a "billing frenzy," the Court thinks it appropriate to impose a modest reduction on Defendants' request.

Again, the fact that Defendants were willing to pay the bills is at least some evidence that the vigor with which Defendants defended the action was warranted under the circumstances. Had Perfect 10 prevailed on its motion for preliminary injunction, for example, Defendants would have likely gone out of business - a fact Mr. Zada acknowledged at the time. (Dkt. No. 655-1, ¶3.) In this regard, Perfect 10 "cannot litigate tenaciously and then be heard to complain about the time necessarily spent by the [the opposing party] in response." Cataphora Inc. v. Parker, 848 F. Supp. 2d 1064, 1070 (N.D. Cal. 2012), quoting Int'l Longshoremen's & Warehousemen's Union v. Los Angeles Export Terminal, Inc., 69 Cal.App.4th 287, 304, 81 Cal.Rptr.2d 456 (1999). However, the Court thinks such billing hours a little steep, even in given the admittedly high stakes in this litigation. While the Court finds no basis to impose the drastic 50 percent across-the-board cut Perfect 10 suggests (Dkt. No. 650, p. 23), the Court finds the pattern of slightly-more-vigorous-than-necessary billing revealed in Perfect 10's calculations warrants a modest reduction. The Court imposes a "haircut," and reduces Defendants' fee request by 10 percent, to bring Defendants' billed hours into line with a more reasonable balance between the costs of litigation and the stakes of losing. Moreno v. City of Sacramento, supra, 534 F.3d at 1112. /// /// /// ///

c. Defendants Reasonable Redactions Are Appropriate

Perfect 10 briefly asserts that Defendants sporadic redactions "have gone too far." (Dkt. No. 650, p. 23.) Upon independent review of Defendants' billing records, the Court finds Defendants minimal redactions well within the realm of what is necessary to protect the attorney-client privilege while still affording the Court sufficient detail to determine the reasonableness of the hours requested. Democratic Party of Washington State v. Reed, supra, 388 F.3d at 1286 (similar redactions appropriate because the "redactions do not impair the ability of the court to judge whether the work was an appropriate basis for fees").

d. Defendants Records Do Not Reflect Block Billing

Perfect 10 also urges the Court to reduce Defendants' bills by an unspecified amount for so-called "block billing." As the Ninth Circuit has explained, "'[b]lock billing' is 'the time-keeping method by which each lawyer and legal assistant enters the total daily time spent working on a case, rather than itemizing the time expended on specific tasks.'" Welch v. Metro. Life Ins. Co., 480 F.3d 942, 945 n.2 (9th Cir. 2007), quoting Harolds Stores, Inc. v. Dillard Dep't Stores, Inc., 82 F.3d 1533, 1554 n. 15 (10th Cir. 1996). Because block billing "makes it more difficult to determine how much time was spent on particular activities," district courts may "reduce hours that are billed in block format" to avoid the potential for padding. Id., at 1554.

Defendant's records are not block billed. Even Perfect 10's "Examples of Defendants' 'Block Billing'" (Dkt. No. 650-2, Exh. 2) reflect itemized statements of the specific tasks defense counsel undertook. It is true that Defendants billing records include a handful of larger time blocks of several hours, but even those sporadic entries include detailed descriptions of the various tasks counsel undertook in that time. For example, Perfect 10 singles out a July 12, 2011 12.8 hour billing entry by Winston & Strawn associate Kevin Oh as an example of block billing. (Dkt. No. 650-2, Exh. 2; Dkt. No. 644-7, p. 39.) But even that billing entry goes well beyond stating the "total daily time" Mr. Oh spent on the case. Instead, the record itemizes a number of tasks Mr. Oh conducted in that 12.8 hour period:

"Review emails from and confer with A. Bridges, J. Golinveaux, T. Kearney regarding Perfect 10's preliminary injunction motion, DMCA notices, and motion to seal; confer with J. Golinveaux concerning discovery; draft email to client concerning preliminary injunction motions; review exhibits, pleadings, and DMCA notices; review and revise draft letters to Perfect 10."

(Dkt. No. 650-2, Exh. 2; Dkt. No. 644-7, p. 39.)

The Court finds such specific itemizations of counsels' tasks sufficient to evaluate the reasonableness of Defendants' request. Lytle v. Carl, 382 F.3d 978, 989 (9th Cir. 2004) ("minimal" descriptions sufficient to support an award of attorneys' fees so long as "they establish that the time was spent on the matters for which" the party seeks fees). Even Perfect 10 does not argue that 12.8 hours was an unreasonable amount of time to perform those itemized tasks. (See Dkt. No. 650, p. 23.) While defense counsel must account for their time, they are "not required to record in great detail how each minute of [their] time was expended." Hensley, supra, 461 U.S. at 437 n.12. Defense counsels' time entries "identify the general subject matter of [their] time expenditures," and are adequate to support an award of attorneys' fees. See Fox v. Vice, ___ U.S. ___, 131 S.Ct. 2205, 2216, 180 L.Ed.2d 45 (2011) ("The essential goal in shifting fees (to either party) is to do rough justice, not to achieve auditing perfection.")

Although Mr. Oh's entry is sufficient to determine the reasonableness of Defendants' request, it is also an outlier. The vast majority of defense counsels' billing entries are for much smaller amounts of time. And even among the larger entries, the majority of larger billing entries include even more specific time statements for individual tasks. Another 8.8-hour billing entry from the same day by Winston & Strawn associate Thomas Kearney, for example, states reads: "Study PI motion and supporting papers (4.8); team meeting (.5); draft opposition to sealing motion (3.5)."

Even if such entries could be construed as "block billing," it would be improper to apply an across-the-board reduction to all of Defendants' requested hours, because the vast majority of the time entries reflect discrete time statements for discrete tasks. Welch v. Metropolitan Life Ins. Co., supra, 480 F.3d 942, p. 948. Nor does Perfect 10 propose what an appropriate cut would be for the limited number of entries Perfect 10 believes to be "block billed." To the extent any reduction would be appropriate because a handful of defense counsel's billing entries permitted some amount of padding, the 10 percent reduction the Court already imposes is more than sufficient to address the problem.

e. Modest Reductions for Clerical Work Are Appropriate

Perfect 10 next identifies a number of billing entries it claims are "secretarial or clerical in nature" and asserts that those activities "generally cannot be recovered as attorney's fees under the lodestar methodology." (Dkt. No. 650, p. 24; Dkt. No. 650-2, Exh. 5.) Perfect 10's assessment of the law is incorrect. "[E]ven purely clerical or secretarial work is compensable if it is customary to bill such work separately... ." Trustees of Const. Indus. & Laborers Health & Welfare Trust v. Redland Ins. Co., 460 F.3d 1253, 1257 (9th Cir. 2006). As the Supreme Court has observed, "[s]uch separate billing appears to be the practice in most communities today," (Missouri v. Jenkins by Agyei, 491 U.S. 274, 289 (1989)) and Perfect 10 offers nothing to suggest a different custom in the central district, where the work of paralegals and legal assistants is routinely billed separately. (See, e.g., cases cited in section III.B.1.b, supra.) The issue is not whether such activities are compensable, but the appropriate rate of compensation. "[P]urely clerical or secretarial tasks should not be billed at a paralegal rate, regardless of who performs them." Missouri v. Jenkins by Agyei, 491 U.S. 274, 288 (1989).

The vast majority of paralegal time reflected in Defendants' billing records is for traditional paralegal work (i.e., reviewing legal documents, cite checking, , etc.). However, Perfect 10 correctly notes that some tasks billed at paralegal rates appear to be purely clerical in nature, like "organiz[ing] selected documents from P10's production for attorney review" and "organiz[ing] attorney set of documents in preparation for hearing." (Id.) Although such activities account for a relatively small portion of the paralegal activities billed, the Court's review of Defendants billing records reveals roughly 20 percent of such tasks should have been billed at a rate more appropriate for clerical tasks. Fox v. Vice, supra, 131 S.Ct. at 2216 ("trial courts may take into account their overall sense of a suit, and may use estimates in calculating and allocating an attorney's time"). In total, Defendants seek 1,592.2 paralegal hours, 20 percent of which is 318.44 hours. As shown in Appendix A below, the weighted average paralegal hourly rate was $287.47 per hour. Reducing that rate by half ($143.74) to account for the estimated 318.44 hours of clerical work, the Court will reduce Defendants' total request by $45,772.57 for those separately billed clerical tasks.

A traditional average rate would simply be the sum of each paralegal's hourly rate divided by total number of paralegals. However, because there is so much variance among the number of hours each paralegal billed, a traditional average would not accurately account for the number of hours each paralegal billed. Instead, the court utilizes a weighted average, which multiplies each paralegal's rate by the number of hours billed, adds the product for each paralegal to a total sum, and then divides by the total number of hours worked. This calculation can be expressed in equation form as x = ?n (rn *hn )/ht where x is the weighted average rate per hour, rn is each paralegal's hourly rate, hn is each paralegal's billed hours, and ht is the total number of hours billed.

f. Perfect 10's Reliance on Other Fee Awards Is Unpersuasive

Citing to a handful of other copyright actions from within and without this district, Perfect 10 also urges the court to reduce Defendants' attorneys' fee request to a purported average of $218,000 for attorneys' fees in copyright cases. (Dkt. No. 650, p. 24; Dkt. No. 650-3, ¶11 & 12, Exh. 1.) The Court finds no basis for such a drastic and unwarranted reduction. Perfect 10 offers no evidence to suggest that any of those cases was litigated nearly as vigorously as this action, despite the fact that the size of the present fee request is, in significant part, the result of Perfect 10's own litigation conduct. Indeed, Perfect 10's own comparison erroneously attempts to compare the overall fee award, without any reference to the amount of time it reasonably took to actually litigate any of those ostensibly "analogous" cases. See, Viveros v. Donahoe, No. CV 10-08593 MMM EX, 2013 WL 1224848, at *10 (C.D. Cal. Mar. 27, 2013) (cross-checking number of hours at billed with the amount of time held reasonable in "analogous cases").

Admittedly, a multi-million dollar attorneys' fee request is objectively large. But a fee award is not unreasonable simply because it involves a lot of money. The operative question isn't the dollar amount, the question is whether the rates charged and the hours spent are reasonable in light of the totality of the litigation. Courts in this district have awarded attorneys' fees in excess of $100,000,000 in copyright litigation depending on the nature of the underlying litigation, and reference to the dollar amount of a particular fee award is not especially helpful without a sense of how much time it took to reasonably litigate the case. See Mattel, Inc. v. MGA Entm't, Inc., No. CV 04-9049 DOC RNBX, 2011 WL 3420603, at *10 (C.D. Cal. Aug. 4, 2011) aff'd sub nom. Mattel, Inc v. MGA Entm't, Inc., 705 F.3d 1108 (9th Cir. 2013) (awarding attorneys' fees of $105,688,073.00 to prevailing copyright defendant). Perfect 10 admits that the size of the fee award at issue in this case is lower than Perfect 10's own costs prosecuting a similar lawsuit against Google. Though large, the fee award comports with the Court's "overall sense" of the amount of effort it took to successfully litigate this high-stakes action over the course of four years, and the Court finds it appropriate under the circumstances. Fox v. Vice, supra, 131 S.Ct. at 2216.

3. Lodestar Calculation

In light of the foregoing, the Court calculates Defendants lodestar at $5,213,117.06. The Court reduces Defendants' initial request of $6,104,372.66 by $266,247.80 (the amount Defendants allocate to their trademark defense; Dkt. No. 644-1, ¶56) to reflect the Court's conclusion that Defendants are not entitled to an award of attorneys' fees under 15 U.S.C. section 1117. The Court further reduces Defendants' initial request by $45,772.57 to account for certain clerical work billed at paralegal rates, as stated in section III.B.2.e, for a subtotal of $5,792,352.29. Finally the Court reduces that subtotal by 10 percent as discussed in section III.B.2.b for an adjusted lodestar amount of $5,213,117.06.

Because neither party urges the court to impose an upward or downward multiplier under Kerr v. Screen Extras Guild, Inc., 526 F.2d 67, 70 (9th Cir. 1975), the Court's analysis of a reasonable attorneys fee ends with the presumptively reasonable lodestar amount. See Ballen v. City of Redmond, 466 F.3d 736, 746 (9th Cir. 2006).

C. Defendants' Non-Taxable Cost Request is Reasonable

As the prevailing party in this action, Defendants are also entitled to their "full costs" under 17 U.S.C. section 505. Although Defendants have separately applied to the clerk for their traditional taxable costs under 28 U.S.C. section 1920 (Dkt. No. 635), they apply to this Court for an award of $424,235.47 in non-taxable costs under the Copyright Act. (Dkt. No. 644, pp. 24-25; Dkt. No. 644-1, ¶59. Defendants provide itemized statements of the expenses for which they seek reimbursement (Dkt. Nos. 644-7; 644-8; 644-10), and Mr. Bridges declares Defendants "necessarily incurred these costs and expenses in the course of providing legal services in this case" and that defense counsel "customarily bill[s] those costs and expenses separately to and apart from attorney hours."

Perfect 10 calculates that $256,960 of that cost bill is for "data extraction and processing costs" and argues that "Defendants provided no detail regarding the activities that generated these alleged costs, why they were necessary to the defense of this case, or how they were priced." (Dkt. No. 650, p. 24; Dkt. No. 650-2, Exh. 6.) However, Perfect 10 offers no evidence or argument to suggest that the scope of that cost bill is unreasonable in the context of this litigation. Instead, Perfect 10 merely suggests that Defendants were required to include a detailed explanation of how each of those electronic discovery costs were necessary to the litigation. But as Perfect 10's own chart shows, the cost entries are self-explanatory, stating Defendants incurred the costs for data storage, data extraction and processing, and account fees with the service provider. (Dkt. No. 650-2, Exh. 6.) Perfect 10 admits that this action involved staggering amounts of discovery - nearly 1.5 terabytes of documents - and data storage and processing would certainly have been essential to effectively litigate. (Dkt. No. 650-8, ¶13.) As one district court has found, "in cases of this complexity, e-discovery saves costs overall by allowing discovery to be conducted in an efficient and cost-effective manner." In re Aspartame Antitrust Litigation, 817 F.Supp.2d 608, 614-15 (E.D. PA. 2011) (awarding costs "for the creation of a litigation database, storage of data, imaging hard drives, keyword searches, deduplication, data extraction and processing"). The Court finds the electronic discovery costs set forth in Mr. Bridges' declaration were reasonably necessary to the litigation.

There is some debate as to whether electronic discovery costs are allowable as taxable costs under 28 U.S.C. §1920. See, e.g., Alzheimer's Inst. of Am. Inc v. Elan Corp. PLC, No. C-10-00482-EDL, 2013 WL 8744216, at *4 (N.D. Cal. Jan. 31, 2013) (disallowing electronic discovery costs under section 1920). However, it is undisputed that an award of "full costs" under 17 U.S.C. §505 is significantly more expansive than section 1920 (Twentieth Century Fox Film Corp. v. Entertainment Distributing, 429 F.3d at 885), and Perfect 10 does not dispute that electronic discovery costs are available under section 505. --------

Perfect 10's objection to $68,634 in Westlaw legal research charges is similarly unpersuasive. Reasonable charges for computerized legal research are recoverable under section 505 so long as the prevailing practice in the district is to separately bill for such expenses. Yue v. Storage Tech. Corp., 2008 WL 4185835, at *6. Defense counsel declares that attorneys in this market customarily bill separately for the costs associated with electronic legal research (Dkt. No. 644-1), and Perfect 10 presents no evidence to the contrary. Instead, Perfect 10's expert Mr. O'Connor opines that law firms frequently overstate the costs actually billed from vendors like Westlaw or Lexis to "recapture not just the charges to the research firm" but also costs for "IT personnel, library personnel, and general overhead expenses." (Dkt. No. 650-7, ¶65.) Notably, however, Perfect 10 offers no evidence to suggest that either Fenwick & West or Winston & Strawn engaged in such conduct, and Perfect 10's conclusion that "[t]hese internally-generated costs were likely just another profit center for Defendants' counsel" is entirely speculative. The research costs are allowable, appear reasonable on their face, and the Court finds no basis in the record to conclude that those costs are inflated above the actual costs Defendants incurred.

The costs items set forth in Mr. Bridges' Declaration (Dkt. Nos. 644-7; 644-8; 644-10) appear to have been reasonably necessary to defend against Perfect 10's copyright claims, and the Court awards Defendants their full request of $424,235.47 in non-taxable costs under 17 U.S.C. section 505.

IV. CONCLUSION

As discussed more fully above, Defendants are prevailing parties in this action, and the Court finds an award of attorneys' fees would serve the purposes of the Copyright Act. As prevailing parties, the Court must also award attorneys' fees under California Civil Code section 3344(a) for Defendants' successful defense against Perfect 10's statutory right of publicity claim. Although Defendants are not entitled to attorneys' fees for their defense of Perfect 10's common law right of publicity claim, the statutory and common law publicity claims were so inextricably intertwined that it would be impossible (and imprudent) to parse fees among those two claims - a fact Perfect 10 does not dispute.

However, Defendants concede they are not entitled to an award of attorneys' fees for their successful defense of Perfect 10's unfair competition claim. And because Perfect 10's conduct in litigating its trademark claim did not make this an "exceptional case," Defendants' are not entitled to an award of attorneys' fees on their trademark claim. Moreover, the Court reduces Defendants' fee request by $45,772.57 to account for certain clerical work billed at paralegal rates, and further reduces the fee award by 10 percent to account for what the Court thinks relatively modest "padding." In light of those reductions, the Court AWARDS Defendants a total of $5,213,117.06 in attorneys' fees as detailed in Appendix A below. 17 U.S.C. §505; Cal. Civ. Code §3344(a). The Court further concludes Defendants' request for allowable costs under 17 U.S.C. section 505 is reasonable under the circumstances and sufficiently documented. Accordingly, the Court further AWARDS Defendants at total of $424,235.47 in non-taxable costs. 17 U.S.C. §505.

The Court OVERRULES Perfect 10's evidentiary objections. (Dkt. No. 662.)

IT IS SO ORDERED Dated: March 24, 2015

/s/_________


HONORABLE ANDRÉ BIROTTE JR.


UNITED STATES DISTRICT COURT JUDGE