In re: High-Tech Employee Antitrust Litigation

Filing 531

ORDER by Judge Lucy H. Koh granting 418 Motion to Certify Class (lhklc1, COURT STAFF) (Filed on 10/24/2013)

1 2 3 4 5 6 7 UNITED STATES DISTRICT COURT 9 NORTHERN DISTRICT OF CALIFORNIA 10 United States District Court For the Northern District of California 8 SAN JOSE DIVISION 11 12 IN RE: HIGH-TECH EMPLOYEE ANTITRUST LITIGATION 13 14 15 16 THIS DOCUMENT RELATES TO: 17 ALL ACTIONS 18 19 ) ) ) ) ) ) ) ) ) ) ) ) ) Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION Plaintiffs Michael Devine, Mark Fichtner, Siddharth Hariharan, Brandon Marshall, and 20 Daniel Stover (collectively, “Plaintiffs”), individually and on behalf of a class of all those similarly 21 situated, allege antitrust claims against their former employers, Adobe Systems Inc. (“Adobe”), 22 Apple Inc. (“Apple”), Google Inc. (“Google”), Intel Corp. (“Intel”), Intuit Inc. (“Intuit”), Lucasfilm 23 Ltd. (“Lucasfilm”), and Pixar (collectively, “Defendants”). Plaintiffs allege that Defendants 24 conspired to suppress, and actually did suppress, employee compensation to artificially low levels 25 by agreeing not to solicit each other’s employees in violation of Section 1 of the Sherman Antitrust 26 Act, 15 U.S.C. § 1, and Section 4 of the Clayton Antitrust Act, 15 U.S.C. § 15. 27 On April 5, 2013, the Court granted in part and denied in part Plaintiffs’ Motion for Class 28 Certification with leave to amend. See Apr. 5 Class Cert. Order, ECF No. 382. Currently before 1 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 the Court is Plaintiffs’ Supplemental Motion for Class Certification. See Pls.’ Suppl. Mot. Class. 2 Cert. (“Suppl. Class Cert. Mot.”), ECF No. 418. Defendants filed an opposition, see Defs.’ Opp’n 3 to Pls.’ Suppl. Mot. Class Cert. (“Suppl. Opp’n”), ECF No. 439, and Plaintiffs filed a reply, Pls.’ 4 Reply Supp. Suppl. Mot. Class. Cert. (“Suppl. Reply”), ECF No. 455. The Court held a hearing on 5 Plaintiffs’ Supplemental Motion for Class Certification on August 8, 2013. See ECF No. 495. 6 Having considered the parties’ submissions, arguments, the relevant law, and the record in this 7 case, the Court GRANTS Plaintiffs’ Supplemental Motion for Class Certification and CERTIFIES 8 Plaintiffs’ proposed class of technical employees (“Technical Class”). 9 I. United States District Court For the Northern District of California 10 11 12 BACKGROUND A. Factual Background 1. The Parties Defendants are leading high tech companies, each with a principal place of business in the 13 San Francisco-Silicon Valley area of California. Apple is a market leader in consumer computer 14 products and software. Defs.’ Opp’n to Pls. Mot. Class. Cert. (“Opp’n”) at 5, ECF No. 209. In 15 2011, Apple’s total revenues exceeded $108 billion. Id. Google is the world’s leading internet 16 search provider. Report of Dr. Edward E. Leamer (“Leamer Rep.”) ¶ 15, ECF No. 190. Google 17 went public in 2004, and reached revenues of nearly $38 billion in 2011. Id. Intel is the world’s 18 largest semiconductor chip maker. Id. ¶ 16. In 2011, Intel earned approximately $54 billion. 19 Adobe specializes in digital media and marketing software. See id. ¶ 13. In 2009, Adobe earned 20 nearly $3 billion in revenues. Id. Intuit specializes in financial planning and tax preparation 21 programs. Opp’n at 5. In 2011, the company’s revenues exceeded $3.8 billion. Leamer Rep. ¶ 17. 22 Lucasfilm is a film production company known for its computer animation expertise and for 23 producing box office hits including the Star Wars films and the Indiana Jones franchise. Id. ¶ 18. 24 Pixar is a leading computer animation film studio. Id. ¶ 19. In 2006, Walt Disney Productions 25 acquired Pixar for approximately $7.4 billion. Id. 26 Named Plaintiffs are software engineers who were former employees of Defendants. 27 Devine worked for Adobe in the State of Washington from October of 2006 to July of 2008. See 28 Consolidated Amended Complaint (“CAC”) ¶ 16, ECF No. 65; Decl. Ann B. Shaver in Supp. Pls.’ 2 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Mot. for Class Cert. (“Shaver Decl.”), Ex. 6 ¶ 1, ECF No. 291. Fichtner worked for Intel in 2 Arizona from July of 1993 through November of 2006 and again from May of 2008 through May 3 of 2011. See CAC ¶ 17; Shaver Decl., Ex. 7 ¶ 1. Hariharan worked for Lucasfilm in California 4 from January of 2007 through August of 2008. See CAC ¶ 18; Shaver Decl., Ex. 8 ¶ 1. Marshall 5 worked for Adobe in California from July of 2006 through December of 2006. See CAC ¶ 19; 6 Shaver Decl., Ex. 9 ¶ 1. Finally, Stover worked for Intuit in California from at least November of 7 2006 through December of 2009. See CAC ¶ 20; Shaver Decl., Ex. 10 ¶ 1. 8 2. Market for High Tech Employees Plaintiffs assert that in a properly functioning and lawfully competitive labor market, each 10 United States District Court For the Northern District of California 9 Defendant would compete for employees by soliciting current employees from one or more of the 11 other Defendants. See CAC ¶ 41. This method of recruiting, to which Defendants refer as “cold 12 calling,” includes communicating directly in any manner—including orally, in writing, 13 telephonically, or electronically—with another company’s employee who has not otherwise 14 applied for a job. Id. 15 Plaintiffs allege that cold calling is a key competitive tool that companies use to recruit 16 employees, particularly high tech employees with advanced skills and abilities. Id. ¶ 45. Through 17 recruiting employees from competitors, a company is able to take advantage of the efforts its rival 18 has expended in soliciting, interviewing, and training skilled labor, while simultaneously inflicting 19 a cost on the rival by removing an employee on whom the rival may depend. Id. ¶ 44. 20 Plaintiffs further contend that the use of cold calling among Defendants commonly 21 increases total compensation and mobility for all of Defendants’ employees. See id. ¶¶ 48, 50. 22 Most directly, Plaintiffs allege that the practice of cold calling provides the recipient of a cold call 23 with opportunities to secure higher wages either by switching to a rival company or by negotiating 24 increased compensation with the recipient’s current employer. Id. ¶ 46. Plaintiffs further allege 25 that the compensation effects of cold calling are not limited to those individuals who receive the 26 calls. Rather, Plaintiffs allege, the effects of cold calling (and the effects of eliminating cold 27 calling) have a broader, common impact on Defendants’ salaried employees, especially their 28 technical employees. Id. ¶ 50. 3 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 3. Defendants’ Alleged Conspiracy Between approximately 2005 and 2009, Defendants Adobe, Apple, Google, Intel, Intuit, 3 Lucasfilm, and Pixar allegedly engaged in an “overarching conspiracy” to eliminate competition 4 among Defendants for skilled labor. Id. ¶ 55; see also id. ¶¶ 1, 2. The conspiracy consisted of an 5 interconnected web of express bilateral agreements among Defendants to abstain from actively 6 soliciting each other’s employees. Id. ¶ 55. 1 Plaintiffs allege that each agreement involved a 7 company under the control of Steve Jobs (Co-Founder, Former Chairman, and Former CEO of 8 Apple) and/or a company that shared at least one director with Apple’s Board of Directors. Id. ¶¶ 9 55, 57. Defendants memorialized these nearly identical agreements in CEO-to-CEO emails and 10 United States District Court For the Northern District of California 2 other documents, including “Do Not Call” lists, thereby putting each Defendant’s employees off- 11 limits to other Defendants. Pls.’ Mot. Class Cert. (“Class Cert. Mot.”) at 1, ECF No. 187. Each 12 bilateral agreement applied to all employees of a given pair of Defendants. See CAC ¶¶ 63, 76, 81, 13 88, 100, 105. These agreements were not limited by geography, job function, product group, or 14 time period. Nor were they related to any specific business or other collaboration between 15 Defendants. Id. 16 Plaintiffs allege that “Defendants entered into the express agreements and entered into the 17 overarching conspiracy with knowledge of the other Defendants’ participation, and with the intent 18 of accomplishing the conspiracy’s objective: to reduce employee compensation and mobility 19 through eliminating competition for skilled labor.” Id. ¶ 55. Plaintiffs also allege that Defendants’ 20 senior executives actively concealed each bilateral agreement and that Defendants’ employees 21 generally were not informed of, nor did they agree to, the terms of any of the agreements. Id. ¶¶ 22 55, 108. B. 23 1. 24 27 28 Department of Justice Investigation From 2009 through 2010, the Antitrust Division of the United States Department of Justice 25 26 Procedural Background (“DOJ”) investigated Defendants’ employment and recruitment practices. Id. ¶¶ 3, 111. 1 The parties refer to these agreements as “Do Not Cold Call” agreements, anti-solicitation agreements, anti-poaching agreements, and anti-competitive agreements. In this Order, the Court refers to these agreements as “anti-solicitation agreements.” 4 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION Following its investigation, the DOJ filed complaints in federal court against Defendants. See 2 United States v. Adobe Systems Inc., No. 10-1629, at 2 (D.D.C. Mar. 7, 2011) (“DOJ Adobe J.”), 3 ECF No. 79-1, Ex. A; United States v. Lucasfilm, Inc., No. 10-2220, 2011 WL 2636850, at *1 4 (D.D.C. June 3, 2011) (“DOJ Lucasfilm J.”), ECF No. 79-1, Ex. B. The DOJ also filed stipulated 5 proposed final judgments in each case. See DOJ Adobe J.; DOJ Lucasfilm J. In these stipulated 6 proposed final judgments, Defendants did not admit any wrongdoing or violation of law, but they 7 agreed to be “enjoined from attempting to enter into, maintaining or enforcing any agreement with 8 any other person or in any way refrain[ing] [from] . . . soliciting, cold calling, recruiting, or 9 otherwise competing for employees of the other person.” DOJ Adobe J. at 5; DOJ Lucasfilm J., at 10 United States District Court For the Northern District of California 1 4; CAC ¶ 115. The District Court for the District of Columbia entered the stipulated proposed final 11 judgments on March 17, 2011, and June 3, 2011. See DOJ Adobe J. at 12; DOJ Lucasfilm J. at 1; 12 CAC ¶ 115. 2 13 2. 14 The Instant Action a. 15 Removal, Case Consolidation, and Preliminary Motions The five cases underlying this consolidated action were initially filed in California Superior 16 Court. Hariharan v. Adobe Sys. Inc., Case No. 11-CV-574066 (Alameda Super. Ct. filed May 4, 17 2011), ECF No. 1; Marshall v. Adobe Sys. Inc., Case No. 11-CV-204052 (Santa Clara Super. Ct. 18 filed June 28, 2011), ECF No. 43-2; Devine v. Adobe Sys. Inc., Case No. 11-CV-204053 (Santa 19 Clara Super. Ct. filed June 28, 2011), ECF No. 43-1; Fichtner v. Adobe Sys. Inc., Case No. 11-CV- 20 204187 (Santa Clara Super. Ct. filed June 30, 2011), ECF No. 43-3; Stover v. Adobe Sys. Inc., Case 21 No. 11-CV-205090 (Santa Clara Super. Ct. filed July 14, 2011), ECF No. 43-4. 3 22 Defendants subsequently removed these five state court actions to the United States District 23 Court for the Northern District of California. Hariharan v. Adobe Sys. Inc., Case No. 11-2509 24 (removed May 23, 2011), see ECF No. 1; Marshall v. Adobe Sys. Inc., Case No. 11-3538 (removed 25 July 19, 2011), see ECF No. 41; Devine v. Adobe Sys. Inc., Case No. 11-3539 (removed July 19, 26 27 28 2 Under the provisions of Section 5(a) of the Clayton Act, 15 U.S.C. § 16(a), the proposed final judgment has no prima facie effect in any subsequent private lawsuit brought against Defendants. 3 While the name of each Superior Court case listed only Adobe as the defendant, the complaints also named Apple, Google, Intel, Intuit, Lucasfilm, Pixar, and Does 1-200, as defendants. 5 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 2011), see ECF No. 41; Fichtner v. Adobe Sys. Inc., Case No. 11-3540 (removed July 19, 2011), 2 see ECF No. 41; and Stover v. Adobe Sys. Inc., Case No. 11-3541 (removed July 19, 2011), see 3 ECF No. 41. On June 1, 2011, the lead case, Hariharan v. Adobe Systems Inc., was reassigned 4 from Magistrate Judge Spero to Judge Armstrong. See ECF No. 24. 5 On July 19, 2011, Defendants collectively filed a motion to relate the five underlying 6 actions. See ECF No. 41. In the Motion to Relate, Defendants stated that, “[b]ecause the cases 7 involve substantially the same parties, events and allegations, and because it appears likely that 8 there would be an unduly burdensome duplication of labor and expense or conflicting results if 9 they were heard before different judges, Defendants believe they are related[.]” Id. at 3. Judge United States District Court For the Northern District of California 10 Armstrong granted the Motion to Relate on July 27, 2011. See ECF No. 52. On August 2, 2011, 11 Plaintiff Siddharth Hariharan moved to transfer the related actions to the San Jose Division. See 12 ECF No. 56. Judge Armstrong granted the Motion to Transfer on August 4, 2011. See ECF No. 13 58. 14 On August 5, 2011, the five related underlying actions were reassigned to the undersigned 15 judge. See ECF No. 60. On September 6, 2011, the parties filed a joint stipulation to consolidate 16 the individual cases to “avoid duplication and unnecessary costs, and . . . [to] promote the efficient 17 conduct of proceedings.” ECF No. 63 at 2. Pursuant to this joint stipulation, the Court 18 consolidated the five underlying actions on September 12, 2011. See ECF No. 64. Plaintiffs filed 19 the Consolidated Amended Complaint on September 13, 2011. See CAC. 20 Defendants filed a Joint Motion to Dismiss the CAC on October 13, 2011, see ECF No. 79, 21 and, with leave of the Court, Lucasfilm filed its separate Motion to Dismiss on October 17, 2011, 22 see ECF No. 83. Following a hearing on January 26, 2012, see ECF No. 108, the Court granted in 23 part and denied in part Defendants’ Joint Motion to Dismiss and denied Lucasfilm’s Motion to 24 Dismiss on April 18, 2012, see ECF No. 119. 25 26 b. Initial Motion for Class Certification On October 1, 2012, Plaintiffs filed their Motion for Class Certification along with an 27 expert report in support of the motion. See Class Cert. Mot.; Leamer Rep. In their class 28 certification motion, Plaintiffs sought certification of an “All Employee” class, which included 6 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 every salaried employee throughout the United States who worked for Defendants between 2005 2 and 2009. Class Cert. Mot. at 1. Plaintiffs estimated that this class included more than 100,000 3 employees. See id. at 5. In the alternative, Plaintiffs sought certification of a more limited class of 4 salaried technical, creative, and research and development employees (“Technical Class”). Class 5 Cert. Mot. at 1. 6 On November 12, 2012, Defendants filed their Opposition to Plaintiffs’ Motion for Class 7 Certification, see Opp’n, as well as a Motion to Strike the expert report and declarations submitted 8 by Plaintiffs, see ECF No. 210. Plaintiffs then filed their Consolidated Reply in Support of Class 9 Certification and in Opposition to the Motion to Strike on December 10, 2012. Pls.’ Consol. Reply United States District Court For the Northern District of California 10 in Supp. of Mot. (“Reply”), ECF No. 247. On January 9, 2013, Defendants filed a Joint 11 Administrative Motion for Leave to Supplement the Record in Support of Defendants’ Opposition 12 to Class Certification, see ECF No. 263, to which Plaintiffs filed an opposition, see ECF No. 270. 13 The Court held a hearing on Plaintiffs’ Motion for Class Certification on January 17, 2013. See 14 ECF No. 282. 15 On April 5, 2013, the Court granted in part and denied in part Plaintiffs’ Motion for Class 16 Certification with leave to amend. See Apr. 5 Class Cert. Order. Specifically, the Court declined 17 to certify the class, but it did confirm as final the Court’s prior interim appointment of Lieff, 18 Cabraser, Heimann & Bernstein, LLP, and the Joseph Saveri Law Firm as Co-Lead Counsel and 19 appointed as Class Counsel the law firms that had served on the Executive Committee, Berger & 20 Montague, P.A. and Grant & Eisenhofer, P.A. Id. at 47. The Court recognized that these firms 21 have vigorously prosecuted this action and will continue to do so. Id. 22 The Court further denied Defendants’ Motion to Strike and granted in part and denied in 23 part Plaintiffs’ request to strike Defendants’ expert report and certain employee declarations. Id. at 24 49-52. Finally, the Court denied Defendants’ Joint Administrative Motion for Leave to 25 Supplement the Record in Support of Defendants’ Opposition to Class Certification. Id. at 52. 26 In granting in part and denying in part Plaintiffs’ Motion for Class Certification, the Court 27 first noted that Defendants did not dispute Plaintiffs’ assertion that both of the proposed classes 28 satisfied the requirements of Rule 23(a): numerosity, commonality, typicality, and adequacy of 7 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 representation. Id. at 9 (citing Class Cert. Mot. at 4-6; Tr. of Jan. 17, 2013 Class Cert H’rg (“Jan. 2 17 Tr.”) at 5:10-15, ECF No. 321). As such, the Court focused its analysis on whether Plaintiffs’ 3 proposed classes satisfied Rule 23(b)(3)’s requirement that common questions predominate. In so 4 doing, the Court found that “the adjudication of Defendants’ alleged antitrust violation will turn on 5 overwhelmingly common legal and factual issues.” Id. at 13. In addition, the Court found that 6 Plaintiffs satisfied their Rule 23(b)(3) burden on the issue of the predominance of common issues 7 with respect to damages. Id. at 44. 8 9 However, the Court could not find, based on the evidence available to Plaintiffs at the time of the initial Class Certification Motion, that Plaintiffs adequately demonstrated that common United States District Court For the Northern District of California 10 issues with regard to the impact of the alleged violation on members of the All Employee Class or 11 Technical Class would predominate under Rule 23(b)(3). Id. at 44-45. Although Plaintiffs’ 12 documentary evidence weighed “heavily in favor of finding that common issues predominate over 13 individual ones for the purpose of being able to prove antitrust impact,” the Court expressed 14 concern that Plaintiffs’ examples—such as email exchanges between CEOs and discrete human 15 resources documents from certain Defendants in particular years—might not be sufficient. See id. 16 at 33. The Court found that Plaintiffs might need “additional [documentary] support or empirical 17 analysis” to demonstrate that common evidence could be used to prove that all or nearly all 18 100,000 members of the All Employee Class were affected by the anti-solicitation agreements. See 19 id. In particular, the Court found that additional documentary support or empirical analysis would 20 be important to assure the Court that common issues predominated over individual issues given 21 that Defendants contested many of the factual bases of Plaintiffs’ theories of harm and actively 22 criticized the reliability, admissibility, and persuasiveness of Plaintiffs’ statistical analyses. Id. 23 The Court also questioned whether Plaintiffs’ All Employee Class was overly broad and noted that 24 Plaintiffs’ initial Motion for Class Certification provided little discussion or analysis to support 25 certifying Plaintiffs’ more limited Technical Class over the All Employee Class. See id. at 29. 26 The Court afforded Plaintiffs leave to amend to address the Court’s concerns. See id. at 52. 27 In so doing, the Court made clear that it was keenly aware that Defendants had failed to produce 28 significant amounts of discovery or make key witnesses available for depositions until after the 8 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 hearing on Plaintiffs’ Motion for Class Certification. Id. at 47. The Court noted that Defendants’ 2 failure to produce documents and witnesses hindered Plaintiffs’ efforts to demonstrate that their 3 proposed classes satisfied the Rule 23 requirements. Id. 4 On May 10, 2013, Plaintiffs filed their Supplemental Motion for Class Certification, which 5 focuses on demonstrating that common issues predominate for the purpose of satisfying the 6 requirements of Rule 23(b)(3) with respect to the Technical Class. Suppl. Class Cert. Mot. at 2. In 7 the Supplemental Motion, Plaintiffs specifically address the Court’s concerns regarding the 8 evidence of predominance with respect to the impact of the antitrust violation on all or nearly all of 9 the Technical Class. See id. at 22-25. Defendants filed their Opposition to Plaintiffs’ United States District Court For the Northern District of California 10 Supplemental Motion for Class Certification on June 21, 2013. See Suppl. Opp’n. Plaintiffs then 11 filed their Reply in support of their Supplemental Motion for Class Certification on July 12, 2013. 12 Suppl. Reply. Subsequently, Defendants filed objections to certain evidence in Plaintiffs’ Reply. 13 ECF No. 469. Plaintiffs then filed a motion to enforce Local Rule 7-3(d)(1) and to strike 14 Defendants’ improper Sur Reply, ECF No. 479, to which Defendants filed an Opposition, ECF No. 15 485. 4 The parties have filed various motions for leave to file statements of recent decisions while 16 the Supplemental Motion for Class Certification has been pending before this Court. ECF Nos. 17 491, 496, 498, 499, 505. 18 On July 12, 2013, Plaintiffs’ Co-Lead Class Counsel filed a letter informing the Court that 19 Plaintiffs and Defendants Pixar and Lucasfilm Ltd. reached an agreement to settle all individual 20 and class claims alleged in the CAC on behalf of Plaintiffs’ proposed Technical Class. See ECF 21 No. 453. On July 30, 2013, Plaintiffs’ Co-Lead Counsel filed a similar letter informing the Court 22 4 23 24 25 26 27 28 The Court GRANTS IN PART and DENIES IN PART Defendants’ Objections to Evidence in Pls.’ Reply Supp. Suppl. Class Cert., ECF No. 469. The Court GRANTS Defendants’ Motion to Strike Ms. Sandberg’s declaration as improper. Id. (citing Contratto v. Ethicon, 227 F.R.D. 304, 308 n.5 (N.D. Cal. 2005) (striking witness declaration because it was an “attempt to introduce new evidence in connection with their reply papers.”)). The Court DENIES Defendants’ request to supplement the record as to the deposition testimony of Dr. Murphy. All of the excerpts to which Defendants cite were properly submitted to the Court by Plaintiffs. Defendants’ request constitutes impermissible “further argument on the motion.” Civ. L. R. 7-3(d)(1). The Court GRANTS Defendants’ request to supplement the record as to the deposition testimony of Dr. Shaw as the excerpts to which Defendants cite were not submitted by Plaintiffs, and thus may be provided to complete the record. The Court DENIES Defendants’ request to strike Dr. Leamer’s analyses regarding salary ranges because it is proper rebuttal analysis. The Court GRANTS Defendants’ request to strike Dr. Leamer’s “superadditive” theory in his rebuttal report. 9 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 that Plaintiffs and Defendant Intuit also reached an agreement to settle all individual and class 2 claims alleged in the CAC on behalf of Plaintiffs’ proposed Technical Class. See ECF No. 489. 3 On September 21, 2013, Plaintiffs and Defendants Pixar, Lucasfilm, and Intuit filed a Motion for 4 Preliminary Approval of Class Settlement. ECF No. 501. That motion is pending before this 5 Court. Due to the settlements, Plaintiffs now only seek certification of a class for litigation 6 purposes against Defendants Adobe, Apple, Google, and Intel. See Suppl. Reply at 1 n.1. 7 Nonetheless, all parties agree that the settlements do not have any impact on Plaintiffs’ Supplemental Motion for Class Certification. See Pls.’ Br. Re Impact of the Proposed Settlement 9 on Pls.’ Suppl. Mot. Class Cert., ECF No. 483 (noting that the settlements “preserve[] Plaintiffs’ 10 United States District Court For the Northern District of California 8 right to litigate against the non-settling Defendants for the entire amount of Plaintiffs’ damages 11 based on joint and several liability under the antitrust laws.”); Defs.’ Joint Br. Re the Impact of the 12 Proposed Pixar and Lucasfilm Settlements on the Suppl. Class Cert. Mot., ECF No. 484 (“Plaintiffs 13 agree that these proposed settlements have no effect on the pending motion for certification of the 14 Technical Class.”). Generally, the same Rule 23 standard applies for certification of a proposed 15 class, whether for litigation or settlement purposes. See Amchem Prod., Inc. v. Windsor, 521 U.S. 16 591, 619 (1997); Hanlon v. Chrysler Corp., 150 F.3d 1011, 1019-23 (9th Cir. 1999). 17 II. 18 PROPOSED CLASS DEFINITION Named Plaintiffs now seek to certify only a nationwide class of salaried technical, creative, 19 and research and development employees who worked for any Defendant while that Defendant 20 participated in at least one anti-solicitation agreement with another Defendant. Thus, Plaintiffs 21 bring before the Court a proposed class “comprising those technical employees whose work 22 contributed to Defendants’ core business functions, whom the Defendants heavily recruited and 23 jealously guarded, and who appear at the very crux of Defendants’ conspiracy and this case.” 24 Suppl. Mot. at 2. Specifically, Plaintiffs seek to certify a Technical Class defined as follows: 25 26 27 28 All natural persons who work in the technical, creative, and/or research and development fields that are employed on a salaried basis in the United States by one or more of the following: (a) Apple from March 2005 through December 2009; (b) Adobe from May 2005 through December 2009; (c) Google from March 2005 through December 2009; (d) Intel from March 2005 through December 2009; (e) Intuit from June 2007 through December 2009; (f) Lucasfilm from January 2005 10 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION through December 2009; or (g) Pixar from January 2005 through December 2009. Excluded from the Class are: retail employees; corporate officers, members of the boards of directors, and senior executives of all Defendants. 1 2 3 Id. at iii. 5 4 The proposed Technical Class consists of job titles identified in Appendix B to the Leamer 5 Report, ECF No. 190, including: (1) Software Engineers, (2) Hardware Engineers and Component 6 Designers, (3) Application Developers, (4) Programmers, (5) Product Developers, (6) User 7 Interface or User Experience Designers, (7) Quality Analysts, (8) Research and Development, (9) 8 Animators, Digital Artists, Creative Directors and Technical Editors, (10) Graphic Designers and 9 Graphic Artists, (11) Web Developers, (12) IT Professionals, (13) Systems Engineers and United States District Court For the Northern District of California 10 Administrators, and (14) employees classified as technical professionals by their employers. See 11 Leamer Rep., App. B, ¶ 155. Plaintiffs believe that this proposed class includes more than 50,000 12 people. Class Cert. Mot. at 5. 13 III. LEGAL STANDARD Class actions are governed by Rule 23 of the Federal Rules of Civil Procedure. Rule 23 14 15 does not set forth a mere pleading standard. To obtain class certification, Plaintiffs bear the burden 16 of showing that they have met each of the four requirements of Rule 23(a) and at least one 17 subsection of Rule 23(b). Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1186, amended by 18 273 F.3d 1266 (9th Cir. 2001). “A party seeking class certification must affirmatively demonstrate 19 . . . compliance with the Rule.” Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2551 (2011). Rule 23(a) provides that a district court may certify a class only if: “(1) the class is so 20 21 numerous that joinder of all members is impracticable; (2) there are questions of law or fact 22 common to the class; (3) the claims or defenses of the representative parties are typical of the 23 claims or defenses of the class; and (4) the representative parties will fairly and adequately protect 24 the interests of the class.” Fed. R. Civ. P. 23(a). That is, the class must satisfy the requirements of 25 numerosity, commonality, typicality, and adequacy of representation to maintain a class action. 26 27 28 5 As discussed above, in the initial Motion for Class Certification, Plaintiffs also sought to certify an “All Employee Class” consisting of more than 100,000 employees. Class Cert. Mot. at 1. The Court denied certification of both the All Employee and Technical classes without prejudice. Plaintiffs have, in their Supplemental Motion, moved to certify only the Technical Class. Accordingly, this Order pertains only to that Class. 11 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Mazza v. Am. Honda Motor Co., Inc., 666 F.3d 581, 588 (9th Cir. 2012). Further, while Rule 23(a) 2 is silent as to whether the class must be ascertainable, courts have held that the Rule implies this 3 requirement as well. See, e.g., Herrera v. LCS Fin. Servs. Corp., 274 F.R.D. 666, 672 (N.D. Cal. 4 2011). 5 If all four prerequisites of Rule 23(a) are satisfied, a court must also find that Plaintiffs “satisfy through evidentiary proof” at least one of the three subsections of Rule 23(b). Comcast 7 Corp. v. Behrend, 133 S. Ct. 1426, 1432 (2013). Rule 23(b) sets forth three general types of class 8 actions. A class may be certified under Rule 23(b)(1) upon a showing that there is a risk of 9 substantial prejudice or inconsistent adjudications from separate actions. Fed. R. Civ. P. 23(b)(1). 10 United States District Court For the Northern District of California 6 A class may be certified under Rule 23(b)(2) if “the party opposing the class has acted or refused to 11 act on grounds that apply generally to the class, so that final injunctive relief or corresponding 12 declaratory relief is appropriate respecting the class as a whole.” Fed. R. Civ. P. 23(b)(2). Finally, 13 a class may be certified under Rule 23(b)(3) if a court finds that “questions of law or fact common 14 to class members predominate over any questions affecting only individual members, and that a 15 class action is superior to other available methods for fairly and efficiently adjudicating the 16 controversy.” Fed. R. Civ. P. 23(b)(3). 17 “[A] court’s class-certification analysis must be ‘rigorous’ and may ‘entail some overlap 18 with the merits of the plaintiff’s underlying claim.’” Amgen Inc. v. Conn. Ret. Plans and Trust 19 Funds, 133 S. Ct. 1184, 1194 (2013) (quoting Dukes, 131 S. Ct. at 2551); see also Mazza, 666 F.3d 20 at 588 (“‘Before certifying a class, the trial court must conduct a ‘rigorous analysis’ to determine 21 whether the party seeking certification has met the prerequisites of Rule 23.’” (quoting Zinser, 253 22 F.3d at 1186)). This “rigorous” analysis applies to both Rule 23(a) and Rule 23(b). Comcast, 133 23 S. Ct. at 1432 (discussing how Congress included “addition[al] . . . procedural safeguards for (b)(3) 24 class members beyond those provided for (b)(1) or (b)(2) class members (e.g., an opportunity to 25 opt out)” and how a court has a “duty to take a ‘close look’ at whether common questions 26 predominate over individual ones”). 27 Nevertheless, “Rule 23 grants courts no license to engage in free-ranging merits inquiries at 28 the certification stage.” Amgen, 133 S.Ct. at 1194-95. “Merits questions may be considered to the 12 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 extent—but only to the extent—that they are relevant to determining whether the Rule 23 2 prerequisites for class certification are satisfied.” Id. at 1195. If a court concludes that the moving 3 party has met its burden of proof, then the court has broad discretion to certify the class. Zinser, 4 253 F.3d at 1186. 5 IV. 6 DISCUSSION The Supreme Court has long recognized that class actions serve a valuable role in the 7 enforcement of antitrust laws. As the Supreme Court stated in Hawaii v. Standard Oil Co., 405 8 U.S. 251, 262 (1972), “[e]very violation of the antitrust laws is a blow to the free-enterprise system 9 envisaged by Congress. This system depends on strong competition for its health and vigor, and United States District Court For the Northern District of California 10 strong competition depends, in turn, on compliance with antitrust legislation.” See also N. Pac. Ry. 11 Co. v. United States, 356 U.S. 1, 4 (1958) (“The Sherman Act was designed to be a comprehensive 12 charter of economic liberty aimed at preserving free and unfettered competition as the rule of trade. 13 It rests on the premise that the unrestrained interaction of competitive forces will yield the best 14 allocation of our economic resources . . . while at the same time providing an environment 15 conductive [sic] to the preservation of our democratic political and social institutions.”). 16 Thus, to “open[] the door of justice” to individuals harmed by antitrust violations while at 17 the same time penalizing antitrust violators, Congress chose to allow individuals to serve as private 18 attorneys general in antitrust actions and to recover treble damages for their injuries. See 19 Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 486 n.10 (1977) (citing the initial 20 Congressional debates concerning the Clayton Act’s damages provisions as evidence that the 21 sponsors saw treble damages both as a means of “giv(ing) the injured party ample damages for the 22 wrong suffered” and “as an important means of enforcing the law”); see also Zenith Radio Corp. 23 v. Hazeltine Research, Inc., 395 U.S. 100, 130-31 (1969) (“The purpose of giving private parties 24 treble-damage and injunctive remedies was not merely to provide private relief, but was to serve as 25 well the high purpose of enforcing the antitrust laws.”). As the Supreme Court noted in Reiter v. 26 Sonotone Corp., 442 U.S. 330, 344 (1979), “[t]hese private suits provide a significant supplement 27 to the limited resources available to the Department of Justice for enforcing the antitrust laws and 28 deterring violations.” 13 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Plaintiffs allege that Defendants entered into an “overarching conspiracy” to suppress 2 employee compensation to artificially low levels. CAC ¶ 55. According to Plaintiffs, Defendants’ 3 agreements restrained trade and were thus per se unlawful under Section 1 of the Sherman 4 Antitrust Act. Id. ¶ 2; Class Cert. Mot. at 1; see 15 U.S.C. § 1 (“Every contract, combination in the 5 form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several 6 States, or with foreign nations, is declared to be illegal.”); see also Rebel Oil Co. v. Atl. Richfield 7 Co., 51 F.3d 1421, 1431 (9th Cir. 1995) (explaining that Section 4 of the Clayton Act allows 8 private parties to sue antitrust violators for damages). 9 Plaintiffs contend that, although the DOJ ultimately put an end to Defendants’ illegal United States District Court For the Northern District of California 10 agreements, the government was unable to compensate the victims of the conspiracy. Plaintiffs 11 now bring this case as private attorneys general “to pick up where the DOJ left off, to seek 12 damages for themselves and for the Class.” Class Cert. Mot. at 1. 13 A. 14 Plaintiffs assert that their proposed Technical Class satisfies the elements of Rule 23(a): Rule 23(a) and Class Representatives 15 numerosity, commonality, typicality, and adequacy of representation. Class Cert. Mot. at 4-6; see 16 Fed. R. Civ. P. 23(a). Defendants do not contest that Plaintiffs have satisfied these requirements. 17 See Jan. 17 Tr. at 5:10-15. Nevertheless, the Court addresses each in turn. 18 First, the Court finds that Plaintiffs have satisfied Rule 23(a)(1)’s numerosity requirement. 19 Pursuant to Rule 23(a)(1), Plaintiffs must show that “the class is so numerous that joinder of all 20 members is impracticable.” Fed. R. Civ. P. 23(a)(1). Plaintiffs need not state the exact number of 21 potential class members, nor is there a bright-line minimum threshold requirement. In re Rubber 22 Chems. Antitrust Litig., 232 F.R.D. 346, 350-51 (N.D. Cal. 2005). Rather, the Court must examine 23 the specific facts of each case. Gen. Tel. Co. v. EEOC, 446 U.S. 318, 330 (1980). In this case, the 24 parties agree that the Technical Class includes approximately 60,000 employees. See Class Cert. 25 Mot. at 5; Opp’n at 4. The Court finds joinder of all members of this proposed class to be 26 impracticable. Thus, the numerosity requirement is satisfied. See Fed. R. Civ. P. 23(a)(1). 27 Second, the Court finds that Plaintiffs have satisfied Rule 23(a)(2)’s commonality 28 requirement. Rule 23(a)(2) requires that “there are questions of law or fact common to the class.” 14 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Dukes, 131 S. Ct. at 2250-51. To satisfy the commonality requirement, Plaintiffs must show that 2 the class members have suffered “the same injury,” meaning that class members’ claims must 3 “depend upon a common contention” of such a nature that “determination of its truth or falsity will 4 resolve an issue that is central to the validity of each [claim] in one stroke.” Id. at 2551 (internal 5 quotation marks and citation omitted). Plaintiffs must demonstrate not merely the existence of a 6 common question, but rather “the capacity of a classwide proceeding to generate common answers 7 apt to drive the resolution of the litigation.” Id. (internal quotation marks omitted) (emphasis in 8 original). Nevertheless, “for purposes of Rule 23(a)(2), ‘[e]ven a single [common] question’ will 9 do.” Id. at 2556 (internal punctuation and citations omitted). United States District Court For the Northern District of California 10 “Where an antitrust conspiracy has been alleged, courts have consistently held that ‘the 11 very nature of a conspiracy antitrust action compels a finding that common questions of law and 12 fact exist.’” In re TFT-LCD (Flat Panel) Antitrust Litig., 267 F.R.D. 583, 593 (N.D. Cal. 2010), 13 amended in part by No. 07-1827, 2011 WL 3268649 (N.D. Cal. July 28, 2011) (quoting In re 14 Dynamic Random Access Memory (DRAM) Antitrust Litig., No. 02-1486, 2006 WL 1530166, at *3 15 (N.D. Cal. June 5, 2006)). Antitrust liability alone constitutes a common question that “will 16 resolve an issue that is central to the validity” of each class member’s claim “in one stroke,” Dukes, 17 131 S. Ct. at 2551, “because proof of an alleged conspiracy will focus on defendants’ conduct and 18 not on the conduct of individual class members.” In re TFT-LCD (Flat Panel) Antitrust Litig., 267 19 F.R.D. at 310 (citing cases). Indeed, the Court has already found that “the adjudication of 20 Defendants’ alleged antitrust violation will turn on overwhelmingly common legal and factual 21 issues.” Apr. 5 Class Cert. Order at 13. Moreover, Defendants do not dispute that there are some 22 common issues of law and fact. See Jan. 17 Tr. at 18:2-8. Because Plaintiffs have demonstrated 23 the existence of at least one common question capable of generating a common answer (antitrust 24 liability), the Court finds that the proposed class meets the commonality requirement of Rule 25 23(a)(2). 26 Third, the Court finds that Plaintiffs have satisfied Rule 23(a)(3)’s typicality requirement. 27 Under the “permissive standards” of Rule 23(a)(3), “representative claims are ‘typical’ if they are 28 reasonably co-extensive with those of absent class members; they need not be substantially 15 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 identical.” Hanlon, 150 F.3d at 1020; accord Staton v. Boeing Inc., 327 F.3d 938, 957 (9th Cir. 2 2003). “The test of typicality is whether other members have the same or similar injury, whether 3 the action is based on conduct which is not unique to the named plaintiffs, and whether other class 4 members have been injured by the same course of conduct.” Hanon v. Dataproducts Corp., 976 5 F.2d 497, 508 (9th Cir. 1992) (internal quotation marks and citation omitted). The purpose of the 6 typicality requirement is to assure that the interests of the named representative align with the 7 interests of the class. See Ellis v. Costco Wholesale Corp., 657 F.3d 970, 984-85 (9th Cir. 2011). 8 In antitrust cases, “typicality usually ‘will be established by plaintiffs and all class members 9 alleging the same antitrust violations by defendants.’” Pecover v. Elec. Arts, Inc., No. 08-2820, United States District Court For the Northern District of California 10 2010 WL 8742757, at *11 (N.D. Cal. Dec. 21, 2010) (quoting In re Playmobil Antitrust Litig., 35 11 F. Supp. 2d 231, 241 (E.D.N.Y. 1998)). In this case, all class members, regardless of their 12 individual employers, allege the same injuries arising from common conduct: suppression of 13 compensation due to Defendants’ anti-solicitation agreements. See Aug. 8, 2013 Class Cert. Hr’g 14 Tr. (“Aug. 8 Tr.”) at 19:11-14, ECF No. 494 (“We’re alleging a single violation of the Sherman 15 Act, a single conspiracy . . . .”). Accordingly, the Court finds that the named Plaintiffs’ interests 16 align with the interests of the class, and the typicality requirement of Rule 23(a)(3) is satisfied. 6 17 Finally, the Court finds that Plaintiffs satisfy Rule 23(a)(4)’s adequacy requirement. Legal 18 adequacy of a class representative under Rule 23(a)(4) turns on two inquiries: (1) whether named 19 plaintiffs and their counsel have “any conflicts of interest with other class members,” and 20 (2) whether named plaintiffs and their counsel will “prosecute the action vigorously on behalf of 21 the class.” Hanlon, 150 F.3d at 1020. As stated previously, the named Plaintiffs and Technical 22 Class members share an interest in proving that Defendants’ conduct violated the antitrust laws and 23 6 24 25 26 27 28 At the August 8 hearing on the Supplemental Motion for Class Certification, the Court expressed concern regarding whether Hariharan, a former employee of Lucasfilm, could satisfy the typicality requirement given that Lucasfilm recently reached a settlement agreement with Plaintiffs. The Court is persuaded that Hariharan still satisfies the typicality requirement because he continues to have an active antitrust claim against the remaining members of the conspiracy. See Aug. 8 Tr. at 21:15-23 ([T]he people who worked for the settled companies during the class period still have active claims against the other members of the conspiracy because . . . all of the members of the combination conspiracy . . . are liable for one another’s conduct.”); see also In re TFT-LCD (Flat Panel) Antitrust Litig., 267 F.R.D. 594 (“[A] conspirator is jointly liable for everything done during the period of the conspiracy’s existence.”). 16 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 suppressed their compensation. In addition, the named Plaintiffs do not have any conflicts of 2 interest with class members. Shaver Decl. Ex. 6 ¶¶ 5-6; id., Ex. 7 ¶¶ 5-6; id., Ex. 8 ¶¶ 5-6; id., Ex. 3 9 ¶¶ 5-6; id., Ex. 10 ¶¶ 5-6. Thus, the Court finds that the Plaintiffs’ proposed Technical Class 4 satisfies the adequacy requirement. In addition to the four requirements explicitly provided in Rule 23(a), courts have held that 6 Rule 23(a) also implicitly requires that the class be ascertainable. See, e.g., Herrera, 274 F.R.D. at 7 672. A class definition is sufficient if the description of the class is “definite enough so that it is 8 administratively feasible for a court to ascertain whether an individual is a member.” O’Connor v. 9 Boeing N. Am. Inc., 184 F.R.D. 311, 319 (C.D. Cal. 1998) (internal citation omitted). In addition, 10 United States District Court For the Northern District of California 5 “the court must be able to [determine that] class members are included or excluded from the class 11 by reference to objective criteria.” 5 James W. Moore, Moore’s Federal Practice, § 23.21[3] 12 (Matthew Bender 3d ed.). 13 Here, Plaintiffs’ proposed Technical Class consists of the job titles identified in Appendix B 14 to the Leamer Report, as discussed above. See Leamer Rep., App. B. Dr. Leamer selected these 15 job titles based on Defendants’ own employment compensation data, which breaks jobs into 16 creative, technical, and research and development “job families.” Id. Kevin F. Hallock, a labor 17 economist and expert in compensation structure and design, reviewed the titles included in the 18 proposed “Technical Class” and confirmed the titles selected for inclusion in the Technical Class in 19 light of Defendants’ job families for technical workers. See Report of Kevin F. Hallock (“Hallock 20 Rep.”) ¶¶ 241-44, ECF No. 417. In addition, within the proposed Technical Class, Plaintiffs seek 21 to include only employees who filled these job titles at Defendants’ companies within a fixed 22 period of time (2005 to 2009 for all Defendants except Intuit, 2007 to 2009 for Intuit). Thus, the 23 Court finds that the class definition is ascertainable. 24 Having undertaken a “‘rigorous analysis’ to determine whether the party seeking 25 certification has met the prerequisites of Rule 23,” Mazza, 666 F.3d at 588, the Court finds that 26 Plaintiffs’ proposed Technical Class satisfies the numerosity, commonality, typicality, and 27 adequacy requirements. In addition, the Court finds that Plaintiffs’ proposed Technical Class is 28 ascertainable. Thus, Plaintiffs have satisfied the requirements set forth by Rule 23(a). 17 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Further, because the named Plaintiffs and Class members share an interest in proving that 2 Defendants’ conduct violated the antitrust laws and suppressed their compensation and do not have 3 any conflicts of interest, the Court now GRANTS Plaintiffs’ request to appoint Michael Devine, 4 Mark Fichtner, Siddharth Hariharan, Brandon Marshall, and Daniel Stover as class representatives. 5 B. Rule 23(b)(3): Predominance 6 Plaintiffs also contend that their proposed Technical Class satisfies the requirements of 7 Rule 23(b)(3). Defendants disagree. Specifically, Defendants argue that Plaintiffs’ proposed class 8 does not satisfy Rule 23(b)(3)’s predominance requirement because neither antitrust impact nor 9 damages can be proven on a classwide basis. Opp’n at 11; Suppl. Opp’n at 3-4. For the reasons United States District Court For the Northern District of California 10 discussed below, the Court finds that questions common to the class are likely to predominate over 11 any individual questions. 12 The predominance analysis focuses on “the legal or factual questions that qualify each class 13 member’s case as a genuine controversy” to determine “whether proposed classes are sufficiently 14 cohesive to warrant adjudication by representation.” Amchen Prods., 521 U.S. at 623; see also 15 Fed. R. Civ. P. 23(b)(3) (holding that, to certify a class, the court must find that “questions of law 16 or fact common to class members predominate over any questions affecting only individual 17 members” (emphasis added)). 18 “Considering whether questions of law or fact common to class members predominate 19 begins . . . with the elements of the underlying causes of action.” Erica P. John Fund, Inc. v. 20 Halliburton Co., 131 S. Ct. 2179, 2184 (2011). A court must analyze these elements to “determine 21 which are subject to common proof and which are subject to individualized proof.” In re TFT- 22 LCD (Flat Panel) Antitrust Litig., 267 F.R.D. at 311-13. 23 In this case, Plaintiffs allege a violation of Section 1 of the Sherman Antitrust Act, 15 24 U.S.C. § 1, and Section 4 of the Clayton Antitrust Act, 15 U.S.C. § 15. See CAC ¶¶ 119-135; 25 Class Cert. Mot. at 1. “[T]o establish an antitrust claim, plaintiffs typically must prove (1) a 26 violation of antitrust laws, (2) an injury they suffered as a result of that violation, and (3) an 27 estimated measure of damages.” In re New Motor Vehicles Canadian Export Antitrust Litigation 28 (“In re New Motors”), 522 F.3d 6, 19 n.18 (1st Cir. 2008). 18 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Before the Court discusses whether common questions predominate with regard to each of 2 these elements, the Court notes that the legal standards with respect to the predominance inquiry 3 are not altogether clear. Specifically, the Court notes that there is no binding authority discussing 4 the standard a court must apply in determining whether common issues predominate in a putative 5 class action alleging wage suppression resulting from antitrust violations.7 Further, recent 6 Supreme Court authority on class certification suggests that the law in this area remains somewhat 7 unsettled. 8 9 In Walmart v. Dukes, the Supreme Court rejected certification of a class of more than one million female Walmart employees in a Title VII case, holding that “[a] party seeking class United States District Court For the Northern District of California 10 certification must affirmatively demonstrate his compliance with the rule.” Dukes, 131 S. Ct. at 11 2551. The Supreme Court further noted that prior to certifying a class, a district court must engage 12 in a “rigorous analysis” that will “entail some overlap with the merits of the plaintiff’s underlying 13 claim.” Id. The Supreme Court held that the class could not be certified because the plaintiffs 14 could not demonstrate commonality under Rule 23(a). Id. at 2554-57. The Dukes plaintiffs had 15 relied on statistical evidence that women were paid less, anecdotal evidence of discrimination from 16 120 women, and a sociologist who opined that there was a culture of sex stereotyping at Walmart. 17 Id. at 2549. The Supreme Court found that this evidence was insufficient to establish commonality 18 7 19 20 21 22 23 24 25 26 27 28 In their initial opposition to Plaintiffs’ class certification motion, Defendants pointed to a series of antitrust class actions alleging wage suppression by horizontal agreement in which district courts outside this circuit denied class certification on the basis that individual issues of antitrust impact and damages would predominate over classwide issues. See Weisfeld v. Sun Chemical Corp., 210 F.R.D. 136 (D.N.J. 2002), aff’d by 84 F. App’x 257 (3d Cir. 2004); Reed v. Advocate Health Care, 268 F.R.D. 573 (N.D. Ill. 2009); Fleischman v. Albany Med. Ctr., No. 06-765, 2008 WL 2945993 (N.D.N.Y. July 28, 2008); In re Comp. of Managerial, Prof’l, & Technical Emps. Antitrust Litig., No. 02-2924, 2003 WL 26115698 (D.N.J. May 27, 2003). However, these cases are inapposite because they lack the comprehensive documentary record present in the instant case. Moreover, in the instant case, the comprehensive documentary record strongly supports the Plaintiffs’ experts’ theories. In addition, intervening authority such as Amgen has refined the predominance inquiry since the cases cited by Defendants were decided. This is particularly true with respect to Reed, a Northern District of Illinois case that predates the Seventh Circuit’s decisions in Butler v. Sears, Roebuck & Co., 722 F.3d 796 (7th Cir. 2013), and Messner v. Northshore Univ. HealthSystem, 669 F.3d 802 (7th Cir. 2012). These two Seventh Circuit cases suggest that the Reed court construed the predominance requirement too stringently. The Court further notes that other district courts have certified wage suppression antitrust classes, including one since the latest Supreme Court cases. See, e.g., Merenda v. VHS of Mich., Inc., No. 06-15601, 2013 WL 5106520 (E.D. Mich. Sept. 13, 2013); Johnson v. Ariz. Hosp. & Healthcare Ass’n., No. 07-1292, 2009 WL 5031334 (D. Ariz. July 14, 2009). 19 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 because it demonstrated no general policy of discrimination and no corporate direction of store 2 managers’ discretion. Id. at 2554. Accordingly, the Supreme Court, finding that commonality 3 required not just common questions but the capacity to generate a common answer in a classwide 4 proceeding, reversed the certification of the class. Id. at 2551. 5 Applying Dukes, the Ninth Circuit, in Ellis v. Costco vacated a district court’s certification 6 of a class. 657 F.3d at 988. The Ellis court, discussing commonality, held that district courts must 7 consider the underlying merits in addressing class certification issues. Id. at 983. The Ninth 8 Circuit further concluded that when there is a battle of the experts on class certification, “rigorous 9 analysis” requires district courts to determine not only admissibility of the experts’ statements, but United States District Court For the Northern District of California 10 11 also the “persuasiveness of the evidence presented.” Id. at 982. Two years after it decided Dukes, the Supreme Court affirmed the certification of a class of 12 plaintiffs who were alleging securities fraud in Amgen. In Amgen, the Supreme Court held that 13 plaintiffs need not prove materiality, one of the elements of plaintiffs’ securities fraud claim, at the 14 class certification stage; rather, the Supreme Court found that plaintiffs needed only demonstrate 15 that common questions would predominate. 133 S. Ct. at 1191. “Rule 23(b)(3) requires a showing 16 that questions common to the class predominate, not that those questions will be answered, on the 17 merits, in favor of the class.” Id. As such, “the office of a Rule 23(b)(3) certification ruling is not 18 to adjudicate the case; rather, it is to select the ‘metho[d]’ best suited to adjudication of the 19 controversy ‘fairly and efficiently.’” Id. Therefore, while class certification may require some 20 inquiry into the merits, “Rule 23 grants courts no license to engage in free-ranging merits inquiries 21 at the certification stage.” Id. at 1194. Importantly, the Supreme Court specified that plaintiffs 22 were not required to demonstrate that common questions would predominate with respect to each 23 element. Id. at 1196 (“Rule 23(b)(3), however, does not require a plaintiff seeking class 24 certification to prove that each ‘elemen[t] of [her] claim [is] susceptible to classwide proof.’” 25 (alterations and emphasis in original)). Rather, the inquiry is more holistic. 26 One month after Amgen, the Supreme Court, in Comcast Corp. v. Behrend, 133 S. Ct. 1426 27 (2013), reversed certification of a damages and liability class under Rule 23(b)(3). The district 28 court in Comcast had certified a class of more than two million Comcast subscribers who sought 20 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 damages for violations of federal antitrust laws. Id. at 1431. Plaintiffs in Comcast had proposed 2 four theories of antitrust impact, but the district court found that only one of those theories could be 3 proven using common evidence. Id. However, the district court certified a damages class 4 notwithstanding the fact that plaintiffs’ expert calculated damages using a model that did not 5 isolate the damages resulting from the one credited theory of antitrust impact. Id. The Supreme 6 Court reversed the certification under a “straightforward application of class-certification 7 principles” due to the disconnect between the theory of impact and the theory of damages. Id. at 8 1433. The Court stated that “[c]alculations [of damages] need not be exact, but at the class- 9 certification stage (as at trial), any model supporting a ‘plaintiff’s damages case must be consistent United States District Court For the Northern District of California 10 with its liability case, particularly with respect to the alleged anticompetitive effect of the 11 violation.” Id. Importantly, the Court suggested that some of the principles applied pursuant to 12 Rule 23(a) in Dukes also applied pursuant to Rule 23(b)(3). Specifically, the Court stated that Rule 13 23(b)(3) required a “rigorous analysis” and that Rule 23(b)(3) may require “inquiry into the merits 14 of the claim.” Id. at 1432-33. 15 The Ninth Circuit, in line with the Supreme Court’s admonition that Comcast created no 16 new law, read Comcast narrowly in Levya v. Medline Industries, 716 F.3d 510 (9th Cir. 2013). 17 There, the Ninth Circuit reversed a district court’s denial of class certification in a case concerning 18 alleged violations of California labor laws. The Ninth Circuit held that even after Comcast, under 19 Ninth Circuit law, the fact that damages calculations would require individualized inquiries does 20 not defeat certification of a Rule 23(b)(3) class. Id. at 513-14. Thus, so long as “damages will be 21 calculated based on the wages each employee lost due to [defendant]’s unlawful practices,” 22 Comcast does not pose a barrier to class certification. Id. 23 Echoing Levya, the Seventh Circuit in Butler v. Sears, Roebuck & Co., 727 F.3d 796 (7th 24 Cir. 2013), a post-Comcast consumer action alleging defects in washing machines, held that 25 individual questions with respect to damages do not defeat class certification. The Seventh Circuit 26 found that “[a] determination of liability could be followed by individual hearings to determine the 27 damages sustained by each class member.” Id. at 789. The Seventh Circuit distinguished Comcast 28 21 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 on the basis that “there is no possibility in this case that damages could be attributed to acts of the 2 defendants that are not challenged on a class-wide basis.” Id. at 800. 3 The Seventh Circuit in Butler approvingly cited its previous decision in Messner v. 4 Northshore Univ. HealthSystem, 669 F.3d 802 (7th Cir. 2012), where the Seventh Circuit had 5 reversed a denial of class certification. In Messner, the Seventh Circuit held that the predominance 6 inquiry does not require the total absence of individual questions, but rather that common questions 7 predominate over any individual questions. Id. at 815. Moreover, the Seventh Circuit in Messner 8 noted that for the purposes of predominance, the inquiry focused on whether common questions 9 predominate over individual questions—not whether plaintiffs could show “common answers to United States District Court For the Northern District of California 10 those questions.” Id. at 819. The Seventh Circuit further cautioned against turning class 11 certification into a “dress rehearsal for the trial on the merits” and stated that in antitrust cases, 12 even rigorous application of the class certification standard “will frequently lead to certification.” 13 Id. at 811, 815; see also Amchem, 521 U.S. at 625 (“Predominance is a test readily met in certain 14 cases alleging consumer or securities fraud or violations of the antitrust laws.”). Importantly, in 15 both Messner and Butler, the Seventh Circuit rejected the proposition that “predominance is 16 determined simply by counting noses: that is, determining whether there are more common issues 17 or more individual issues, regardless of relative importance.” Butler, 727 F.3d at 801. Rather, 18 “predominance requires a qualitative assessment too; it is not bean counting.” Id.; see also 19 Messner, 669 F.3d at 814 (“There is no mathematical or mechanical test for evaluating 20 predominance.”). 21 The Seventh Circuit decision in Butler cited and was in accord with the Sixth Circuit’s 22 post-Comcast decision in In re Whirlpool Corp. Front-Loading Washer Products Liability 23 Litigation, 722 F.3d 838 (6th Cir. 2013). There, the Sixth Circuit affirmed the certification of a 24 class, finding that the district court had adequately considered the merits in determining that 25 common questions would predominate over individual questions. The Sixth Circuit found that like 26 Amgen, the defendant’s liability in the consumer class action would be classwide or non-existent. 27 Id. at 859. The Sixth Circuit further distinguished Comcast on the basis that Comcast concerned a 28 class certified for damages purposes in addition to liability purposes. Id. Accordingly, the Sixth 22 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Circuit concluded that “the principles we glean from Amgen and Comcast Corp. include that to 2 satisfy Rule 23(b)(3), named plaintiffs must show, and district courts must find, that questions of 3 law or fact common to members of the class predominate over any questions that affect only 4 individual members.” Id. at 860. 5 The D.C. Circuit has further elaborated on the qualitative assessment required by the Sixth 6 and Seventh Circuits and concluded that such an assessment requires district courts to closely 7 scrutinize factual evidence and expert reports that demonstrate that impact can be proven on a 8 classwide basis. See In re Rail Freight Fuel Surcharge Antitrust Litig., 725 F.3d 244, 247 (D.C. 9 Cir. 2013). The D.C. Circuit held that “[c]ommon questions of fact cannot predominate where United States District Court For the Northern District of California 10 there exists no reliable means of proving classwide injury in fact.” Id. at 252-53. The D.C. Circuit 11 further stated that “[i]t is now indisputably the role of the district court to scrutinize the evidence 12 before granting certification.” Id. at 253. The D.C. Circuit therefore found that “[i]t is now clear . . 13 . that Rule 23 not only authorizes a hard look at the soundness of statistical models that purport to 14 show predominance—the rule commands it.” Id. at 255. Accordingly, the D.C. Circuit vacated the 15 district court’s certification of the class because there were methodological problems with the 16 plaintiffs’ expert reports that the district court had not considered. Id. at 252. 8 17 18 8 19 20 21 22 23 24 25 26 27 28 The First and Third Circuits in pre-Dukes, Amgen, and Comcast cases vacated district courts’ certification of classes under Rule 23(b)(3) and remanded for reconsideration based in part on theories that have been superseded by intervening Supreme Court authority. See In re New Motors, 522 F.3d at 8; In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 307 (3d Cir. 2008). For example, the First and Third Circuits relied on a theory that “[i]n antitrust class actions, common issues do not predominate if the fact of antitrust violation and the fact of antitrust impact cannot be established through common proof.” In re New Motors, 522 F.3d at 20; see also In re Hydrogen Peroxide Antitrust Litig., 552 F.3d at 311 (same). This appears to conflict with the Supreme Court’s conclusion in Amgen that “Rule 23(b)(3) . . . does not require a plaintiff seeking class certification to prove that each “elemen[t] of [her] claim [is] susceptible to classwide proof.” 133 S. Ct. at 1191 (emphasis in original). Nevertheless, some of the theories in In re New Motors and In re Hydrogen Peroxide Antitrust Litigation are consonant with Amgen. Specifically, the Third Circuit held that a rigorous analysis required resolving conflicts in expert opinions that go to certification and that a party’s indication that it intends to use common evidence is insufficient for certification. In re Hydrogen Peroxide Antitrust Litig., 552 F.3d at 321-23. Similarly, the First Circuit held that a district court should have engaged in a searching inquiry into the validity of a novel and complex theory of impact. In re New Motors, 522 F.3d at 27. Accordingly, the First Circuit found that a district court should have addressed criticisms of the plaintiffs’ expert’s methodology. Id. In light of the fact that these theories appear to be consistent with Supreme Court authority, the Court applies them in the instant action. 23 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Certain principles regarding the legal standard that this Court must apply in determining whether the Technical Class should be certified emerge from Walmart, Amgen, Comcast, and the 3 circuit court cases applying this Supreme Court authority. First, and most importantly, the critical 4 question that this Court must answer is whether common questions predominate over individual 5 questions. Amgen, 133 S. Ct. at 1191. In essence, this Court must determine whether common 6 evidence and common methodology could be used to prove the elements of the underlying cause of 7 action. Id. Second, in answering this question, this Court must conduct a “rigorous” analysis. 8 Comcast Corp., 133 S. Ct. at 1432. This analysis may overlap with the merits, but the inquiry 9 cannot require Plaintiffs to prove elements of their substantive case at the class certification stage. 10 United States District Court For the Northern District of California 2 Amgen, 133 S. Ct. at 1194. Third, this Court must determine not only the admissibility of expert 11 evidence that forms the basis of the methodology that demonstrates whether common questions 12 predominate. Ellis, 657 F.3d at 982. Rather, this Court must also determine whether that expert 13 evidence is persuasive, which may require the Court to resolve methodological disputes. Id.; see 14 also In re Rail Freight Fuel Surcharge Antitrust Litig., 725 F.3d at 255. Fourth, the predominance 15 inquiry is not a mechanical inquiry of “bean counting” to determine whether there are more 16 individual questions than common questions. Butler, 727 F.3d at 801. Instead, the inquiry 17 contemplates a qualitative assessment, which includes a hard look at the soundness of statistical 18 models. Id.; In re Rail Freight Fuel Surcharge Antitrust Litig., 725 F.3d at 255. Fifth, Plaintiffs 19 are not required to show that each element of the underlying cause of action is susceptible to 20 classwide proof. Amgen, 133 S. Ct. at 1196. Rather, they need only show that common questions 21 will predominate with respect to their case as a whole. Id. 22 With these principles in mind, this Court now turns to the elements and finds that common 23 questions predominate overall and with regard to all three elements—antitrust violation, antitrust 24 impact, and damages. 25 1. Antitrust Violation 26 Plaintiffs allege that Defendants engaged in an “overarching conspiracy” to eliminate 27 competition among one another for skilled labor, with the intent and effect of suppressing the 28 compensation and mobility of Defendants’ employees. CAC ¶¶ 1, 2, 55. Accordingly, Plaintiffs 24 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 contend that, at trial, “[p]roving Defendants’ conspiracy will be the overriding common issue for 2 every Class Member.” Class Cert. Mot. at 2. In support of their allegations, Plaintiffs have set 3 forth copious common evidence in the form of Defendants’ internal work documents, deposition 4 transcripts, and email exchanges between Defendants’ CEOs as well as other directors, officers, 5 and senior managers, all of which support Plaintiffs’ allegations that Defendants entered into 6 express agreements not to compete for one another’s employees. 7 Defendants concede that adjudication of Defendants’ alleged antitrust violations will turn 8 on common legal and factual issues. See Jan. 17 Tr. at 17:1-4 (Court: “Do you contest [the 9 antitrust violation] prong of the analysis?” Defendants’ Counsel: “Not for purposes of this United States District Court For the Northern District of California 10 motion.”). As stated in this Court’s April 5 Class Certification Order, this Court agrees and finds 11 that Plaintiffs have demonstrated that “adjudication of Defendants’ alleged antitrust violation will 12 turn on overwhelmingly common legal and factual issues.” Apr. 5 Class Cert. Order at 13. 13 To prevail on a cause of action for violation of Section 1 of the Sherman Act, a plaintiff 14 must show that: “(1) there was an agreement, conspiracy, or combination between two or more 15 entities; (2) the agreement was an unreasonable restraint of trade under either a per se or rule of 16 reason analysis; and (3) the restraint affected interstate commerce.” 9 Am. Ad Mgmt., Inc. v. GTE 17 Corp., 92 F.3d 781, 784 (9th Cir. 1996); see also Tanaka v. Univ. of S. Cal., 252 F.3d 1059, 1062 18 (9th Cir. 2001). Plaintiffs here present substantial evidence of Defendants’ antitrust violations, and 19 all of that evidence is common to the Technical Class as a whole. 20 Plaintiffs’ evidence indicates that the roots of Defendants’ conspiracy appear to reach back 21 to the mid-1980s, shortly after George Lucas (former Lucasfilm Chairman of the Board and CEO) 22 sold Lucasfilm’s “computer division,” a “tech, research, and development company,” to Steve Jobs 23 (Co-Founder, Former Chairman, Former CEO of Apple), who then renamed the division “Pixar.” 24 Decl. of Lisa Cisneros (“Cisneros Decl.”), Ex. NN (Lucas Depo.) at 16, 59, ECF No. 418-2. 25 George Lucas believed that companies should not compete against each other for employees, 26 because “[i]t’s not a normal industrial competitive situation.” Id. at 52. As George Lucas 27 explained, “I always—the rule we had, or the rule that I put down for everybody,” was that “we 28 9 The third factor, whether the alleged restraint affected interstate commerce, is not disputed. 25 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 cannot get into a bidding war with other companies because we don’t have the margins for that sort 2 of thing.” Id. at 44. Edward Catmull (Pixar President) agreed with George Lucas that the newly 3 independent Pixar would reciprocate this non-compete “rule” with Lucasfilm. The companies thus 4 agreed: (1) not to cold call each other’s employees; (2) to notify each other when making an offer 5 to an employee of the other company even if that employee applied for a job on his or her own 6 initiative; and (3) that any offer would be “final” and would not be improved in response to a 7 counter-offer by the employee’s current employer (whether Lucasfilm or Pixar). Shaver Decl., Ex. 8 3 (McAdams Depo.) at 145-46; Def. Lucasfilm Ltd.’s Am. Ans. to Consol. Am. Compl. ¶ 59, ECF 9 No. 168 (“Lucasfilm and Pixar had a general understanding that they would not actively solicit United States District Court For the Northern District of California 10 candidates from the other via cold-calling.”). 11 Steve Jobs (Co-Founder, Former Chairman, Former CEO of Apple), who was “very 12 adamant about protecting his employee force,” proceeded to expand the Pixar-Lucasfilm agreement 13 to include Apple and its labor competitors. Cisneros Decl., Ex. RR (Catmull Depo.) at 195. As 14 such, beginning no later than 2004, Pixar sought Steve Jobs’ permission before making offers of 15 employment to Apple employees, regardless of whether Pixar solicited the employee or the 16 employee applied independently. See Shaver Decl., Ex. 62 (showing email from Rob Cook (Pixar 17 Vice President of Advanced Technology) to Steve Jobs in 2004 requesting permission to make an 18 offer to an Apple employee). On April 30, 2007, Lori McAdams (Pixar Vice President of Human 19 Resources and Administration) and Danielle Lambert (Apple Head of Human Resources), 20 formalized the two companies’ understanding on the same terms as the “gentlemen’s agreement” 21 between Pixar and Lucasfilm. See id., Ex. 66 (Lori McAdams informing recruiting team about her 22 phone call with Danielle Lambert and that, “effective now, we’ll follow a gentleman’s agreement 23 with Apple that is similar to our Lucasfilm agreement. That is . . . we won’t directly solicit any 24 Apple employee (including outside recruiters if we use them) . . . . Danielle will ask her Recruiting 25 team to follow the same procedure.” (emphasis added)). 26 These agreements extended to other Defendants. On February 18, 2005, Bill Campbell 27 (Chairman of Intuit Board of Directors, Co-Lead Director of Apple, and advisor to Google) 28 assisted Steve Jobs (Co-Founder, Former Chairman, Former CEO of Apple) in entering into an 26 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 agreement with Eric Schmidt (Google Executive Chairman, Member of the Board of Directors, and 2 former CEO) at Google. See id., Ex. 17 (email from Bill Campbell to Steve Jobs informing Steve 3 Jobs that Eric Schmidt “got directly involved and firmly stopped all efforts to recruit anyone from 4 Apple”). That same day, Danielle Lambert (Apple Head of Human Resources) ordered her staff to 5 “[p]lease add Google to your ‘hands-off’ list. We recently agreed not to recruit from one another 6 so if you hear of any recruiting they are doing against us, please be sure to let me know. Please be 7 sure to honor our side of the deal.” Id., Ex. 23. Later that year, Arnnon Geshuri (Google 8 Recruiting Director) was asked to create a formal “Do Not Cold Call” list of companies, including 9 Apple, which had “special agreements” with Google not to compete for employees. See id., Ex. United States District Court For the Northern District of California 10 27. The draft was presented to Google’s Executive Management Group, a committee consisting of 11 Google’s senior executives, including Eric Schmidt, Larry Page (Google Co-Founder), Sergey Brin 12 (Google Co-Founder), and Shona Brown (former Google Senior Vice President, Business 13 Operations). See id., Ex. 28. Eric Schmidt approved the list. See id. (email from Eric Schmidt 14 stating “[t]his looks very good.”). When Shona Brown asked Eric Schmidt whether he had any 15 concerns with sharing information regarding the “Do Not Call” list with Google’s competitors, Eric 16 Schmidt responded that he preferred that it be shared “verbally[,] since I don’t want to create a 17 paper trail over which we can be sued later?” Id., Ex. 41. Shona Brown responded: “makes sense 18 to do orally. i agree.” Id. 19 Two months after entering into an agreement with Google, Steve Jobs (Co-Founder, Former 20 Chairman, Former CEO of Apple) persuaded Adobe to enter into a nearly identical agreement with 21 Apple. On May 26, 2005, Steve Jobs complained to Bruce Chizen (former Adobe CEO) that 22 Adobe was recruiting Apple employees. Id., Ex. 18. Bruce Chizen responded by saying, “I 23 thought we agreed not to recruit any senior level employees . . . . I would propose we keep it that 24 way. Open to discuss. It would be good to agree.” Id. Steve Jobs was not satisfied, and replied 25 by threatening to send Apple recruiters after Adobe’s employees: “OK, I’ll tell our recruiters that 26 they are free to approach any Adobe employee who is not a Sr. Director or VP. Am I 27 understanding your position correctly?” Id. Bruce Chizen immediately gave in: “I’d rather agree 28 NOT to actively solicit any employee from either company . . . . If you are in agreement I will let 27 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 my folks know.” Id. (emphasis in original). The next day, Theresa Townsley (Adobe Vice 2 President Human Resources) announced to her recruiting team, “Bruce and Steve Jobs have an 3 agreement that we are not to solicit ANY Apple employees, and vice versa.” Id., Ex. 19 (emphasis 4 in original). Adobe then placed Apple on its “[c]ompanies that are off limits” list, which instructed 5 Adobe employees not to cold call Apple employees. Id., Ex. 11. 6 In addition to Google’s anti-solicitation agreement with Apple, Google also entered into a 7 “no poaching policy” with Intel. See, e.g., id., Ex. 51 (email from Paul Otellini (CEO of Intel and 8 Member of the Google Board of Directors) to Intel recruiter, dated April 16, 2007, stating, “I have 9 an unofficial no poaching policy with [Google]”). Eric Schmidt (Google Executive Chairman, United States District Court For the Northern District of California 10 Member of the Board of Directors, and former CEO) confirmed this policy in a June 4, 2007, email 11 to Paul Otellini in which Eric Schmidt wrote, “I checked as to our recruiting policy with Intel. 12 ‘Intel has been listed on the Do Not Call List since the policy was created. No one in staffing 13 directly calls, networks, or emails into the company or its subsidiaries looking for talent.’ 14 Hopefully there are no exceptions to this policy and if you become aware of this please let me 15 know immediately!” Id., Ex. 56. Bill Campbell (Chairman of Intuit Board of Directors, Co-Lead 16 Director of Apple, and advisor to Google) was also involved in the Google-Intel agreement, as 17 reflected in an email exchange from 2006 in which Bill Campbell agreed with Jonathan Rosenberg 18 (Google Advisor to the Office of CEO and former Senior Vice President of Product Management) 19 that Google should call Paul Otellini before making an offer to an Intel employee, regardless of 20 whether the Intel employee first approached Google. See id., Ex. 37. Paul Otellini then forwarded 21 the email to Patricia Murray (Intel Senior Vice President and Director of Leadership Strategy and 22 former President of Human Resources), with a note stating, “FYI . . . Do not fwd.” Id. Two days 23 later, in an email entitled “global gentleman agreement with Google,” an Intel recruiter asked Paul 24 Otellini and another senior executive, “Are either of you aware of any agreement with Google that 25 prohibits us from recruiting Google’s senior talent?” Id., Ex. 52. Paul Otellini replied, “Let me 26 clarify. We have nothing signed. We have a handshake ‘no recruit’ between eric and myself. I 27 would not like this broadly known.” Id. Plaintiffs note that, while the DOJ alleged that the 28 Google-Intel agreement began no later than September of 2007, other evidence suggests that the 28 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 agreement began in 2005, the same time as other Defendants’ bilateral agreements. See id., Ex. 28 2 (showing Intel on Google’s “Do Not Call” list as early as 2005). In October of 2008, Intel also 3 agreed with Pixar that it “will not proactively pursue any Pixar employees going forward.” Shaver 4 Decl., Ex. 70. 5 Bill Campbell (Chairman of Intuit Board of Directors, Co-Lead Director of Apple, and advisor to Google) also insisted that Google and Intuit enter into a non-compete agreement. 7 Although Google’s “non-solicit policy” initially “cover[ed] only 18 Intuit employees,” “Bill 8 [Campbell] requested that Intuit be added fully to the Do Not Call list.” Id., Ex. 31 (email, dated 9 June 6, 2007, between Arnnon Geshuri (Google Recruiting Director) and Eric Schmidt (Google 10 United States District Court For the Northern District of California 6 Executive Chairman, Member of the Board of Directors, and former CEO)). Thus, by June 12, 11 2007, Intuit was added fully to the list. Id., Ex. 26; id. Ex. 30 (“please update the DNC list to now 12 include Intuit 100% do not call.”); see also Decl. of Dean Harvey (“Harvey Decl.”), Ex. 25 at 13, 13 ECF No. 248 (stating that “Google’s [Do Not Cold Call] policy for Intuit began in April 2006 and 14 was broadened in June 2007 to include all Intuit employees.”). 15 Ultimately, by the time that the DOJ investigation began in the summer of 2009, Apple’s 16 own “Hands Off (Do Not Call List)” included every Defendant. See Shaver Decl., Ex. 22. Steve 17 Jobs (Co-Founder, Former Chairman, and Former CEO of Apple) also tried, unsuccessfully, to 18 enter into a similar agreement with Palm, Inc. (“Palm”) following a period in which several 19 employees moved between the two companies. On August 22, 2007, Steve Jobs called Edward 20 Colligan (former President and CEO of Palm) to propose “an arrangement between Palm and 21 Apple by which neither company would hire the other’s employees, including high tech 22 employees.” Decl. of Edward Colligan (“Colligan Decl.”) ¶ 6, ECF No. 293. Steve Jobs also 23 threatened to retaliate against Palm if Palm resisted. Id. (“Mr. Jobs also suggested that if Palm did 24 not agree to such an arrangement, Palm could face lawsuits alleging infringement of Apple’s many 25 patents.”). Edward Colligan nevertheless refused, writing to Steve Jobs, “[y]our proposal that we 26 agree that neither company will hire the other’s employees, regardless of the individual’s desires, is 27 not only wrong, it is likely illegal.” Id., Ex. A. Edward Colligan stated further, “I can’t deny 28 people who elect to pursue their livelihood at Palm the right to do so simply because they now 29 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 work for Apple, and I wouldn’t want you to do that to current Palm employees.” Id. Edward 2 Colligan held firm against Steve Jobs’ threats, stating, “I want to be clear that we are not 3 intimidated by your threat . . . . If you choose the litigation route, we can respond with our claims 4 based on [Palm’s] patent assets, but I don’t think litigation is the answer.” Id. Steve Jobs 5 responded: 6 7 8 This is not satisfactory to Apple . . . . We must do whatever we can to stop this. I’m sure you realize the asymmetry in the financial resources of our respective companies when you say: “We will both just end up paying a lot of lawyers a lot of money.” . . . My advice is to take a look at our patent portfolio before you make a final decision [on an anti-solicitation agreement] here. 9 United States District Court For the Northern District of California 10 Id., Ex. B. Edward Colligan did not agree and did not communicate with Steve Jobs further 11 regarding his proposal. Id. ¶ 8. 12 Defendants generally structured their agreements with each other to apply to all employees, 13 regardless of job type, department, or geography. See Shaver Decl., Ex. 17 (Google would not 14 recruit anyone from Apple); id., Ex. 56 (Google would not call, network, or email into Intel or its 15 subsidiaries looking for talent); id., Ex. 19 (Apple and Adobe agreed not to solicit any employee 16 from either company); id., Ex. 60 (Lucasfilm and Pixar agreed “not to solicit each other’s 17 employees”); id., Ex. 66 (Pixar agreed not to “directly solicit any Apple employee”); see also 18 Harvey Decl., Ex. 25 at 13 (Google’s Do Not Cold Call policy “include[d] all Intuit employees.”). 19 However, Defendants Apple and Intel agreed simply “NOT to hire top talent (esp technical) away 20 from each other,” rather than to avoid all anti-solicitation efforts. Shaver Decl., Ex. 55 (emphasis 21 in original). 22 Plaintiffs maintain that Defendants’ alleged conspiracy went on for years until revealed by 23 the DOJ. After investigating Defendants’ alleged conspiracy, “the DOJ concluded that Defendants 24 reached ‘facially anticompetitive’ agreements that ‘eliminated a significant form of competition . . . 25 to the detriment of the affected employees who were likely deprived of competitively important 26 information and access to better job opportunities.’” MTD Order at 3-4; CAC ¶ 112; see also 27 Dep’t of Just. Compl. Against Adobe, et al. (“DOJ Adobe Compl.”) ¶¶ 2, 14, ECF No. 93-1; Dep’t 28 of Just. Compl. Against Lucasfilm (“DOJ Lucasfilm Compl.”) ¶¶ 2, 15, 22, ECF No. 93-4. The 30 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 DOJ also determined that the agreements “were not ancillary to any legitimate collaboration,” 2 “were broader than reasonably necessary for the formation or implementation of any collaborative 3 effort,” and “disrupted the normal price-setting mechanisms that apply in the labor setting.” DOJ 4 Adobe Compl. ¶ 16; DOJ Lucasfilm Compl. ¶ 17; CAC ¶ 112. The DOJ concluded that 5 Defendants entered into agreements that were restraints of trade that were per se unlawful under the 6 antitrust laws. DOJ Adobe Compl. ¶ 35; DOJ Lucasfilm Compl. ¶ 3; CAC ¶ 112. However, the 7 government only obtained injunctive relief from the Defendants. It did not obtain any 8 compensation for employees injured by the allegedly collusive activities. 9 This substantial evidence presented by Plaintiffs suggests that adjudication of Defendants’ United States District Court For the Northern District of California 10 alleged antitrust violation will turn on legal and factual issues that are common to the Technical 11 Class. Accordingly, the Court finds that common questions will predominate with respect to the 12 alleged antitrust violation. 13 14 2. Antitrust Impact Having found that common questions will predominate with respect to the first element, 15 antitrust violation, the Court now turns to the second element, impact. “Antitrust ‘impact’—also 16 referred to as antitrust injury—is the ‘fact of damage’ that results from a violation of the antitrust 17 laws.” In re DRAM Antitrust Litig., 2006 WL 1530166 at *7. “It is the causal link between the 18 antitrust violation and the damages sought by plaintiffs.” In re New Motors, 522 F.3d at 19 n.18. 19 Plaintiffs marshal substantial evidence, including documentary evidence and expert reports 20 using statistical modeling, economic theory, and data, to demonstrate that common questions will 21 predominate over individual questions in determining the impact of the antitrust violations. The 22 Court finds that the documentary evidence and expert reports paint a picture of Defendants’ 23 business practices and the market in which Defendants operate that suggests that common proof 24 could be used to demonstrate the impact of Defendants’ actions on Technical Class members. 25 Accordingly, the Court finds that Plaintiffs’ proposed methodology satisfies the predominance 26 standard. 27 28 Specifically, the record suggests that all technical employees—not just those who would have received cold calls but for the anti-solicitation agreements—may have been impacted by the 31 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 agreements. Plaintiffs note that cold calling, a recruitment tool that Defendants viewed favorably, 2 has the effect of spreading information about salaries and benefits from recruiters of one firm to 3 employees of another. Leamer Rep. ¶ 71-76. Such information could then spread to other 4 employees within a firm and beyond, leading to widespread increases in employee compensation 5 across the labor market due to increased access to information. Id. Further, Plaintiffs contend that Defendants had company-wide compensation structures, 7 which organized employees into job groups, levels, and families that were evaluated and paid in 8 relationship to all other groups. Suppl. Mot. at 15-22. In addition, Defendants valued internal 9 equity (the idea that similarly situated employees should be compensated similarly) within their 10 United States District Court For the Northern District of California 6 firms. Id. Because of a desire to maintain equity between employees, the upward pressure that 11 cold calls placed on the salaries of individual employees who would have received the calls would 12 have also affected other employees who were part of the same salary structure. As such, variances 13 in individual employees’ salaries would affect other employees who were in a similar position. 14 Each Defendant’s compensation structure could then have been influenced by the other 15 Defendants’ structures as Defendants saw each other as competitors for the same labor pool. 16 Finally, Plaintiffs point to the fact that Defendants were motivated to retain their 17 employees. This, Plaintiffs contend, would have motivated each Defendant to provide financial 18 incentives to employees to respond to and to prevent poaching by other Defendants. Leamer Rep. 19 ¶ 105. Yet, because of the anti-solicitation agreements, Defendants did not need to initiate such 20 measures, which would have benefitted the entire Technical Class. 21 Ultimately, the Court is not tasked at this phase with determining whether Plaintiffs will 22 prevail on these theories. Rather, the question is narrower: whether Plaintiffs have presented a 23 sufficiently reliable theory to demonstrate that common evidence can be used to demonstrate 24 impact. The Court finds that, based on the extensive documentary evidence, economic theory, 25 data, and expert statistical modeling, Plaintiffs’ methodology demonstrates that common issues are 26 likely to predominate over individual issues. The Court first discusses the substantial documentary 27 evidence, which supports Plaintiffs’ theory of common impact, and then proceeds to discuss the 28 32 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 expert reports. The Court concludes by rejecting Defendants’ attempt to identify flaws that would 2 undermine Plaintiffs’ entire methodology. 3 4 5 a. Documentary Evidence In Plaintiffs’ Supplemental Motion for Class Certification, Plaintiffs submitted thousands of 6 pages of documents—all common evidence—which support Plaintiffs’ theories of classwide harm 7 and undermine many of the representations previously made by Defendants. Plaintiffs’ common 8 evidence suggests that, for the purpose of ultimately proving impact, common issues will 9 predominate over individual ones. United States District Court For the Northern District of California 10 The Court begins by discussing the documentary evidence on the importance of cold calling 11 as a recruitment tool and the effect of the preclusion of cold calling on the Technical Class as a 12 whole. The Court then discusses the evidence of Defendants’ rigid compensation structure and 13 importance of internal equity. The Court finally turns to the documentary evidence that 14 Defendants viewed each other as labor competitors, which may have resulted in individual 15 Defendants’ wage suppression depressing other Defendants’ employees’ wages. 16 i. Cold Calling and Recruitment 17 Plaintiffs produce significant evidence that cold calling was an important part of 18 Defendants’ recruitment practices and contend that the elimination of such recruitment through 19 cold calling had adverse effects on all Technical Class members. 20 21 22 23 24 25 26 27 28 a) The Importance of Cold Calling as a Recruitment Practice to Defendants Plaintiffs’ documents support the allegation that, throughout the class period, Defendants viewed recruitment, particularly of “passive candidates”—that is, employees who were not actively looking for a new job—as crucial to their growth and development. Donna Morris (AdobeAdobe Senior Vice President, Global Human Resources) described recruiting talent as “critical” to company growth. Harvey Decl., Ex 1 (Morris Depo.), at 56:17-19 (“Q: Why is recruiting talent important to Adobe? A: So our critical, most critical asset is people. So really we’re an [intellectual property]-based company.”). Adobe also believed that an important way to source 33 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 “top talent” was to focus on “passive” talent, which it defined as “top performers [who] tend to be 2 entrenched” but “may be ‘willing to listen’ if the right opportunity is presented.” Shaver Decl., Ex. 3 14 at 3. As explained by Donna Morris, “often the very best candidates might not necessarily be 4 looking for Adobe. They might not even know that Adobe is a company where they can leverage 5 their capabilities. And so recruiting is a big, big aspect.” Id. at 57:13-17. To support Google’s 6 rapid growth, which included hiring “several thousand employees per year from 2006 to 2009,” 7 Google employed “as many as 800 recruiters while also working with external recruiting 8 agencies.” See id., Ex. 25, at 7-8. Google also determined that “[p]assive sourcing will play an 9 increasingly larger role in recruiting as we move forward as a company.” Harvey Suppl. Decl., Ex. United States District Court For the Northern District of California 10 11 14 (Google 2006 Sourcing Diagnostic). The record also indicates that Defendants viewed cold calling as a key method to attract 12 potential employees. Intel estimated that, historically, competitive sourcing, including cold calling 13 and research, accounted for 14 “Complete Guide to Sourcing” that “[Cold] Calling candidates is one of the most efficient and 15 effective ways to recruit, ” Shaver Decl., Ex. 54. Similarly, Google found that although referrals 16 were the largest source of hires, “agencies and passively sourced candidates offer[ed] the highest 17 yield.” Harvey Suppl. Decl., Ex. 14. Consequently, in response to concerns over slow hiring, 18 Google’s Chief Culture Officer stated that “[c]old calling into companies to recruit is to be 19 expected unless they’re on our ‘don’t call’ list.” Shaver Decl., Ex. 42. 20 percent of hires, Harvey Decl., Ex. 27, and stated in its Further, Defendants appear to have been particularly concerned about their ability to recruit 21 employees for positions within the Technical Class. For example, shortly prior to Google’s anti- 22 solicitation agreement with Apple, Google determined that it needed to “dramatically increase the 23 engineering hiring rate.” Cisneros Decl., Ex. 1753. Thus, Google stated that it would “need to 24 drain competitors to accomplish this rate of hiring.” Id.; see also Harvey Suppl. Decl., Ex. 14 25 (Google 2006 Sourcing Diagnostic) (documenting a large “hiring gap” for engineering positions). 26 Relatedly, when a Senior Apple Executive stated in 2007 that his biggest challenge in the “EE 27 hiring plan” was finding high quality people, Mark Bentley (former Apple Director of Executive 28 34 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Recruiting and Interim Human Resources Director) responded by stating: “Bottom line is that we 2 need to do more targeted recruiting of ‘passive’ candidates.” Harvey Suppl. Decl., Ex. 7. 3 Google also closely tracked the decline rate of its top technical candidates as well as the 4 loss of its technical employees. See Cisneros Decl., Ex. 173, at 1 (showing that “Google’s offer 5 decline rates 6 Shaver Decl., Ex. 45 (documenting how, during 2010, “ 7 leaving Google for another company went to a startup organization” and that Facebook accounted 8 for the highest portion of overall departures). , especially among senior and top technical candidates.”); see also 9 b) United States District Court For the Northern District of California 10 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Enforcement of the Anti-Solicitation Agreements by Defendants While Defendants dispute that this absence of cold calling due to their anti-solicitation 11 12 of technical employees who reported agreements had any effect on job opportunities or flow of information to the class members, see Opp’n at 17, Defendants’ own documents created during the alleged conspiracy tell a different story. First, Plaintiffs offer evidence indicating that, but for Defendants’ anti-solicitation agreements, Defendants would have been cold calling one another’s employees. For example, in November of 2005, Howard Look (former Pixar Vice President of Software) stated that Pixar was struggling to find candidates, but “of course cannot recruit out of Apple.” Shaver Decl., Ex. 64. Adobe personnel recognized that “Apple would be a great target to look into” for the purpose of recruiting, but knew that they could not do so because, “[u]nfortunately, Bruce [Chizen (former Adobe CEO)] and Apple CEO Steve Jobs have a gentleman’s agreement not to poach each other’s talent.” Shaver Decl., Ex. 13. As Bill Campbell (Chairman of Intuit Board of Directors, Co-Lead Director of Apple, and advisor to Google) explained at his deposition, anti-solicitation agreements prevented a competitor from going “A through Z” and calling “everybody that was a mid-level engineer and above . . . that was what I objected to.” Cisneros Decl., Ex. EE (Campbell Depo.) at 30. Thus, by virtue of these anti-solicitation agreements, Defendants’ employees were deprived of job information and opportunities. 28 35 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Second, the evidence indicates that Defendants actively and aggressively enforced these 2 anti-solicitation agreements, which further demonstrates their harmful effects. In an email from 3 Arnnon Geshuri (Google Recruiting Director) to Eric Schmidt (Google Executive Chairman, 4 Member of the Board of Directors, and former CEO) and Laszlo Bock (Google Senior Vice 5 President of People Operations), Arnnon Geshuri confirmed: “[O]ur [Google] recruiters are strictly 6 following the Do Not Call policy regarding Intel and no one has called, networked, or emailed into 7 the company or its subsidiaries looking for talent.” Shaver Decl., Ex 35. Intel memorialized its 8 agreement with Pixar in a document which states, “We cannot recruit (including calling up, 9 emailing or enticing in any way) current Pixar employees to come to work for Intel. If a Pixar United States District Court For the Northern District of California 10 employee applies to Intel without being recruited by Intel . . . [Pat Geslinger (former Intel Senior 11 Corporate Vice President)] will contact the CEO of Pixar for approval to hire.” Id., Ex. 53. 12 Plaintiffs also offer evidence supporting their assertion that Defendants’ agreements were 13 particularly concerned with preventing the recruitment of one another’s technical employees. 14 Steve Jobs (Co-Founder, Former Chairman, and Former CEO of Apple) repeatedly contacted the 15 CEOs of co-Defendants to thwart the recruitment of Apple’s employees. For instance, when a 16 recruiter from Google’s engineering team contacted an Apple employee in 2007, Steve Jobs 17 forwarded the message to Eric Schmidt (Google Executive Chairman, Member of the Board of 18 Directors, and former CEO) and stated, “I would be very pleased if your recruiting department 19 would stop doing this.” Id., Ex. 24. Google responded by making a “public example” out of the 20 recruiter and “terminat[ing] [the recruiter] within the hour.” Id. The aim of this public spectacle 21 was to “(hopefully) prevent future occurrences.” Id.; see also Cisneros Decl., Ex. 1869 (email 22 from Bill Campbell (Chairman of Intuit Board of Directors, Co-Lead Director of Apple, and 23 advisor to Google) to Sergey Brin (Google Co-Founder), stating “Steve just called me again and is 24 pissed that we are still recruiting his browser guy.”); Shaver Decl., Ex. 25 (email from Steve Jobs 25 to Eric Schmidt) (“I am told that Googles [sic] new cell phone software group is relentlessly 26 recruiting in our iPod group. If this is indeed true, can you put a stop to it?”). 27 28 By “prevent[ing] future occurrences” of Google’s recruitment of Apple employees, see Shaver Decl., Ex. 24, Apple employees were deprived of learning about potential job opportunities 36 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 at Google from more than 800 Google recruiters, as well as Google’s external recruiting agencies. 2 See Harvey Decl., Ex. 25, at 7-8. In fact, Google even declined to hire some former Apple 3 engineers after Steve Jobs let it be known that he’d “strongly prefer that [Google] not hire these 4 guys.” Cisneros Decl., Ex. 653. 5 Plaintiffs offer further evidence supporting their assertion that Defendants’ CEOs personally monitored and enforced their anti-solicitation agreements, especially as the agreements 7 applied to members of the Technical Class. For example, on September 26, 2007, Paul Otellini 8 (CEO of Intel and Member of the Google Board of Directors) received an internal complaint 9 regarding Google’s successful recruiting efforts of Intel’s technical employees. Shaver Decl., Ex. 10 United States District Court For the Northern District of California 6 57 (“Paul, I am losing so many people to Google . . . . We are countering but thought you should 11 know.”). Paul Otellini’s answer was to forward the email to Eric Schmidt (Google Executive 12 Chairman, Member of the Board of Directors, and former CEO): “Eric, can you pls help here???” 13 Id. Eric Schmidt obliged and forwarded the email to his recruiting team, who prepared a report for 14 Eric Schmidt on Google’s activities. Id., Ex. 35. The next day, Eric Schmidt replied to Paul 15 Otellini, “If we find that a recruiter called into Intel, we will terminate the recruiter.” Ex. 50; see 16 also Cisneros Dec’l, Ex. 651 (May 4, 2006, e-mail from Paul Otellini to Eric Schmidt: “Sorry to 17 bother you again on this topic, but my guys are very troubled by Google continuing to recruit our 18 key players.”). 19 Additionally, an email forwarded to Edward Catmull (Pixar President) indicated that Pixar 20 felt compelled to check with Steve Jobs (Co-Founder, Former Chairman, and Former CEO of 21 Apple) before extending a job offer to even an administrative assistant. Shaver Decl., Ex. 68. In 22 response, Edward Catmull emphasized to Rob Cook (Pixar Vice President of Advanced 23 Technology) that “[t]he key is to stay away from the engineers.” Id. Consistent with this position, 24 Edward Catmull informed Steve Jobs via email that Pixar had received an application from an 25 Apple employee to work as a test automation engineer, but “[w]e declined.” Cisneros Decl., Ex. 26 424. Several months later, that employee contacted Pixar again informing Pixar that he had 27 another offer and still planned to leave Apple. Id. Edward Catmull emailed Steve Jobs asking 28 whether Jobs would “object” if Pixar communicated with the employee, and Steve Jobs gave 37 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 permission. Id. Thus, Plaintiffs’ evidence supports their claim that these anti-solicitation 2 agreements, enforced by Defendants’ top officers, stifled recruitment efforts of Technical Class 3 members. 4 c) 5 6 7 8 9 United States District Court For the Northern District of California 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 In addition to showing that Defendants valued cold calling and that the absence of cold calling hampered recruitment of members of the Technical Class, the documentary evidence also suggests that the lack of cold calls had a profound and common effect on all members of the Technical Class. Plaintiffs allege that the elimination of cold calling deprived all employees of information regarding pay packages that the employees could have used to obtain more lucrative employment or to gain leverage over their existing employers in negotiating pay increases. Class Cert. Mot. at 16. The earning potential of a valuable employee who knows her market worth is illustrated by an email exchange at Adobe. Out of concern that one employee—a “star performer” due to his technical skills, intelligence, and collaborative abilities—might leave Adobe because “he could easily get a great job elsewhere if he desired,” Adobe considered how best to retain him. Cisneros Decl., Ex.1250. In so doing, Adobe expressed concern about the fact that this employee had already interviewed with 28 other companies and communicated with friends who worked there. Id. Thus, Adobe noted that the employee “was aware of his value in the market” as well as the fact that the employee’s friends from college were “ Id. In response, Adobe decided to give the employee an immediate pay raise. Id. Similarly, as explained by Alex Lintner (Intuit Head of Global Business Division), “[w]henever somebody’s being targeted by an outside company and we want to retain them, we have a conversation around how we can retain them so they don’t take the offer from the outside company.” Hallock Rep. ¶ 199. In the example of one employee that Intuit wanted to retain, Intuit which was an 26 27 The Effect of the Absence of Cold Calling on the Technical Class as a Whole but, in light of the employee’s skills and contributions to the company, Intuit was “willing to make an investment.” Id. 38 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 While Defendants claim that counter-offers were generally made only to particular 2 individuals to retain key talent, Plaintiffs’ evidence supports their allegation that the risk imposed 3 by cold calls and subsequent counteroffers may have had much broader effects. As noted by one 4 Google employee in response to Google’s decision to make counteroffers to some individuals who 5 were recruited to go elsewhere, “[i]t’s impossible to keep something like this a secret. The people 6 getting counter offers talk, not just to Googlers and ex-Googlers, but also to the competitors where 7 they received their offers (in the hopes of improving them), and those competitors talk too, using it 8 as a tool to recruit more Googlers.” Shaver Decl., Ex. 59. This employee expressed frustration, 9 stating that by staying at Google, “it feels like my loyalty is being punished.” Id. United States District Court For the Northern District of California 10 Alan Eustace (Google Senior Vice President) commented on concerns regarding 11 competition for workers and Google’s approach to counteroffers by noting that, “it sometimes 12 makes sense to make changes in compensation, even if it introduces discontinuities in your current 13 comp, to save your best people, and send a message to the hiring company that we’ll fight for our 14 best people.” Id. Because recruiting “a few really good people” could inspire “many, many others 15 [to] follow,” Alan Eustace concluded, “[y]ou can’t afford to be a rich target for other companies.” 16 Id. According to him, the “long-term . . . right approach is not to deal with these situations as one- 17 off’s but to have a systematic approach to compensation that makes it very difficult for anyone to 18 get a better offer.” Id. (emphasis added). 19 This documentary evidence suggests that Defendants would have responded to the 20 information spread by cold calls not merely on an individual basis, but with a structural response 21 that affected all members of the Technical Class. Thus, Plaintiffs’ evidence suggests not only that 22 the anti-solicitation agreements eliminated a key tool of recruitment, cold calling, but also that the 23 impact of this elimination affected the entire Technical Class. The documentary evidence supports 24 Plaintiffs’ theory that Defendants’ elimination of cold calling allowed Defendants not to have to 25 put in place structural incentives to retain employees. This common evidence provides support for 26 Plaintiffs’ theory that if the anti-solicitation agreements did not exist, Defendants would have had 27 to take actions not only to retain the particular employees who may have received the cold calls, 28 39 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 but also that Defendants would have had to take broader action that would have affected the 2 Technical Class as a whole. 3 ii. Compensation Structure and Internal Equity 4 As set forth below, Plaintiffs’ documentary evidence further shows that Defendants 5 maintained formal compensation structures and made significant efforts to maintain internal equity 6 within those structures. This additional documentary evidence further supports Plaintiffs’ theory 7 that the anti-solicitation agreements’ downward pressure on individual employees’ salaries would 8 have applied similar downward pressure across Defendants’ salary structure and on all Technical 9 Class employees’ salaries. United States District Court For the Northern District of California 10 First, Plaintiffs’ evidence supports their claim that, during the class period, all Defendants 11 used formal administrative compensation structures and divided jobs into pay bands, zones, grades, 12 and ranges by which they evaluated and paid employees in groups in relationship to other groups. 13 At Adobe, every job position was assigned a job title, and every job title had a corresponding 14 salary range within Adobe’s salary structure, which included a salary minimum, middle, and 15 maximum. See Cisneros Decl., Ex. C (Arriada-Keiper Depo.) at 16, 20, 159 189-90, 259; id., Ex. 16 G (Vijungco Depo.) at 29. Adobe expected that the distribution of its existing employees’ salaries 17 would fit 18 shows that, for each year in the Class Period, Apple had a “job structure system,” which included 19 categorizing and compensating its workforce according to a discrete set of company-wide job 20 levels assigned to all salaried employees and 21 to 22 14-15, 52-53; Brown Decl., Ex. 16 (Burmeister Decl.) ¶¶ 6, 10 & Ex. B. Every salary range 23 comprised centrally established “min,” “mid” and “max” amounts. See ids. 24 Id., Ex. F (Streeter Depo.) at 57. Similarly, Apple’s compensation data sets of base salary ranges applicable . Id., Ex. M (Burmeister Depo.) at Google also had many job families, many grades within job families, and many job titles 25 within grades. See, e.g., Harvey Suppl. Decl., Exs. 15, 16; see also Cisneros Decl., Ex. S (Brown 26 Depo.) at 74-76 (discussing salary ranges utilized by Google); id., Ex. X (Wagner Depo.) at 49-50 27 (testifying that Google’s 28 ranges). Throughout the class period, Google utilized salary ranges and pay bands with minima 40 salary ranges had generally the same structure” as the Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION salary 1 and maxima and either means or medians. Hallock Rep. ¶ 66; see Cisenors Decl., Ex. S (Brown 2 Depo.) at 74-76, id., Ex. X (Wagner Depo.) at 49-50 (testifying that Google’s 3 had generally the same structure as the 4 Google Senior Vice President, Business Operations), “if you discussed a specific role [at Google], 5 you could understand that role was at a specific level on a certain job ladder.” Cisneros Decl., Ex. 6 S (Brown Depo.) at 94-95; id., Ex. Y (Conrad Depo.) at 23. 7 salary ranges salary ranges). As explained by Shona Brown (former Likewise, Intel had a “compensation structure,” with job grades and job classifications. See id., Ex. BB (McKell Depo.) at 73 (“[W]e break jobs into one of three categories—job families, we 9 call them—R&D, tech, and nontech, there’s a lot more . . . .”). The company assigned employees 10 United States District Court For the Northern District of California 8 to a grade level based on their skills and experience. Id., Ex. Y (Conrad Depo.) at 23; see also id., 11 Ex. CC (Murray Depo.) at 45 (explaining that everyone at Intel is assigned a “classification” 12 similar to a job grade). Intel standardized its salary ranges throughout the company; each range 13 applied to multiple jobs and most jobs spanned multiple salary grades. Id., Ex. BB (McKell Depo.) 14 at 59. Intel further 15 Adobe, , and compensation at Intel, like 16 17 Id. at 62-63. The other Defendants had similarly set salary structures. Intuit had job families and job 18 titles—as indicated by documents tracking salary low, mid, and high information, job codes, and 19 percentiles—and categorized jobs into 20 Lucasfilm’s compensation scheme included job titles that were matched to job families, see 21 Cisneros Decl., Ex. LL (Coker Depo.) at 246, as well as set salary ranges for employees who had 22 similar job titles or job classifications, id., Ex. NN (Lucas Depo.) at 137-138. Pixar used job 23 families and groups, see id., Ex. VV (Sheehy Depo.) at 78, 136, and established salary ranges for 24 each position, id., Ex. SS (McAdams Depo.) at 29. formal bands. Hallock Rep. ¶¶ 85, 89. 25 Second, to ensure that employees were paid within the prescribed salary ranges, Defendants 26 used specific guidelines and tools. For instance, to assist managers in staying within the prescribed 27 ranges for setting and adjusting salaries, Adobe had an online salary planning tool as well as 28 “salary matrixes” which provided managers with guidelines based on market salary data. See 41 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Cisneros Decl., Ex. C (Arriada-Keiper Depo.) at 82-3 (“[E]ssentially the salary planning tool is 2 populated with employee information for a particular manager, so the employees on their team 3 [sic]. You have the ability to kind of look at their current compensation. It shows them what the 4 range is for the current role that they’re in . . . . The tool also has the ability to provide kind of the 5 guidelines that we recommend in terms of how managers might want to think about spending their 6 allocated budget.”). 7 Apple also created a Human Resources and recruiting tool called which was an 8 internal system for tracking employee records and performance, and required managers to grade 9 employees at pre-set levels. See id., Ex. I (Baja Depo.) at 142-43, 145-46); id., Ex. N United States District Court For the Northern District of California 10 (Fadell Depo.) at 52-53; id., Ex. O (Mansfield Depo.) at 33. As explained by Tony Fadell (former 11 Apple Senior Vice President, Ipod Division, and advisor to Steve Jobs), 12 the employee, this is the level, here are the salary ranges, and through that tool we were then – we 13 understood what the boundaries were.” Id., Ex. N (Fadell Depo.) at 53. “would say, this is 14 Intel also used a software tool to provide guidance to managers about an employee’s pay 15 range which would also take into account market reference ranges and merit. See Harvey Suppl. 16 Decl., Ex. 9. As explained by Randall Goodwin (Intel Technology Development Manager), “[i]f 17 the tool recommended something and we thought we wanted to make a proposed change that was 18 outside its guidelines, we would write some justification.” Cisneros Decl., Ex. Z (Goodwin Depo.) 19 at 52; accord Hallock Rep. ¶ 77. 20 Intuit similarly gave compensation guidelines to managers throughout the company. See 21 Cisneros Decl., Ex. GG (McNeal Depo.) at 76, 99. Michael McNeal (Intuit Vice President of 22 Talent Development, former Vice President of Talent Strategy, Vice President of Talent 23 Acquisition, Director of Talent Acquisition, and Manager of Executive Recruitment) 24 acknowledged, for example, that Intuit provided guidance about “the variables that [the company] 25 usually use[s] to the make . . . decision[s]” about compensation. Id. at 99. 26 Similarly, Frank Wagner (Google Director of Compensation), testified that he could locate 27 the target salary range for jobs at Google through an internal company website. See id., Ex. X 28 (Wagner Depo.) at 57-58 (“Q: And if you wanted to identify what the target salary would be for a 42 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 certain job within a certain grade, could you go online or go to some place . . . and pull up what 2 that was for that job family and that grade? . . . A: Yes.”). 3 Third, Plaintiffs’ evidence indicates that Defendants expected compensation to be set within 4 their salary ranges; deviations required special approval. Adobe’s practice, if employees were 5 below the minimum recommended salary range, was to “adjust them to the minimum as part of the 6 annual review” and “red flag them.” Id., Ex. C (Arriada-Keiper Depo.) at 24. Deviations from the 7 salary ranges would also result in conversations with managers, wherein Adobe’s officers 8 explained, “we have a minimum for a reason because we believe you need to be in this range to be 9 competitive.” Id. United States District Court For the Northern District of California 10 Similarly, Intel regularly ran reports showing the salary range distribution of its employees, 11 id., Ex. BB (McKell Depo.) at 64, and, at Intuit, recruiters could not deviate from salary guidelines 12 without express approval, id., Ex. HH (Nguyen Depo.) at 72-73, 90-92. At Apple, going outside 13 prescribed “guidelines” also required extra approval. See id., Ex. J (Bechtel Depo.) at 217; id., Ex. 14 N (Fadell Depo.) at 53 (“And if we were to go outside of that, then we would have to pull in a 15 bunch of people to then approve anything outside of that range.”). Google’s compensation 16 programs were designed, monitored, and overseen by a special department called “People Ops.” 17 See id., Ex. 5 (Brown Depo.) at 24. And finally, at Pixar, Stephanie Sheehy (Pixar Manager of 18 Human Resources Analysis) and Lori McAdams (Pixar Vice President of Human Resources and 19 Administration) were responsible for ensuring that salaries for each job group remained within 20 their allocated pool. Cisneros Decl., Ex. VV (Sheehy Depo.) at 77-78. 21 Fourth, Plaintiffs’ evidence shows not only that Defendants maintained formal 22 compensation structures, but that the details of these structures were driven by concerns about 23 maintaining internal equity—the idea that employees doing the same work would generally be paid 24 similarly—in both hiring and promotions. 25 As explained by Debbie Streeter (Adobe Vice President, Total Rewards), Adobe “always 26 looked at internal equity as a data point, because if you are going to go hire somebody externally 27 that’s making . . . more than somebody who’s an existing employee that’s a high performer, you 28 need to know that before you bring them in.” Id., Ex. F (Streeter Depo.) at 175. Similarly, when 43 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 considering whether to extend a counteroffer, Adobe advised “internal equity should ALWAYS be 2 considered.” Id., Ex. 216.5. 3 Concerns about internal equity also permeated Apple’s compensation program. Steven Burmeister (Apple Senior Director of Compensation) testified that internal equity—which 5 Burmeister defined as the notion of whether an employee’s compensation is “fair based on the 6 individual’s contribution relative to the other employees in your group, or across your 7 organization”—inheres in some, “if not all,” of the guidelines that managers consider in 8 determining starting salaries. Id., Ex. M (Burmeister Depo.) at 61-64; id., Ex. 1856. In fact, as 9 explained by Patrick Burke (former Apple Technical Recruiter and Staffing Manager), when hiring 10 United States District Court For the Northern District of California 4 a new employee at Apple, “compar[ing] the candidate” to the other people on the team they would 11 join “was the biggest determining factor on what salary we gave.” Id., Ex. L (Burke Depo.) at 279 12 (emphasis added). 13 Likewise, Google considered internal equity to be an important goal. Google utilized a 14 salary algorithm in part for the purpose of “[e]nsur[ing] internal equity by managing salaries within 15 a reasonable range.” Id., Ex. 1613. Furthermore, because Google “strive[d] to achieve fairness in 16 overall salary distribution,” “high performers with low salaries [would] get larger percentage 17 increases than high performers with high salaries.” Id., Ex. 1618.14. 18 Similarly, Intel used internal equity “to determine wage rates for new hires and current 19 employees that corresponded to each job’s relative value to Intel.” Id., Ex. BB (McKell Depo.) at 20 210-211; id., Ex. 398.8. To assist in that process, Intel used a tool that generates an “Internal 21 Equity Report” when making offers to new employees. Id., Ex. BB (McKell Depo.) at 212-13. In 22 the words of Ogden Reid (Intel Director of Compensation and Benefits), “[m]uch of our culture 23 screams egalitarianism . . . . While we play lip service to meritocracy, we really believe more in 24 treating everyone the same within broad bands.” Id., Ex. 2035.4. 25 At Lucasfilm, all new positions and out-of-cycle compensation adjustments presented to its 26 compensation committee for approval were to be accompanied by “Peer Relationship” information 27 regarding how the subject employee’s (or candidate’s) colleagues inside the company were 28 compensated, and this factored heavily into committee decisions. See id., Exs. 710, 729, 2084, 44 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 2092, 2094, 2096; id., Ex. MM (Condiotti Depo.) at 41-42 (“[I]n most cases, when we got a comp 2 request . . . it would have the survey data and all of the internal people that had—were in similar 3 positions.”). Chris Galy (Lucasfilm Director of Talent Acquisition) testified that internal equity is 4 “always one of the considerations” in determining pay for new hires and that he always discusses 5 internal equity with the manager requesting the new hire. See id., Ex. FF (Galy Depo.) at 200-03. 6 Chris Galy explained that maintaining internal equity was important because “[y]ou don’t want to . 7 . . hire one person and lose ten.” Id. at 201. Chris Galy also testified about a specific recent 8 situation in which bringing in a new person at a higher salary required raising another employee’s 9 salary in order to preserve internal equity. Id. at 194-95 (“[A]nd so we did an action [pay increase] United States District Court For the Northern District of California 10 for 11 .”). 10 Pixar similarly expressed concerns with maintaining internal equity. See, e.g., id., Ex. QQ 12 (Batali Depo.) at 67 (“[I]f someone feels like they’re being paid more than someone I know who 13 has more value, it raises a bit of a flag”); id., Ex. UU (Zissimos Depo.) at 71 (discussing comparing 14 salaries of similar employees to ensure they were not “out of whack”). 15 Due to Defendants’ formalized pay structures and compensation design, Plaintiffs’ 16 evidence indicates that Defendants’ concerns with internal equity could lead to classwide changes 17 in compensation levels as a result of the anti-solicitation agreements. In the face of inequities 18 between new hires and existing employees, Defendants considered increasing compensation for 19 existing employees. This precise dynamic is best reflected in Defendants’ own internal documents 20 and emails before and after the anti-solicitation agreements. Lucasfilm regularly and proactively 21 reviewed employee salaries to ensure its workforce was within range and implemented “[C]all- 22 [O]ut [E]quity [A]djustment[s]”—individual compensation increases for the explicit purpose of 23 “align[ing] the employee more appropriately in their salary range . . . [and] based on how that 24 10 25 26 27 28 Chris Galy stated in his deposition: “Where a manager would come in and say ‘I believe that I have an [sic] high-performing’ -- in fact, I just had one of these about a month ago, couple months ago, where we went out and hired somebody, and as we were looking at some of the folks on the team, we recognized that the person is -- we were at risk of potentially having this person feel like they were, you know, not in the market range, so we did an action for . . . . We gave a salary increase.” Cisneros Decl., Ex. FF (Galy Depo.) at 195. Defendants contest this anecdote by contending that the employee in question received a salary increase for purposes unrelated to internal equity. Yet the portions of Chris Galy’s deposition that the Defendants cite are not in the record. The Court relies on the materials that the parties placed in the record. 45 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 employee aligns with their internal peer group based on the same set of criteria.” Id., Ex. OO 2 (Maupin Depo.) at 194; id., Ex. 730. 3 4 At Intel, a human resources document from 2002—prior to the anti-solicitation agreements —recognized “ 5 ” and “ 6 ” Id., Ex. 392. In response, Intel planned to: (1) “ 7 and (2) 8 9 Id. An Intel human resources document confirms that, United States District Court For the Northern District of California 10 11 12 Harvey Suppl. Decl., Ex. 10 at 7 (emphasis added). At Adobe, five months before Bruce Chizen (former Adobe CEO) entered into an 13 agreement with Steve Jobs (Co-Founder, former Chairman, and Former CEO of Apple) to 14 eliminate cold calling between the companies, Donna Morris (Adobe Senior Vice President, Global 15 Human Resources Division) expressed concern “about internal equity due to 16 Harvey Decl., Ex. 17 (“ 17 18 19 20 ”). Adobe personnel stated that, because of the , they may not be able to respond to the problem immediately “ ” Id. 21 The effects of internal equity may have been even more concentrated within the Technical 22 Class. Intuit, for example, differentiated between compensation for engineers and other technical 23 positions versus “all other[s].” Cisneros Decl., Ex. 2739.70. Technical employees had a 24 range of salaries and, even for employees of similar salaries, technical employees 25 Id. Pixar had a “Tools Software Engineer leveling matrix” that it used “to give [Pixar] a consistent 26 framework for evaluating the expected contribution of [their] software engineers” and to justify 27 adjusting salaries. See id., Ex. 1309.1. After Pixar determined that some of its 28 , Pixar decided to make “ .” Id. From Pixar’s 46 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION . 1 perspective, “[t]he goal of the new salary proposals is [to] compensate the lowest paid team- 2 members who are performing at the highest levels. This is a ‘pre-emptive strike.’ We want to send 3 a clear message to these 4 seeing much more competitive offers from other companies.” Id. 5 that we value them at least as much as some new hires who are In sum, Plaintiffs’ evidence supports their theory that Defendants’ formal compensation 6 structures combined with the premium Defendants’ placed on internal equity created a market for 7 the Technical Class of employees in which any individual’s compensation was intertwined with 8 that of her peers. The Court finds persuasive Plaintiffs’ contention that common questions about 9 the impact of Defendants’ compensation structures, their focus on internal equity, and the effects of United States District Court For the Northern District of California 10 these factors on the Technical Class as a whole are likely to predominate over any individual 11 questions. 12 iii. Impact of Labor Market Competition 13 Thus far, the Court has discussed Plaintiffs’ documentary evidence of the effects of cold 14 calling, Defendants’ compensation structures, and Defendants’ internal equity concerns on wage 15 suppression across the Technical Class. Now, the Court turns to documentary evidence that 16 suggests that wage suppression within an individual Defendant firm may have affected Technical 17 Class members employed by other Defendant firms because the Defendants viewed each other as 18 competitors for the same employees. This competition often meant that Defendants benchmarked 19 compensation based on each other or based on common external sources. 20 Adobe, for example, viewed Google and Apple to be among its top competitors for talent 21 and expressed concerned about whether Adobe was “winning the talent war.” Shaver Decl., Ex. 22 14. Adobe further considered itself in a “ 23 Google, Apple, and Intuit as among the other “horses.” See Shaver Decl., Ex. 15. In 2008, Adobe 24 benchmarked its compensation against 25 Decl., Ex. 15; cf. Cisneros Decl., Ex. 2800 (showing that, in 2010, Adobe considered Intuit to be a 26 “direct peer,” and considered Apple, Google, and Intel to be “reference peers,” though Adobe did 27 not actually benchmark compensation against these latter companies). -horse race from a benefits standpoint,” which included companies including Google, Apple, and Intel. Shaver 28 47 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Similarly, throughout the class period, Google analyzed and compared its equity 2 compensation to Apple, Intel, Adobe, and Intuit, among other companies, each of which it 3 designated as a “peer company” based on meeting criteria such as being a “high-tech company,” a 4 “high-growth company,” and a “key labor market competitor.” Id., Ex. 173. In 2007, based in part 5 on an analysis of Google as compared to its peer companies, Laszlo Bock (Google Senior Vice 6 President of People Operations) and Dave Rolefson (Google Equity Compensation Manager) wrote 7 that “[o]ur biggest labor market competitors are 8 beat Google for talent.” Id. 9 to Apple identified Google, Intel, as well as The Walt Disney Company (which now owns United States District Court For the Northern District of California 10 Pixar and Lucasfilm), as “peer companies,” a phrase Apple defined as “U.S.-based, stand-alone, 11 public companies that, in [the Apple compensation committee’s] view, compete with [Apple] for 12 talent, have revenue, market capitalization, and performance that are generally comparable to 13 [Apple].” Cisneros Decl., Ex. 1855. 14 Intel also benchmarked compensation against other “tech companies generally considered 15 comparable to Intel,” which Intel defined as a “[b]lend of semi, software, networking, 16 communications, and diversified computer companies.” Cisneros Decl., Ex. 2030.115. According 17 to Intel, in 2007, these comparable companies included Apple and Google. Id. 18 At Lucasfilm, a 2007 “Recruiting and Human Resources Update” prepared for a Board of 19 Directors meeting corroborates many of the dynamics that appear to have been at play for 20 Defendants. The update states that (1) passive talent was “difficult to find;” (2) Lucasfilm had 21 “[e]xtremely diverse needs for each division;” (3) its Bay Area competition included other 22 Defendants in this case including Pixar, Google, and “Silicon Valley” generally; and (4) some of 23 its most difficult positions to fill included members of the Technical Class, including 24 Id., Ex. 690.19-.22. 25 Further, there is evidence that Defendants also benchmarked their compensation data to 26 common external sources, most commonly Radford or Croner. Adobe, for example, pegged its 27 compensation structure as a 28 . See id., Ex. C (Arriada-Keiper Depo.) at 16; see also id., Ex. M 48 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 (Burmeister Depo.) at 52-55 (discussing how Apple used external market data, such as Radford, as 2 a “reference” when determining salary ranges); id., Ex. 1309.1 (showing that Pixar compared itself 3 against the Radford survey). Bruce Chizen (former Adobe CEO) explained that the particular 4 market targets that Adobe used as benchmarks for setting salary ranges “tended to be software, 5 high-tech, those that were geographically similar to wherever the position existed.” Id., Ex. A 6 (Chizen Depo.) at 98. At times, Defendants would make adjustments to their salaries in order to 7 stay competitive. For example, Intel would “ 8 9 ,” id., Ex. BB at 89, and use that as part of the process for determining its “own focal process or pay delivery,” id., Ex. CC at 23. United States District Court For the Northern District of California 10 11 Id., Ex. BB at 90. Plaintiffs’ documentary evidence further suggests that when determining the 12 competitiveness of their own pay practices, Defendants matched job title compensation within the 13 company to similar titles across multiple companies and shared compensation information. See 14 Shaver Suppl. Decl., Ex. 122 (email from Lori McAdams (Pixar Vice President of Human 15 Resources and Administration) to Sharon Coker (former Lucasfilm Director and Senior Director of 16 Human Resources), among others, asking about others’ “salary increase budget for FY ’07” and 17 stating “[o]urs is 18 close, more or less?”); Cisneros Decl., Ex. 621 (email from Laszlo Bock (Google Senior Vice 19 President of People Operations) stating, “[o]ur budget is comparable to other tech companies . . . . 20 We called tech companies this week to check merit budgets to compare to our 3.9%. They told us 21 their merit budgets are: . . . Adobe – 22 could safely share this data only because they were not in fact competing for employees due to the 23 anti-solicitation agreements. 24 , but we may manage it closer to . . . Apple – on average. Are you doing anything . . . Intel – . . .). Defendants The pressure Defendants would have experienced due to market competition, including 25 competition against other Defendants in this case, is reflected in a 2006 email from Howard Look 26 (former Pixar Vice President of Software) to Lori McAdams (Pixar Vice President of Human 27 Resources and Administration), Ed Catmull (Pixar President), and Ali Rowghani (Pixar CFO and 28 Senior Vice President of Strategic Planning): 49 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 2 3 4 5 6 7 This new market data [from Radford] corroborates what we’ve been feeling about the bay area tech market heating up, and helps partially explain why recruiting has been so hard, and why we’ve lost two people to other tech companies ( ) . . . . We obviously want to be fiscally prudent and do our best to stay within our budget, but we should also acknowledge the much higher cost of backfilling for experienced engineers. We have lots going for us beyond base salary . . . but for a software apps engineer, there’s a lot going on out there, and with , there is risk that we may lose more. Cisneros Decl., Ex. 1306. The concern Defendants felt about competition for employees contributed to Defendants’ 8 decisions to enter into anti-solicitation agreements. For example, in 2005, after hearing that 9 Google was trying to recruit employees from Apple’s Safari team, Steve Jobs (Co-Founder, Former United States District Court For the Northern District of California 10 Chairman, Former CEO of Apple) threatened Sergey Brin (Google Co-Founder), stating, as Brin 11 recounted, “if you [Brin] hire a single one of these people that means war.” Id., Ex. 1871. In an 12 email to Google’s Executive Management Team as well as Bill Campbell (Chairman of Intuit 13 Board of Directors, Co-Lead Director of Apple, and advisor to Google), Sergey Brin advised: “lets 14 [sic] not make any new offers or contact new people at Apple until we have had a chance to 15 discuss.” Id. Similarly, in 2005, when considering whether to enter into an anti-solicitation 16 agreement with Apple, Bruce Chizen (former Adobe CEO), expressed concerns about the loss of 17 “top talent” if Adobe did not enter into an anti-solicitation agreement with Apple, stating, “if I tell 18 Steve [Jobs] it’s open season (other than senior managers), he will deliberately poach Adobe just to 19 prove a point. Knowing Steve, he will go after some of our top Mac talent like 20 will do it in a way in which they will be enticed to come (extraordinary packages and Steve 21 wooing).” Harvey Decl., Ex. 14. 22 and he In addition, Defendants appear to have perceived the anti-solicitation agreements as a way 23 to stifle rising costs. From the perspective of George Lucas (former Lucasfilm Chairman of the 24 Board and CEO), Lucasfilm “c[ould]not get into a bidding war with other companies because we 25 don’t have the margins for that sort of thing.” Cisneros Decl., Ex. NN at 44; see also Shaver Decl., 26 Ex. 60 (stating in an email that Pixar and Lucasfilm “have agreed that we want to avoid bidding 27 wars”). As expressed by Edward Catmull (Pixar President), “[e]very time a studio tries to grow 28 rapidly . . . it seriously messes up the pay structure . . . by offering high salaries to grow at the rate 50 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION [a company] desire[s], people will hear about it and leave.” Id., Ex. 61; see also Cisneros Decl., 2 Ex. RR (Catmull Depo.) at 179 (“So it messes up the pay structure. It does. It makes it very high. . 3 . . That’s just the reality we’ve got. And I do feel strongly about it.”). Writing in response to the 4 head of Disney Studios, Edward Catmull explained, “[w]e have avoided wars up here in 5 Norther[n] California because all of the companies up here - Pixar, ILM [Lucasfilm], Dreamworks, 6 and a couple smaller places - have conscientiously avoided raiding each other.” Shaver Decl., Ex. 7 61. Notably, shortly after Steve Jobs entered into an anti-solicitation agreement with Google, Meg 8 Whitman (former CEO of eBay) called Eric Schmidt (Google Executive Chairman, Member of the 9 Board of Directors, and former CEO) “to talk about [Google’s] hiring practices.” Cisneros Decl., 10 United States District Court For the Northern District of California 1 Ex. 872. As Eric Schmidt told Google’s senior executives, Ms. Whitman said “Google is the talk 11 of the valley because [you] are driving up salaries across the board.” Id. 12 The evidence therefore indicates that Defendants sought to enter into anti-solicitation 13 agreements in an effort to stifle increased competition for labor and rising wages. To the extent 14 that they were successful, Defendants did not need to increase compensation as much as they 15 otherwise would have to attract and retain employees. This common evidence further suggests that 16 the anti-solicitation agreements reached beyond individual members of the Technical Class and 17 affected the compensation of the Technical Class as a whole, including across Defendant firms. 18 The extensive documentary evidence Plaintiffs present therefore supports their theory that they will 19 be able to prove the impact of the antitrust violations on a classwide basis. 20 b. 21 Expert Reports and Statistical Evidence To show that common issues predominate for the purpose of assessing classwide impact, 22 Plaintiffs further retained the services of two experts: Edward E. Leamer, Ph.D., 11 and Kevin F. 23 Hallock, Ph.D. 12 Defendants presented reports from their own experts, Kevin M. Murphy, Ph.D., 13 24 11 25 26 27 28 Edward E. Leamer, Ph.D, is the Chauncey J. Medberry Professor of Management, Professor of Economics, and Professor of Statistics at the University of California, Los Angeles. Dr. Leamer earned a B.A. in Mathematics from Princeton University in 1966, and a Masters in Mathematics and a Ph.D. in Economics at the University of Michigan in 1970. He has published on the topics of econometric methodology and statistical analysis, international economics, and macro-economic forecasting, including on the subject of inferences that may appropriately be drawn from nonexperimental data. 12 Kevin F. Hallock, Ph.D., is the Donald C. Opatrny ’74 Chair of the Department of Economics, Joseph R. Rich ’80 Professor, Professor of Economics, Professor of Human Resources Studies, and 51 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 and Kathryn Shaw, Ph.D., 14 to attack Dr. Leamer’s and Dr. Hallock’s analyses and conclusions. 2 The Court begins by describing the methodologies and analyses of Dr. Leamer and Dr. Hallock, 3 which support Plaintiffs’ theories of common impact of harm. The Court then turns to Defendants’ 4 criticisms of Dr. Leamer, Dr. Hallock, and the materials on which Dr. Leamer and Dr. Hallock 5 rely. The Court finds that methodological deficiencies in Defendants’ expert reports render the 6 criticisms unpersuasive. The Court therefore finds that the methodologies and theories of Dr. 7 Leamer and Dr. Hallock demonstrate that common questions are likely to predominate over 8 individual questions. 9 United States District Court For the Northern District of California 10 11 12 13 14 15 16 17 18 i. Dr. Leamer’s Opinions Based on Economic Theory, Documentary Evidence, Data, and Statistical Analyses In Dr. Leamer’s first expert report, which was presented in support of Plaintiff’s Motion for Class Certification, Plaintiffs asked Dr. Leamer to evaluate whether classwide evidence was capable of showing that the anti-solicitation agreements artificially reduced the compensation of: (1) members of the Technical and All Employee classes generally, and (2) all or most members of each class. See Leamer Rep. ¶ 10(a). In addition, Plaintiffs asked Dr. Leamer to assess whether there was a reliable classwide or formulaic method capable of quantifying the amount of suppressed compensation suffered by each class member. Leamer Rep. ¶ 10(b). Dr. Leamer answered these questions in the affirmative. 19 20 21 22 23 24 25 26 27 28 Director of the Cornell Institute for Compensation Studies at Cornell University. Dr. Hallock earned a B.A. in Economics at the University of Massachusetts at Amherst in 1991, and a Ph.D. in Economics from Princeton University in 1995. He is a leading labor economist and an expert in compensation structure and design. 13 Kevin M. Murphy, Ph.D., is the George J. Stigler Distinguished Service Professor of Economics in the Booth School of Business and the Department of Economics at the University of Chicago. Murphy received a bachelor’s degree in economics from the University of California, Los Angeles, in 1981, and a Ph.D. in economics from the University of Chicago in 1986. Dr. Murphy has published on labor markets and the determinants of wages and compensation. His work in labor economics has addressed the market determinants of wage by skill level as well as the determination of relative wages across industries and occupations. 14 Kathryn Shaw, Ph.D., is the Ernst C. Arbuckle Professor of Economics at the Stanford Graduate School of Business. Dr. Shaw received an A.B. degree from Occidental College and Ph.D. in Economics from Harvard University. Dr. Shaw has published on the topic of personnel economics. She also co-pioneered the field of “insider econometrics,” a research field in personnel economics in which researchers go within companies and use insider knowledge and data to identify the performance gains from management practices. 52 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Dr. Leamer’s analysis proceeded in two steps. First, Dr. Leamer explained that economic 2 studies and theory, documentary evidence, and statistical analyses were capable of showing that the 3 anti-solicitation agreements “tend[ed] to suppress employee compensation generally, by preventing 4 class members from discovering the true value of their work.” Class Cert. Mot. at 16. In other 5 words, Dr. Leamer illustrated how classwide evidence was capable of showing that, at the very 6 least, Defendants were paying some members of the class less than they would have been paid in 7 the absence of the anti-solicitation agreements. Second, Dr. Leamer illustrated how economic 8 studies and theory, documentary evidence, and statistical analyses are capable of showing that this 9 suppression of compensation affected all or nearly all class members. Plaintiffs noted that Dr. United States District Court For the Northern District of California 10 Leamer’s approach followed a roadmap widely accepted in antitrust class actions that use evidence 11 of general price effects plus evidence of a price structure to conclude that common evidence is 12 capable of showing widespread harm to the class. See, e.g., Johnson, 2009 WL 5031334 at *8, 11 13 (finding predominance where conduct was alleged to suppress bill rates for nurses generally and 14 evidence was presented that bill rates were correlated with nurse pay rates); see also In re 15 Linerboard Antitrust Litig., 305 F.3d 145, 153-55 (3d Cir. 2002) (endorsing regression plus pricing 16 structure study to show classwide impact). 17 In Dr. Leamer’s supplemental expert report, which was prepared in support of the 18 supplemental motion for class certification, Plaintiffs asked Dr. Leamer to respond to questions 19 raised by the Court related to whether Dr. Leamer’s initial methodology could show classwide 20 impact. Dr. Leamer focused his supplemental report on the Technical Class and found that his 21 additional analyses confirmed his “original finding of a somewhat rigid pay structure at each 22 Defendant that would have transmitted the effects of the agreements broadly, including throughout 23 the Technical Class.” Leamer Suppl. Rep. ¶ 13. 24 25 a) Suppressed Compensation Generally Dr. Leamer first concluded that classwide evidence was capable of showing that the anti- 26 solicitation agreements suppressed compensation of Technical Class members generally. 27 According to Dr. Leamer, this first step was supported by principles of information economics, 28 such as “market price discovery.” Dr. Leamer noted that, when evaluating the functioning of labor 53 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 markets, economists often use a market equilibrium model, which “presume[s] that market forces 2 are powerful enough and work rapidly enough that virtually all transactions occur at approximately 3 the same price—the ‘market price’ which equilibrates supply and demand.” Leamer Rep. ¶ 71. In 4 reality, when labor market conditions change, high transaction costs and limited information flow 5 can slow the process by which transaction prices reach market equilibrium. Id. ¶¶ 72-73. “Market 6 price discovery” is the process by which participants in a market search for this equilibrium. Id. 7 ¶ 71. 8 Dr. Leamer opined that the high transaction costs—including time, money, and personal dislocation—involved in searching for high tech jobs limit the number of existing workers seeking 10 United States District Court For the Northern District of California 9 new employment. Id. ¶ 74. Defendants and other high tech companies value potential employees 11 who are not actively looking for new employment opportunities (“passive candidates”) more than 12 those who are looking for new jobs (“active candidates”) because currently satisfied employees: (1) 13 tend to be perceived as more qualified, diligent, and reliable; (2) often have training, on-the-job 14 experience, and track records that save the hiring company search and training costs; and (3) are 15 valuable assets that, if hired away from rivals, can harm competitors. Id. ¶ 62. Thus, recruiting 16 these passive candidates by cold calling is both an important tool for employers and a key channel 17 of information for employees about outside opportunities. Id. ¶¶ 57-62, 75. 18 Dr. Leamer hypothesized that, by restricting cold calling and other competition over 19 employees, Defendants’ anti-solicitation agreements impaired information flow about 20 compensation and job offers. Class Cert. Mot. at 3. Defendants’ inhibition of employees’ ability 21 to discover and obtain the competitive value of their services meant employees were afforded 22 fewer opportunities to increase their salaries by moving between firms and deprived of information 23 that could have been used to negotiate higher wages and benefits within a firm. See Leamer Rep. 24 ¶¶ 71-76. In addition, Dr. Leamer opined that, by limiting the information available to employees, 25 Defendants could avoid taking affirmative steps, such as offering their employees financial rewards 26 and other forms of profit sharing, to retain employees with valuable firm-specific skills. Id. ¶¶ 77- 27 80. 28 54 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 In support of this hypothesis, Dr. Leamer cited to the work of Nobel Prize winning 2 economists, such as Joseph Stiglitz, for the proposition that “even a small amount of information 3 imperfection could have a profound effect on the nature of the equilibrium.” Joseph Stiglitz, 4 Information and the Change in the Paradigm in Economics, 92 Am. Econ. Rev. 460, 461 (2002); 5 see Leamer Reply Rep. ¶¶ 37, 38; see also id. ¶ 40 (“The fact that actions convey information leads 6 people to alter their behavior, and changes how markets function. This is why information 7 imperfections have such profound effects.”). 8 Dr. Leamer also relied on the documentary evidence—common to the class as a whole—as further support for the link between the anti-solicitation agreements and compensation reduction. 10 United States District Court For the Northern District of California 9 See Leamer Rep. ¶¶ 81-88. For example, he cited to Defendants’ internal documents indicating 11 that, but for the anti-solicitation agreements, Defendants would have been competing for labor and 12 cold calling each other’s employees, and that Defendants recognized that cold calling and other 13 forms of employee solicitation had the potential to drive up the cost of specific employees, which 14 could then have broader effects. Id. 15 Finally, Dr. Leamer bolstered his findings with standard econometric analysis utilizing 16 solely classwide evidence and methods. Dr. Leamer performed an analysis to show that employees 17 who changed firms received higher compensation than those who stayed, reflecting the economic 18 theory of price discovery at work. Id. ¶¶ 89-93. Dr. Leamer also performed multiple regression 19 analyses, 15 utilizing Defendants’ internal compensation data, to illustrate class members’ 20 undercompensation by comparing compensation during the conspiracy with compensation in a 21 conspiracy-free, but-for world. Dr. Leamer concluded that the multiple regression analyses 22 showed that the anti-solicitation agreements artificially suppressed compensation at each 23 Defendant. Leamer Rep. ¶¶ 145-46, Figs. 20-24. 24 b) Widespread Effect 25 15 26 27 28 “A regression is a statistical tool designed to express the relationship between one variable, such as price, and explanatory variables that may affect the first variable. Regression analysis can be used to isolate the effect of an alleged conspiracy on price, taking into consideration other factors that might also influence price, like costs and demand.” In re Aftermarket Auto. Lighting Prods. Antitrust Litig., 276 F.R.D. 364, 371 (C.D. Cal. 2011) (internal quotation marks and citation omitted). 55 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Second, Dr. Leamer opined that economic studies and theory, documentary evidence, and 2 statistical analyses were capable of showing that this compensation suppression had widespread 3 effects. In both his initial report and his supplemental report, Dr. Leamer relied on economic 4 studies and theories of loyalty, fairness, and internal equity, as well as documentary evidence and 5 data, to explain how the adverse effects on compensation due to Defendants’ anti-solicitation 6 agreements would have been felt by employees who would have received a cold call or had a 7 significant chance of receiving a cold call and employees who are linked to these groups due to 8 internal equity considerations. Leamer Rebuttal Suppl. Expert Decl. (“Leamer Suppl. Reply Rep.”) 9 ¶¶ 27-28. United States District Court For the Northern District of California 10 Based on economic studies and theories involving loyalty, fairness, and internal equity, Dr. 11 Leamer contended that labor markets do not behave like commodity markets. Rather, labor 12 markets rely on committed long-term relationships built on trust, understanding, and mutual 13 interests. Leamer Rep. ¶ 102. As Dr. Leamer explained: “If workers were commodities, every 14 small change to external or internal conditions would lead to recontracting, separation, or 15 termination. This would create enormous uncertainty and disruption and insecurity for employer 16 and employee.” Id. Thus, both employers and employees seek ways to turn the market transaction 17 into secure long-term relationships, which “can come either from commitment (emotional or 18 financial) to the mission of the organization, or from jointly owned firm-specific assets.” Id. 19 Companies thus attempt to create loyalty “by getting buy-in from the firm’s mission and by making 20 the place of work as appealing as possible.” Id. ¶ 103. 21 “One foundation of employee loyalty is a feeling of fairness that can translate into a sharing 22 of . . . [a firm’s] rewards with more equality than a market might otherwise produce.” Id. ¶ 104. 23 Firms seek to promote a feeling of fairness among employees to maintain or to increase 24 employees’ commitment and contentment, which also leads to higher levels of productivity. 25 Leamer Suppl. Rep. ¶ 16. Dr. Leamer explained that, “[t]o maintain loyalty, it is usually better for 26 a firm to anticipate rather than to react to outside opportunities, since if a worker were to move to 27 another firm at a much higher level of compensation, coworkers left behind might feel they have 28 56 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 not been fairly compensated. That can have an adverse effect on worker loyalty, reducing 2 productivity and increasing interest in employment elsewhere.” Leamer Rep. ¶ 105. 3 Dr. Leamer opined that the information conveyed by an outside offer or a cold call could 4 stimulate a response by management that could extend beyond the specific individual who received 5 the cold call. As Dr. Leamer explained, “when management becomes aware of an attractive 6 outside opportunity for one individual this may make management aware also of the implicit 7 competitive threat to similar individuals and management may feel it wise to make a preemptive 8 move against that threat by an increase in compensation for these newly-threatened similar 9 employees.” Leamer Suppl. Rep. ¶ 15. Even though the market may not mandate a rise in United States District Court For the Northern District of California 10 compensation for these similar individuals until they actually receive an outside offer, “preemptive 11 improvements” can minimize the disruption to employee loyalty that might occur when an 12 employee discovers the she was undercompensated. Leamer Rep. ¶ 105. Thus, “[c]old-[c]alling— 13 as well as just the threat of [c]old-[c]alling—puts upward pressure on compensation.” Id. ¶ 106. 14 Dr. Leamer opined that “a broad preemptive response is completely analogous to salary increases 15 that are tied to information provided by employment services regarding the compensation offered 16 by the ‘market.’” Leamer Suppl. Rep. ¶ 15. Essentially, Dr. Leamer opined that the “response to 17 bursts of cold calls and, even more, the response to the threat of cold calls” would raise internal 18 equity concerns that would spread the impact throughout the Technical Class. Leamer Suppl. 19 Reply Rep. ¶ 27. 20 In further support of his opinion, Dr. Leamer relied on documentary evidence, including 21 Defendants’ compensation data. This documentary evidence showed Dr. Leamer that Defendants 22 each employed company-wide compensation structures that included grades and titles, and that 23 high-level management established ranges of salaries for grades and titles, which left little scope 24 for individual variation. Id. ¶¶ 121-22. Defendants also established and regularly updated 25 compensation levels with the goals of: (1) providing similar compensation for all employees in the 26 same employment category; (2) providing specific relative compensation levels for employees in 27 different, hierarchically ordered, employment categories; (3) retaining employees; and (4) 28 maintaining employee productivity and contentment. See id. ¶ 122. 57 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Dr. Leamer looked to standard economic labor theory and statistical analyses as additional 2 evidence that the anti-solicitation agreements would broadly affect members of the Technical 3 Class. He explained that his “statistical task is to identify the common factors in the individual 4 data and to apportion these common factors between internal and external forces.” Leamer Suppl. 5 Reply Rep. ¶ 29. 6 In his first expert report, Dr. Leamer conducted regression analyses based on Defendants’ 7 salary structures and compensation data, to which the Court referred as Dr. Leamer’s “Common 8 Factors Analyses.” See Leamer Rep., Figs. 11-14. Specifically, Dr. Leamer’s Common Factors 9 Analyses assessed Defendants’ “firmwide compensation structures, and the formulaic way in United States District Court For the Northern District of California 10 which total compensation was varied over time.” Id. ¶ 128. According to Dr. Leamer, 11 approximately 90 percent of the variation in any individual employee’s compensation can be 12 explained by common factors “such as age, number of months in the company, gender, location, 13 title, and employer.” Id.; see also id., Figs. 11-14. Defendants do not dispute the “fact that job 14 titles explain a large fraction of the firm-wide variation in compensation.” Murphy Expert Report 15 (“Murphy Rep.”) ¶ 92, ECF No. 230. Dr. Leamer concludes that “[t]he fact that nearly all 16 variability in class member compensation at any point in time can be explained by common 17 variables means there was a systematic structure to employee compensation at each of the 18 Defendant firms.” Leamer Rep. ¶ 130. Dr. Leamer opined that these rigid wage structures, and the 19 fact that the coefficients in his regressions did not vary substantially over time, suggested that 20 “compensation of class members tended to move together over time and in response to common 21 factors,” such that the effects of the anti-solicitation agreements would be expected to be 22 experienced broadly. Id. 23 Second, Dr. Leamer opined that the evidence showed “a persistent salary structure across 24 employees consistent with important elements of equity in the Defendants’ compensation 25 practices.” Id. ¶ 134. Dr. Leamer specifically relied on five charts that depicted changes in the 26 base salaries and total compensation for ten major job titles at Apple between 2006 and 2009, and 27 the ten major job titles at Google between 2005 and 2009. See id., Figs. 15-17. The Court referred 28 to these five charts as Dr. Leamer’s “Compensation Movement Charts.” Apr. 5 Class Cert. Order 58 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 at 36. Dr. Leamer contended that these charts offered further evidence that compensation for 2 different positions tended to move together over time (i.e., if software engineers received a raise, so 3 did account executives). See Leamer Rep. ¶¶ 133-34. Based on this evidence, Dr. Leamer opined 4 that the anti-solicitation agreements that focused on subsets of workers would nonetheless have 5 broader effects because of a desire on Defendants’ part to maintain the overall salary structure. Id. 6 ¶ 134. 7 Third, Dr. Leamer used a regression model to show that the anti-solicitation agreements had 8 some general impact on the Class and to quantify the total amount of that impact on the Class. See 9 Leamer Rep., Figs. 20-24. This model, to which the Court previously referred as the “Conduct United States District Court For the Northern District of California 10 Regression” analysis, incorporated a range of variables designed to account for factors including: 11 (1) age, sex, and years at the company; (2) the effects on compensation caused by the anti- 12 solicitation agreements; and (3) the effects caused by factors specific to each Defendant (e.g., firm 13 revenue, total number of new hires, etc.). See id., Figs. 20, 23. Dr. Leamer used the model to 14 estimate the average or net under-compensation at each firm during the conspiracy period. See id., 15 Fig. 22 and 24; Reply at 33. 16 In the Court’s April 5 Class Certification Order, the Court stated that, “[a]ccepting 17 arguendo that the Common Factors Analyses are accurate, they show that factors such as where an 18 employee works and what an employee does play a large role in determining the employee’s 19 salary.” Apr. 5 Class Cert. Order at 36. However, the Court found that Dr. Leamer did not 20 establish that this fact implied that Defendants’ salary structures were so rigid that compensation 21 for employees with different titles would move together through time such that a detrimental 22 impact to an employee with one job title would result in an impact to other employees in entirely 23 different jobs (i.e., that any impact would ripple across the entire salary structure). Id. 24 The Court also found that Dr. Leamer’s Compensation Movement Charts shed little light on 25 whether compensation of members of the All Employee Class (e.g., a custodian at an Intel office in 26 Texas and an engineer at an Intel office in California) moved together over time because the 27 Compensation Movement Chart included only twenty job titles, primarily job titles from the 28 Technical Class at two companies out of the thousands of job titles at the seven companies 59 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 included in the classes. Id. at 36-38. The Court expressed concern that the twenty positions 2 reflected on the chart were not representative of the compensation movement of all Class members. 3 See id. at 36-38. The Court also found that these charts did not provide particularly compelling 4 evidence regarding whether salaries at each company were linked because Dr. Leamer admitted 5 that the allegedly parallel movement reflected in the charts was also consistent with a “non[-]rigid 6 wage structure.” Brown Decl., Ex. 1 (Leamer Depo.) at 283:23-25. 7 Finally, the Court found that Dr. Leamer’s Conduct Regression analysis was capable of 8 showing that Defendants’ total expenditures on compensation was less than they would have been 9 in the absence of anti-solicitation agreements and thus capable of showing classwide damages. United States District Court For the Northern District of California 10 Apr. 5 Class Cert. Order at 38. The Court further found that the Conduct Regression was capable 11 of showing that the anti-solicitation agreements had a general impact on class members. Id. The 12 Court rejected Defendants’ contentions to the contrary. Id. at 39-42. In this Order, the Court 13 discusses Conduct Regression primarily for the analysis’s utility in demonstrating classwide 14 damages. However, the Court also notes, as it did in its previous order, that the Conduct 15 Regression analysis is also capable of demonstrating a general classwide impact. 16 In response to the Court’s concerns that the extant statistical analyses could not show a 17 rigid wage structure, Dr. Leamer submitted additional statistical analyses in his supplemental 18 expert report. These new analyses focused on demonstrating that Defendants maintained a 19 somewhat rigid wage structure not only within job titles, but also that Defendants maintained such 20 a rigid wage structure across job titles. To demonstrate this, Dr. Leamer performed a correlation 21 analysis on a job-title-by-job-title basis that compared the “movement over time of the average 22 compensation of each title with the average compensation of the firm’s Technical Class.” Leamer 23 Suppl. Rep. ¶ 4 (emphasis omitted). Dr. Leamer looked to the correlation analysis to measure 24 “statistically how closely different variables move together.” Id. ¶ 22; see also Suppl. Reply at 4 25 (stating that this correlation analysis is the quantitative equivalent of the co-movement charts). Dr. 26 Leamer conducted the correlation analysis for all job titles, not just the twenty that Dr. Leamer 27 presented in his initial report. Leamer Suppl. Rep. ¶ 4. This included all titles for which Dr. 28 Leamer had at least six observations (a statistical threshold), which included 94% of Class Period 60 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 employee years. Id. ¶¶ 14, 30. He analyzed correlation over time in two dimensions: “correlation 2 of compensation levels and correlations of compensation changes.” Id. ¶ 23 (emphasis in original). 3 The former focuses on long-term movements, while the latter focuses on year-by-year movements. 4 Id. With respect to both, Dr. Leamer found that the “vast majority” of Technical Class 6 employee job titles (weighted by number of employee years) at each firm correlated positively over 7 time with the compensation of the overall set of Technical Class employees at that firm. See id. ¶ 8 32, Figs. 2, 3. 16 As such, an increase in the compensation of each job title within any given firm 9 was correlated with an increase in the overall compensation of Technical Class employees in that 10 United States District Court For the Northern District of California 5 firm. In combination with the fact that 90% of employee total compensation is driven by common 11 factors, Dr. Leamer contended that this sharing of gains across job titles over time further 12 supported an inference of a somewhat rigid salary structure. See id. ¶ 4, Figs. 2, 3. Accordingly, 13 Dr. Leamer’s supplemental analysis bolstered his finding of a rigid salary structure because it 14 demonstrated that not only was there a rigid salary structure within job titles (which his initial 15 analysis showed) but also that there was a rigid salary structure across job titles. 16 Dr. Leamer next presented a multiple regression model for each company designed to 17 detect the effect of internal forces acting on class member compensation (i.e., a rigid compensation 18 structure), as opposed to external market forces. See id. ¶¶ 24-29. The model measured the effect 19 of a number of explanatory variables on job title compensation. One variable was average 20 Technical Class compensation at a particular company. See id. The effect of this variable reflects 21 the degree to which compensation increases for the group are shared broadly at the same time. Id. 22 ¶ 25. The next variable measured the effect of the previous year’s compensation, showing the 23 degree to which gains in one year are later shared with other members of the Technical Class at the 24 same company. See id. ¶ 26. Dr. Leamer’s model also included variables for the firm’s revenue 25 and job growth in the San Jose-Sunnyvale-Santa Clara Metropolitan Statistical Area to allow for 26 the possibility of alternative explanations for compensation increases. Id. ¶¶ 27-28. 27 16 28 To account for titles with insufficient data to run the title-by-title analysis, Dr. Leamer also divides the employee groups into deciles and measures the correlation of each decile to the mean; these groups exhibit the same positive relationship. See Leamer Suppl. Rep. Figs. 9, 10. 61 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Dr. Leamer estimated the regression on a title-by-title basis for job titles with adequate data 2 within each company. Id. ¶¶ 24, 34-42; see Fig. 1 (Intel example), Figs. 6-8 (results). 17 Id. ¶¶ 43- 3 49, Figs. 11-12. The regressions indicated that the “vast majority” of employees fall within titles 4 or groups that show: (1) that gains for the titles or groups are shared broadly at the same time and 5 (2) that gains for some are shared with others in different job titles in a subsequent year. See id. ¶ 6 8. 18 Dr. Leamer contended that this is consistent with his previous opinion that “all or almost all 7 Defendants’ employees would have been impacted by the non-compete agreements.” Id. 8 Moreover, Dr. Leamer opined that the fact that gains were shared over time strongly indicated that 9 an internal sharing force, rather than only external market forces, drove the structure of class United States District Court For the Northern District of California 10 member compensation. Id. Dr. Leamer also demonstrated there is a much stronger correlation 11 between compensation of job titles within a firm than there is between compensation of job titles 12 between firms. This finding further reinforced his conclusion that each Defendant maintained a 13 somewhat rigid pay structure and undermined Defendants’ contention that internal forces and pay 14 structure play no role in setting compensation. Id. ¶¶ 65-68. 15 Dr. Leamer recognized that his analyses do include outliers—job titles that do not 16 positively correlate to the average or do not show sharing over time. Id. ¶ 12. He noted, however, 17 that the number of outliers was small, and in most cases the outliers involved titles with incomplete 18 data. See id. ¶¶ 12, 50-64. Thus, the presence of a few outliers did not undermine his basic 19 conclusions about how Defendants paid their employees. Such conclusions were also supported by 20 17 21 22 23 24 25 26 27 28 Dr. Leamer also estimated regressions by splitting the Technical Class titles into deciles. Dr. Leamer explains that, to form the ten groups, he ranked titles on the basis of average (inflation adjusted) total compensation over the lifetime of the title and then divided these up into deciles based on employee-years. Although Dr. Leamer attempted to break the firms up into 10 equal sized groups (equal based on employee years), some groups ended up being larger than others because some titles were more populous than others. Leamer Suppl. Rep. ¶ 43, n.8. 18 Plaintiffs argue that this latter result is particularly significant. Defendants’ expert, Dr. Murphy, initially contended that an “alternative hypothesis that the level of compensation of Defendants’ employees is broadly determined by competition in a vast labor market for similar employees and that adjustments for unique circumstances of particular employees are highly individualized.” Murphy Rep. ¶ 89. Plaintiffs argue that this contention becomes unsupportable when used to explain why gains for some are shared with others in a subsequent year; there is not a sensible reason that an external force such as increased demand for computers would affect some employees in one year and the rest in the next, without resort to internal forces such as fairness concerns. Leamer Suppl. Rep. ¶ 8 (“Furthermore, the sharing of gains over time strongly indicates the existence of an internal sharing force driving the structure of class member compensation, rather than only external market forces.”). 62 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 economic theory and the evidentiary record. See id. ¶ 64. Importantly, Dr. Leamer also contended 2 that he had not seen any evidence that any of the titles within the Technical Class would not have 3 been harmed by the anti-solicitation agreements. Id. 4 Ultimately, Dr. Leamer concluded that common proof, “in the form of documents, data, 5 economic theory, and statistical methodologies,” were capable of demonstrating that the anti- 6 solicitation agreements artificially suppressed compensation of all or nearly all members of the 7 Technical Class. Leamer Rep. ¶ 149. 8 9 United States District Court For the Northern District of California 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 ii. Dr. Hallock’s Opinions Based on Defendants’ Testimony, Contemporaneous Documents, and Data In further support of Class Certification, Plaintiffs present a second expert report from Dr. Kevin F. Hallock, a leading labor economist and an expert in compensation structure and design. See Hallock Rep. ¶¶ 1-3. Dr. Hallock investigated whether Defendants used formal administrative pay systems, and whether the anti-solicitation agreements at issue would have suppressed the compensation of all or nearly all members of the Technical Class. In forming his opinions, Dr. Hallock reviewed only common evidence: Defendants’ testimony, contemporaneous documents, and data. Suppl. Class Cert. Mot. at 2. Dr. Hallock found that Defendants all used formalized compensation systems that organized employees into pay ranges, grades, or families under umbrella systems. Hallock Rep. ¶ 45. In finding that all Defendants utilized formalized pay systems, Dr. Hallock relied on evidence that, among other things: (1) Defendants sorted their employees into job families and/or grades; (2) Defendants utilized salary ranges with a minimum, mid-point and maximum set based on external employment market data; and (3) Defendants used internal tools to assist managers with setting other employees’ compensation levels. See id. ¶¶ 45-109. An important feature of these formal systems is that job titles, levels, and grades are valued relative to all other employee categories in the company. Employees who receive compensation outside of their guideline ranges are identified and corrected to bring them in line with the company structure. See, e.g., id. ¶¶ 114, 140, 160, 166, 181. According to Dr. Hallock, “[i]f the ratios [between a person and someone else who is similarly situated] diverge from each other, the 63 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 person will experience reactions of unfairness and inequity,” thus making internal equity 2 “important not only in setting up the original structure of a compensation system but also when 3 managing it.” Id. ¶ 110 (internal quotation marks omitted). 4 Dr. Hallock also found that Defendants used their compensation system to pay their 5 employees in systematic and structured ways. Dr. Hallock found that nearly all compensation 6 decisions were made company-wide on an annual basis and in a fashion that preserves existing 7 compensation relationships. When Defendants made “out of cycle” adjustments to retain certain 8 employees (such as to make counteroffers or pay retention bonuses in light of a competitor’s 9 solicitation), Defendants were careful to adjust the system to take the exceptions into account. Dr. United States District Court For the Northern District of California 10 Hallock also found that Defendants adhered to principles of internal equity whereby similarly 11 situated and similarly performing employees were paid similarly. See id. ¶¶ 111-81. 12 Given Defendants’ formalized pay structures and compensation design, as well as issues of 13 equity and fairness present in the Defendant firms, Dr. Hallock opined that the anti-solicitation 14 agreements would have a widespread and systematic impact on compensation. Id. ¶ 237. First, Dr. 15 Hallock opined that “[a] direct impact on pay could occur if an employee did not receive a cold 16 call, or if the upward wage pressures on any of the employees in related groups or job families 17 were disrupted.” Id. ¶ 238. For example, Dr. Hallock hypothesized that one way that pay could be 18 lowered at Defendants for nearly all workers has to do with extraordinary employees. Dr. Hallock 19 noted that Defendants employed certain elite employees, and that cold calling often targeted these 20 top employees. Id. ¶ 239. Since the anti-solicitation agreements suppressed salaries of these top 21 employees and therefore lowered “the top of the box” in terms of the salary range, Hallock opined 22 that the entire box may be lowered as well, thus impacting “nearly all other workers.” Id. 23 Dr. Hallock also opined that external market data could influence wages, as there is 24 evidence that Defendants benchmark their data to external sources, most commonly Radford or 25 Croner. See id. ¶ 240. “[T]o the extent that pay is lowered at other firms through anti-competitive 26 and other behavior of firms,” Dr. Hallock found that “the market data they use for their own 27 structure will be lower” and hence “their own pay levels will be lower than they would be in the 28 absence of such agreements.” Id. 64 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Finally, in support of Plaintiffs’ Technical Class definition, Dr. Hallock examined 2 Defendants’ pay structures and compensation design as they pertain to the Technical Class, and 3 concluded that the same mechanisms that would have transmitted pay suppression throughout the 4 Defendants’ firms apply with even greater force to technical employees. Id. ¶ 246. Thus, if the 5 anti-solicitation agreements suppressed the pay of certain members of the Technical Class, all or 6 nearly all other members would be expected to have also been impacted. See Suppl. Mot. at 3. 7 c. The Court’s Conclusions and Defendants’ Contentions 8 Defendants argue that this Court should not certify the Technical Class because 9 individualized inquiries regarding who was impacted will predominate over common questions. United States District Court For the Northern District of California 10 Defendants contend that their compensation policies and practices were highly individualized with 11 wide variation in compensation. According to Defendants, compensation was set by hundreds of 12 different managers who were directed to differentiate pay and reward high achieving employees. 13 As such, Defendants argue that pay raises to one employee would not necessarily affect the salary 14 of all other employees in the Technical Class. 15 In furtherance of their contention that individualized issues predominate, Defendants 16 contend that the individual pieces of evidence offered by Plaintiffs are unpersuasive. However, as 17 discussed below, Defendants cannot rebut the voluminous documentary evidence from Defendants’ 18 internal files and the expert reports that rely on this documentary evidence. 19 First, the Court finds, as it did previously, that Dr. Leamer’s market price discovery and 20 internal equity hypotheses offer theories subject to common proof for how Defendants’ anti- 21 solicitation agreements suppressed compensation broadly. See Apr. 5 Class Cert. Order at 21. 22 While Defendants’ expert, Dr. Murphy, criticized the economic literature upon which Dr. Leamer 23 relied, Dr. Murphy did not dispute the basic principles of information economics undergirding Dr. 24 Leamer’s hypothesis. See. e.g., Harvey Decl., ECF No. 297, Ex. 13 (Murphy Depo.) at 188:6-14; 25 192:25-193:6; 194:10-196:10; 197:7-19. Similarly, although Dr. Murphy criticized Dr. Leamer’s 26 initial expert report because it did not compare the importance of maintaining internal equity to 27 other goals, such as procedural equity or the value of rewards for individual contributions as a 28 loyalty motivator, see Murphy Rep. ¶ 81, the Court does not find that this undermines Dr. Leamer’s 65 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 hypothesis that internal equity played some role in affecting employment compensation. This is 2 particularly true in light of the extensive documentary evidence showing that Defendants valued 3 internal equity. 4 Second, the Court finds that Plaintiffs’ documentary evidence provides substantial further 5 support for Plaintiffs’ method of proving impact. Indeed, at trial, the Court predicts that this 6 evidence is likely to be among the most persuasive to a jury as it illustrates and confirms many of 7 the actual dynamics at play within Defendants’ firms. While Defendants characterize Plaintiffs’ 8 new evidence as “mostly old and off point,” see Suppl. Opp’n at 13, the Court finds that this 9 evidence significantly bolsters Plaintiffs’ showing that their method of proving impact will turn on United States District Court For the Northern District of California 10 common evidence. For example, rather than a few documents showing that some Defendants 11 valued internal equity in their compensation practices, Plaintiffs’ documentary evidence now 12 indicates that all Defendants valued internal equity. Further, the evidence now suggests that 13 internal equity was such an important aspect of Defendants’ compensation practices that: (1) 14 Defendants utilized software tools to generate internal equity reports and to compare each 15 employee to his or her peers; (2) Defendants advised managers that internal equity was a prime 16 consideration when setting and adjusting salaries; and (3) Defendants actively monitored their 17 compensation structure to identify discrepancies within and beyond job titles and groups and to 18 make adjustments as necessary. 19 Despite this documentary evidence, Defendants contend that managers exercised broad 20 discretion when setting and adjusting salaries and that Defendants valued performance—to which 21 Defendants refer as “pay for performance”—over internal equity. However, Defendants’ 22 documents indicate otherwise. For example, in a 2004 Human Resources presentation, Intel states 23 that, although “[c]ompensation differentiation is desired by Intel’s Meritocracy philosophy,” “short 24 and long term high performer differentiation is questionable.” Harvey Suppl. Decl., Ex. 10 at 13. 25 Indeed, Intel notes that “[l]ack of differentiation has existed 26 .” Id. at 19. As key “[v]ulnerability [c]hallenges,” Intel identifies: (1) “[m]anagers (in)ability 27 to distinguish at [f]ocal”—“actual merit increases are significantly reduced from system generated 28 increases,” “[l]ong term threat to retention of key players”; (2) “[l]ittle to no actual pay 66 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 differentiation for 2 ”; and (3) “[n]o explicit strategy to differentiate.” Id. at 24 (emphasis added). Further, the documentary evidence indicates that even where “pay for performance” exists, 4 such differentiation is not inconsistent with Dr. Leamer’s theories of internal equity. For example, 5 a chart from Google shows that for a given level of performance, the higher the pre-adjustment 6 position, the lower the merit increase. See, e.g., Cisneros, Ex. 1609.9 (Google documents noting 7 that it “tries to manage salaries in the 8 strong performers who are paid relatively high.”). Specifically, Google’s documents show that an 9 employee whose salary is below the target salary would receive a larger percentage pay increase 10 United States District Court For the Northern District of California 3 than an employee whose salary exceeds the target even though both employees receive the same 11 performance review. See Cisneros, Ex. 1855.107. This demonstrates the coexistence of 12 performance-based pay and concerns of internal equity. 13 range and thus give minimal increases to very To contend that they valued “pay for performance” over internal equity, Defendants rely 14 principally on declarations from top management in their human resources, recruitment, 15 compensation, and benefits departments. However, the Court has already recognized that these 16 declarations were drafted for the specific purpose of opposing Plaintiffs’ class certification motion, 17 and accordingly, the Court finds that these documents are of a diminished probative value. Cf. In 18 re Wells Fargo Home Mortg. Overtime Pay Litig., 527 F. Supp. 2d 1053, 1061 (N.D. Cal. 2007) 19 (scrutinizing carefully declarations from Defendants’ employees that appeared “litigation driven”). 20 Indeed, many of the claims made in those declarations are inconsistent with the discovery Plaintiffs 21 obtained from Defendants after the hearing on Plaintiffs’ initial class certification motion. For 22 example, in a declaration created for the purposes of opposing the initial class certification motion, 23 Mason Stubblefield (Intuit Vice President of Human Resources) stated that “Intuit does not use 24 salary bands or ranges, either for existing employees or new hires.” Stubblefield Decl. ¶ 10. Yet, 25 in his deposition after the hearing on the class certification motion, Mason Stubblefield took the 26 diametrically opposite position, when he stated that “[t]here are 27 company . . . . Each job that we have fits within a band . . . . Jobs fit into levels . . . . And different 28 numbers of levels get used in different job families based on business needs.” Stubblefield Depo. bands [at Intuit] inside the 67 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 87:8-88:24. In light of these demonstrated inconsistencies, the Court is more persuaded by the 2 internal documents Defendants created before and during the anti-solicitation agreements, such as 3 CEO-to-CEO emails, presentations regarding compensation and recruitment from the heads of 4 Defendants’ human resources departments, and inter-office communications about internal equity 5 concerns, than the declarations Defendants created to oppose class certification and testimony that 6 is litigation driven. 7 Defendants further rely on Dr. Shaw’s report to rebut Plaintiffs’ contentions that 8 Defendants maintained formalized compensation structures. Dr. Shaw opines that Defendants 9 maintain a “pay for performance” philosophy implemented by individual managers based on United States District Court For the Northern District of California 10 subjective evaluations of employees. Shaw Rep. ¶ 16. Dr. Shaw further contends that Dr. 11 Hallock’s “top of the box” theory is unsupportable. See id. ¶¶ 41, 42, 54, 62, 66. Dr. Shaw opines 12 that, “[b]ased on Defendants’ compensation systems, pay practices, and pay philosophy,” she 13 “would not expect that a suppression of wages to some employees would affect all or nearly all 14 Technical Class members.” Id. at 27. 15 The Court finds Dr. Shaw’s criticisms of Plaintiffs’ theories and of Dr. Hallock’s report 16 unpersuasive. Dr. Shaw’s report is conclusory and contrary to the overwhelming evidence in the 17 record. For example, Dr. Shaw relies heavily on the declarations Defendants created to oppose 18 class certification even though many of the claims in those declarations are inconsistent with 19 Defendants’ own internal documents. See id. at 20-21 n.25 & n.26; 21 n.30 & n.32; and 23 n.35; 20 Shaw App. C ¶¶ 1, 2, 3, 4, 7, 8, 10, 11, 18, 24; Shaw App. D ¶¶ 1, 9. Further, although Dr. Shaw 21 emphasizes managers’ broad discretion in setting compensation, Dr. Shaw did not systematically 22 investigate whether Defendants supervised and controlled their managers’ use of discretion, and 23 Dr. Shaw admits that she did not assess whether managerial discretion made any significant 24 difference to employee pay. See Shaver Suppl. Decl. Ex. O (Shaw Depo.) 74:1-75:16, 93:16-22, 25 98:14-15. Dr. Shaw also asserts that, in technology-based firms, pay ranges assigned to job codes 26 are “mere guidelines for managers” and that the pay of workers is highly individualized. See Shaw 27 Rep. ¶ 30. However, the Court finds that Dr. Shaw’s emphasis on managerial discretion is contrary 28 to the extensive documentary evidence discussed above, which suggests that the exercise of any 68 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 discretion was limited and that managers had to obtain authorization to deviate from Defendants’ 2 compensation structure. See supra section IV.B.2.a.ii; see, e.g., Cisneros Decl., Ex. N (Fadell 3 Depo.) at 53 (stating that to deviate from a prescribed salary range, Apple managers “would have 4 to pull in a bunch of people to then approve anything outside of that range”); id., Ex. C (Arriada- 5 Keiper Depo.) at 24 (Adobe admonished managers who deviated from salary ranges that Adobe 6 had “a minimum for a reason”). Moreover, Dr. Shaw’s contentions regarding the individualized 7 nature of compensation are contrary to the statistical analysis provided by Dr. Leamer—which 8 show that in 97% of class member employee-years, pay was within the prescribed range. Leamer 9 Suppl. Reply Rep. ¶¶ 31, 67. United States District Court For the Northern District of California 10 Defendants’ principal contentions, however, are challenges to Dr. Leamer’s statistical 11 analyses. 19 Before the Court turns to an analysis of the competing methodologies of Dr. Leamer 12 and Defendants’ expert, Dr. Murphy, the Court notes that the importance of these statistical models 13 is diminished in light of the extensive documentary evidence that supports Plaintiffs’ theory of 14 impact. In other contexts, courts have long noted that statistical and anecdotal evidence must be 15 considered in tandem. See Coral Const. Co. v. King Cnty., 941 F.2d 910, 919 (9th Cir. 1991) 16 (“[T]he combination of convincing anecdotal and statistical evidence is potent.”). This Court could 17 not identify a case at the class certification stage with the level of documentary evidence Plaintiffs 18 have presented in the instant case. The Court agrees with Dr. Leamer that “interpretation of non- 19 experimental data needs to be sensitive to the context in which the data were generated, and 20 persuasive conclusions from the numerical data require the information in the numerical data and 21 the documents to be aligned.” Leamer Suppl. Rebuttal Rep. ¶ 11. After all, class certification 22 23 19 24 25 26 27 28 Defendants moved to strike Dr. Leamer’s report in support of Plaintiffs’ initial motion for class certification for failure to provide reliable, relevant, and admissible testimony under Daubert v. Merrell Dow Pharmaceutical, 509 U.S. 579 (1993), and Rule 702 of the Federal Rules of Evidence. Defs.’ Mot. to Strike Rep. of Dr. Edward E. Leamer (“Mot. to Strike”). ECF No. 210. The Court rejected Defendants’ motion in its April 5 Class Certification Order. See Apr. 5 Class Cert. Order at 49-50 (“While the Court has concerns about the probativeness of some of Dr. Leamer’s statistical evidence . . . the Court does not find this evidence is so methodologically flawed as to warrant exclusion.”). In their Opposition to the Supplemental Motion for Class Certification, Defendants do not raise any new Daubert challenges to the expert opinions set forth by Dr. Leamer or Dr. Hallock. See Suppl. Opp’n. 69 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 requires a holistic, qualitative assessment; as the Seventh Circuit has noted, the class certification 2 analysis is not “bean counting.” Butler, 727 F.3d at 801. The Court now turns to Defendants’ contentions that Dr. Leamer’s methodologies are 4 unpersuasive. Defendants contend that Dr. Leamer should not have relied on averages in his 5 correlation and multiple regression analyses because (1) averaging masks individualized issues, (2) 6 Dr. Leamer’s regression analysis is faulty due to an endogeneity problem, and (3) Dr. Leamer’s 7 statistical evidence cannot show causation. Suppl. Opp’n 13-14. Defendants argue that these flaws 8 render dubious Dr. Leamer’s conclusion that Defendants maintained a “rigid wage structure.” See 9 id. at 1-2. The Court finds that several of Defendants’ arguments are contrary to their own internal 10 United States District Court For the Northern District of California 3 documents and methodologies. Moreover, the Court finds that Defendants’ remaining contentions 11 do not suggest that Dr. Leamer’s statistical analyses should be rejected wholesale, particularly in 12 light of the extensive documentary evidence that supports Dr. Leamer’s conclusions. Thus, as 13 discussed below, the Court rejects Defendants’ contentions and accepts Dr. Leamer’s methodology. 14 Defendants’ primary criticisms of Dr. Leamer’s supplemental report turn on his use of 15 averaging in his correlation and multiple regression analyses. See Suppl. Opp’n at 1-2. Defendants 16 take issue with Dr. Leamer’s use of averages in his job-title-by-job-title correlation analysis, which 17 analyzes the “movement over time of the average compensation of each title with the average 18 compensation of the firm’s Technical Class.” Leamer Suppl. Rep. ¶ 4. Defendants contend that, 19 “[b]y averaging the compensation of all employees who hold the same job title or fall into the same 20 decile,” Dr. Leamer “necessarily wipes out the very thing he is supposed to be measuring—the 21 significant variation in individual employees’ compensation.” Suppl. Opp’n at 5. According to 22 Defendants, Dr. Leamer’s “correlation of averages would reach the same conclusion regardless of 23 whether all employees with the same job title received identical or vastly different compensation 24 over time and whether their compensation moved in lockstep or in opposite directions.” Id. at 6. 25 Defendants also object to Dr. Leamer’s use of averages in his company-specific multiple regression 26 models, designed to detect the effects of internal forces (i.e., a wage structure) acting on class 27 member compensation, as opposed to merely external market forces. See id. at 5-6; 11-13. 28 Specifically, Defendants claim that Dr. Leamer’s multiple regression analysis masks individual 70 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 variation by using average job title compensation data rather than individual compensation data. 2 Id. at 13. 3 However, the Court finds that Dr. Leamer’s averaging of the data appears to yield results that, in the context of the correlation and multiple regression analyses, are consistent with 5 Plaintiffs’ theory that there is a somewhat rigid wage structure. Dr. Leamer notes in his rebuttal 6 supplemental report that “the inherent noise in the individual level data tends to drown out the 7 signal of the internal pay structure [Plaintiffs] are trying to detect.” Leamer Suppl. Rebuttal Rep. ¶ 8 32. Dr. Murphy admits that averaging aggregate data is an appropriate statistical tool for the same 9 reasons given by Dr. Leamer. Shaver Decl., Ex. N (Murphy Depo.) 553:18-20 (“The reason you 10 United States District Court For the Northern District of California 4 do the averaging is so that you are left with a more systematic part and the idiosyncratic parts get 11 averaged out.”). 12 With respect to both correlation and multiple regression analyses, Dr. Leamer averaged the 13 compensation of employees within each job title. Leamer Suppl. Rep. ¶¶ 18-29. While this kind of 14 averaging may have masked some of the individual variations within each job title, it was 15 necessary to determine whether there was a wage structure across job titles. See Aug. 8 Tr. at 16 97:11-22; 106:5-13. 20 The Court is therefore not persuaded that it should disregard the correlation 17 analysis for the purposes of showing a wage structure across job titles. The Common Factors 18 Analysis shows that approximately 90 percent of each employee’s compensation is explained by 19 common factors—primarily by job title. Leamer Rep. ¶ 128; Aug. 8 Tr. at 105:10-24. Because 20 individual compensation is primarily determined by job title, as shown by the Common Factors 21 Analysis (which was run on an employee-by-employee basis without averaging, see Leamer Rep. ¶ 22 20 23 24 25 26 27 28 Dr. Leamer explains in his supplemental report that he chose to work with title averages because: individual data is likely to be dominated by forces that operate at the individual level, which can make it difficult to detect the firm wide effects including the spread of the anti-cold-calling agreements broadly across the firms. Averaging across individuals in a title can average out the individual effects, thus making the firmwide effects more transparent. In addition, a title-level analysis provides a clearer perspective on the compensation structures the documentary evidence shows Defendants used to maintain their many employees and maintain internal equity among their employees. Leamer Suppl. Rep. ¶ 19. 71 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 128, Fig. 11), and the correlation and multiple regression analyses show that compensation across 2 job titles move together, the Court believes that Dr. Leamer’s analysis bolsters Plaintiffs’ theory 3 that there is a wage structure in place under which an impact on some employees would have 4 resulted in an impact to all or nearly all employees. The Court further notes that Dr. Leamer’s 5 conclusions in the Common Factors and correlation analyses are consistent with the documentary 6 evidence that suggests that Defendants maintained a formal wage structure and valued internal 7 equity. 8 9 Defendants’ contention that averaging rendered unpersuasive Dr. Leamer’s analyses is particularly unconvincing considering that Defendants themselves often aggregated their entire United States District Court For the Northern District of California 10 compensation budget and compared it to the budgets of other firms, or matched job title 11 compensation within the company to similar titles across multiple companies. See, e.g., Bock 12 Depo. 44:12-20 (Google Senior Vice President of People Operations, testifying that Google’s 13 bonus amounts were calculated by taking “a set of people in like jobs, you calculate the 14 15 . All those people, because they’re in like jobs, will have the same will be based on the . . . . The rather than on their 16 .”); see also Shaver Decl., Ex. 122 (email from Lori McAdams (Pixar Vice President of 17 Human Resources and Administration) to Sharon Coker (former Lucasfilm Director and Senior 18 Director of Human Resources), among others, asking about others’ salary increase budget for FY 19 ’07 and stating “Ours is 20 anything close, more or less?”); Cisneros Decl., Ex. F (Streeter Depo.) at 46-47 (discussing how 21 Adobe uses aggregated and average data from Radford for job titles and classifications, which it 22 then uses in developing its own compensation structure). 23 , but we may manage it closer to on average. Are you doing Defendants further criticize Dr. Leamer’s supplemental expert report on another similar 24 basis. Defendants contend that Dr. Leamer ignores data that shows substantial variation in the 25 levels and changes in individual employees’ compensation. In contrast to Dr. Leamer, Dr. Murphy 26 examines variation at the individual employee level and finds substantial divergence in 27 compensation over time for individuals who start with the same job title. Defendants argue that 28 these results show that it is wrong to infer that an increase in compensation for some employees 72 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 would result in an impact to all or nearly all employees. Rather, Defendants contend that the 2 results reflect managerial discretion. Opp’n at 10 (citing Murphy Rebuttal Rep., Exs. 2-3, & App. 3 B). 4 However, that Defendants differentiated pay is not inconsistent with Dr. Leamer’s finding 5 that the Defendants maintained compensation structures that restrained that differentiation. Indeed, 6 as discussed above, the evidence supports the notion that differentiation, such as based on a “pay 7 for performance” philosophy, is not inconsistent with theories of internal equity. See, e.g., 8 Cisneros Decl., Ex. 1609 (Google Chart showing that a high performance rating will result in an 9 employee compensated at of Google’s target salary having his/her salary increased by a United States District Court For the Northern District of California 10 higher percentage than an individual with the same rating but compensated at a level higher than 11 Google’s target salary). For example, Google notes that it “tries to manage salaries in the 12 range and thus give minimal increases to very strong performers who are paid relatively 13 high.” Id.; see also Cisneros Decl., Ex. 1855.107 (showing that managerial discretion and 14 performance based compensation is not necessarily inconsistent with principles of internal equity at 15 Google). 16 Second, Defendants contend that Dr. Leamer’s multiple regression analysis is unpersuasive 17 because it suffers from an endogeneity problem. Suppl. Opp’n at 13. An “‘endogeneity problem[]’ 18 . . . arises when some of the same unmeasured common factors drive both the independent and 19 dependent variables.” Murphy Suppl. Rep. ¶ 43. Defendants argue that Dr. Leamer’s “external” 20 San Jose-Sunnyvale-Santa Clara Metropolitan Statistical Area variable and unnamed omitted 21 variables affect “both average job title compensation [the dependent variable] and Leamer’s 22 ‘internal’ firm-wide average compensation variable [the independent variable].” Suppl. Opp’n at 23 13. Defendants argue that by “[i]gnoring this ‘endogeneity’ among his variables, Dr. Leamer 24 improperly concludes that the “internal” variable meaningfully explains average job title 25 compensation . . . while the ‘external’ variable does not.” Id. Defendants contend that this is a 26 fundamental error that renders Dr. Leamer’s model uninformative and his inferences from it 27 unsound. Id. 28 73 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 The Court is not persuaded by this concern. With respect to the San Jose variable, 2 Defendants have not provided quantitative analysis to show that the San Jose employment variable 3 skews the results with respect to the internal firm-wide average compensation variable. Rather, 4 Defendants offer the endogeneity theory without any support in the form of data, analysis, or case 5 law. A similar lack of specificity plagues Defendants’ argument with respect to the unknown 6 variables that Defendants contend cause this endogeneity problem. Specifically, as Plaintiffs note, 7 Defendants have not “identif[ied] a single omitted variable, or show[n] how adding one would 8 change the results.” Suppl. Reply at 9; see also Leamer Suppl. Rebuttal Rep. ¶ 61 (noting that Dr. 9 Murphy has “not presented any analysis showing that omitted non-sharing external or internal United States District Court For the Northern District of California 10 11 effects are responsible for the positive sharing in [Dr. Leamer’s] results”). Finally, Defendants argue that Dr. Leamer’s analyses cannot address the question of 12 causation underlying Plaintiffs’ theory of classwide impact, which Defendants characterize as 13 “whether compensation for class members was so rigidly interlinked that a wage increase for some 14 would cause a wage increase for substantially all.” Suppl. Opp’n at 14. That is, Defendants 15 contend that Plaintiffs have not demonstrated that the anti-solicitation agreements caused any 16 depression in wages. However, Plaintiffs clarify that “Plaintiffs never argued that the impact of the 17 agreements would have been ‘lockstep’—that a $5 raise to one employee would have required a 18 simultaneous $5 raise across the firm.” Suppl. Reply at 8. Rather, as shown by the documentary 19 evidence, “by shielding their employees from waves of recruiting, Defendants not only avoided 20 individual raises, they also avoided having to make across-the-board preemptive increases to 21 compensation.” Id. 22 As the documentary evidence above demonstrates, there is compelling evidence that in the 23 absence of the anti-solicitation agreements, Defendants would have had to make structural 24 preemptive or reactive changes. See, e.g., Cisneros Decl., Ex. VV at 106 (Sheehy Depo.) 25 (explaining that a group of Pixar employees’ base salary needed to be increased because Pixar was 26 “competing with technology companies in the Bay Area, and [its] recruiting team was hearing from 27 candidates that they were getting better offers elsewhere.”); Leamer Suppl. Rebuttal Rep. ¶ 23 28 (citing deposition of Patricia Murray (Intel Senior Vice President and Director of Leadership 74 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Strategy and former President of Human Resources), in which Murray explained, in response to a 2 question regarding whether Intel raised compensation for particular job categories or job ranges to 3 preemptively prevent attrition, “[i]f attrition was high in a particular job category somewhere in the 4 globe, that would be a fact considered in the many facts we assessed”). The Court finds persuasive 5 Dr. Leamer’s statement that economists “analyze correlations, which are routinely used. . . to draw 6 causal conclusions when supported by compelling frameworks and complementary information.” 7 Leamer Suppl. Rebuttal Rep. ¶ 13. In the instant case, the compelling documentary evidence along 8 with Plaintiffs’ expert theories and correlation analyses are capable of demonstrating causation on 9 a classwide basis. United States District Court For the Northern District of California 10 In sum, the Court does not find persuasive Defendants’ criticisms of Dr. Leamer’s 11 methodology. Defendants have not presented any analysis that undermines Dr. Leamer’s analysis, 12 and many of Defendants’ arguments are contradicted by the documentary evidence. Accordingly, 13 the Court finds that Dr. Leamer’s methodology, in conjunction with and bolstered by the extensive 14 documentary evidence, is sufficient to meet the predominance standard with respect to impact. 15 d. Conclusion on Impact 16 Plaintiffs’ documentary evidence, along with the expert reports and statistical analyses that 17 rely on this evidence, establish that common issues between class members will predominate over 18 individual issues in proving antitrust impact. The documentary evidence supports Plaintiffs’ theory 19 that Defendants had formal compensation structures on a company-wide basis that placed a 20 premium on internal equity concerns, and that collusive communications between various 21 Defendants limited the proportion of each Defendants’ budget that would be dedicated to merit 22 increases. Further, the evidence suggests that the Defendants benchmarked their compensation 23 structures to external data and to each other. The documentary evidence and the expert reports also 24 support Plaintiffs’ theory that the top companies and top employees at these companies set a 25 ceiling, based on which all or nearly all employees of the Technical Class’s compensation was set. 26 This extensive evidence supports Plaintiffs’ theory that each technical employee’s compensation 27 was linked to those of her peers within and across Defendants’ firms. 28 75 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Plaintiffs have also presented documentary and expert evidence about labor market 2 dynamics that would have affected compensation on a classwide basis. For example, Defendants 3 would have had to take preemptive steps, through bonuses and raises to ward off threats of 4 poaching by other Defendants. Further, Defendants would have had to react to any cold calling 5 and poaching of employees by increasing compensation in response to the spread of information 6 from outside and within Defendants’ firms. Because of the structural concerns regarding internal 7 equity and the existing rigid compensation structures, the evidence suggests that any such 8 preemptive or reactive steps, in the absence of the anti-solicitation agreements, would have had 9 classwide effect and would have impacted all or nearly all members of the Technical Class. United States District Court For the Northern District of California 10 One example, which concerns interactions between Google and non-party Facebook, 11 illustrates many of these theories, including the significance of the threat imposed by cold calling, 12 the disruption of internal equity, and how these dynamics would have, in the absence of the anti- 13 solicitation agreements, prompted Defendants to offer employees incentives to minimize attrition. 14 In March of 2008, Arnnon Geshuri (Google Recruiting Director) discovered that non-party 15 Facebook had been cold calling into Google’s Site Reliability Engineering (“SRE”) team. 16 Geshuri’s first response was to suggest contacting Sheryl Sandberg (Chief Operating Officer for 17 non-party Facebook) in an effort to “ask her to put a stop to the targeted sourcing effort directed at 18 our SRE team” and “to consider establishing a mutual ‘Do Not Call’ agreement that specifies that 19 we will not cold-call into each other.” Cisneros Decl., Ex. 614. Arnnon Geshuri also suggested 20 “look[ing] internally and review[ing] the attrition rate for the SRE group,” stating, “[w]e may want 21 to consider additional individual retention incentives or team incentives to keep attrition as low as 22 possible in SRE.” Id. (emphasis added). Finally, an alternative suggestion was to “[s]tart an 23 aggressive campaign to call into their company and go after their folks—no holds barred. We 24 would be unrelenting and a force of nature.” Id. 25 In August of 2008, after losing one of many employees to Facebook, Google’s Vice 26 President of Communications emailed Google’s executive management group and Bill Campbell 27 (Chairman of Intuit Board of Directors, Co-Lead Director of Apple, and advisor to Google). In this 28 email, the Google Vice President expressed concern about Facebook’s “poaching” and stated that 76 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 she had “offered [the employee] different roles and discussed his future at Google” but that she had 2 “gone as far as [she could] without making promises about pay or title that would cause significant 3 problems across [her] team.” Id. Bill Campbell’s response was to ask, “Who should contact 4 Sheryl [Sandberg] (or [Facebook Founder] Mark [Zuckerberg]) to get a cease fire? We have to get 5 a truce.” Id. Facebook refused. 6 Facebook continued to poach Google’s employees. In 2010, for example, of 7 Facebook’s new employees were recruited from Google. Leamer Rep. ¶ 109. Accordingly, in 8 October 2010, Google began studying Facebook’s solicitation strategy. Id. A month later (and two 9 months after the DOJ made public its investigation of Defendants), Google announced its “Big United States District Court For the Northern District of California 10 Bang,” which involved an increase to the base salary of all of its salaried employees by 10% and 11 provided an immediate cash bonus of $1,000 to all employees. Shaver Decl., Ex. 46. In an internal 12 email, Laszlo Bock (Google Senior Vice President of People Operations) explained that the 13 rationale for the Big Bang included: (1) being “responsive to rising attrition;” (2) supporting higher 14 retention because “higher salaries generate higher fixed costs;” and (3) being “very strategic 15 because start-ups don’t have the cash flow to match, and big companies are (a) too worried about 16 internal equity and scalability to do this and (b) don’t have the margins to do this.” Shaver Decl., 17 Ex. 48. 18 The increased salaries within Google impacted the other Defendants as well. For example, 19 Google’s “Big Bang” disrupted Apple’s compensation structure by forcing Apple to pay more for 20 new hires which then created internal equity issues with current employees. David Alvarez (Apple 21 Recruiting Manager and former Research Manager) responded as follows: 22 23 24 25 26 We will go back and review internal equity again and come up with a new proposal for your review. . . . This is due to the market changing. This will most likely create an internal comp issue with current employees. In order to stay ahead and be competitive, we must be aggressive with our offers in order to close good talent. Having these new data points will help justify adjusting current employees. Great talent does come at a cost. . . . Just as a data point, Google gave its employees a 10% increase to deal with the aggressive offers out in the market. 27 Cisneros Decl., Ex. 1376. Google’s reaction to Facebook’s solicitation of its employees, and 28 Apple’s subsequent reaction to Google’s actions, demonstrates the weakness of Defendants’ 77 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 contention that the impact of the anti-solicitation agreements was localized to individual employees 2 who would have been solicited but for the agreements or that class members would have been 3 impacted differently depending on their individual circumstances. These contemporaneous 4 reactions of the Defendants to genuine competition for labor suggest instead that their response to 5 widespread solicitation of their employees would have been structural rather than individual. 6 7 3. Damages In addition to disputing whether Plaintiffs can show impact on a classwide basis, the parties dispute whether Plaintiffs can show damages on a classwide basis. The Supreme Court has held 9 that damages “[c]alculations need not be exact, but at the class-certification stage (as at trial), any 10 United States District Court For the Northern District of California 8 model supporting a ‘plaintiff’s damages case must be consistent with its liability case, particularly 11 with respect to the alleged anticompetitive effect of the violation.’” Comcast, 133 S. Ct. at 1433 12 (citing ABA Section of Antitrust Law, Proving Antitrust Damages: Legal and Economic Issues 57, 13 62 (2d ed. 2010)). In other words, “a damages suit cannot be certified to proceed as a class action 14 unless the damages sought are the result of the class-wide injury that the suit alleges.” Butler, 727 15 F.3d at 799 (emphasis in original). 16 Here, Plaintiffs rely on their expert, Dr. Leamer, to demonstrate that they can use reliable 17 methods to compute damages by applying classwide methods and analyses. See Class Cert. Mot. at 18 22 (citing Leamer Rep. ¶¶ 135-48). Dr. Leamer concluded that common evidence and a regression 19 approach could be used to create a model for quantifying the estimated cost to Technical Class 20 members resulting from Defendants’ challenged conduct. Id. ¶¶ 141-48. This model generated 21 percentages by which Defendants undercompensated their Technical Class employees in each of 22 the conspiracy years. Id., Fig. 24. Specifically, Dr. Leamer estimated the effect of the anti- 23 solicitation agreements by contrasting compensation during the periods when the anti-solicitation 24 agreements were in effect with compensation before and after the anti-solicitation agreements. Id. 25 ¶ 136. Dr. Leamer’s model incorporated a range of variables designed to account for factors 26 including: (1) age, sex, and years at the company; (2) the effects on compensation caused by the 27 anti-solicitation agreements; (3) the effects caused by factors specific to each Defendant (e.g., firm 28 revenue, total number of new hires, etc.); and (4) the effects caused by the industry. See id., Fig. 78 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 23. Dr. Leamer used the model to show that the anti-solicitation agreements had some impact on 2 the Technical Class generally, and to estimate the average or net under-compensation at each 3 Defendant firm during the period in which the anti-solicitation agreements were in effect. See id. 4 Fig. 24; Reply at 33. Dr. Leamer’s model accounted for variations in the effect of the anti- 5 solicitation agreements over time and for variations among different kinds of employees. Id. ¶ 6 146. Dr. Leamer contended that this econometric model of employee compensation could be used 7 in a straightforward formulaic fashion in conjunction with Defendants’ compensation data to 8 calculate damages for employees in the Technical Class. See id. ¶ 148; see also id. Figs. 3 & 4. 9 In the Court’s April 5 Class Certification Order, the Court considered Defendants’ United States District Court For the Northern District of California 10 criticisms of Dr. Leamer’s Conduct Regression, but ultimately found the Conduct Regression 11 model sufficient to satisfy Plaintiffs’ burden for the purpose of Rule 23(b)(3) on the issue of 12 damages. Apr. 5 Class Cert. Order at 43-44. 13 First, the Court rejected Defendants’ criticism that Dr. Leamer failed to conduct a 14 sensitivity analysis—an “exploration of how sensitive [a model’s] conclusions are to a choice of 15 variables.” Leamer Depo. at 351:4-6. Defendants argued that Dr. Leamer should have performed 16 “disaggregated” analyses for each Defendant using only data from that Defendant’s employees. 17 However, in light of the limited compensation data available to Dr. Leamer, Leamer Reply Rep. 18 ¶ 99, including the relatively short length of the data period, 2001-2011, the Court found that 19 aggregation may provide “a [more] robust analysis and yield more reliable and more meaningful 20 statistical results.” Ellis v. Costco Wholesale Corp., 285 F.R.D. 492, 523 (N.D. Cal. 2012). The 21 Court also was not persuaded by Dr. Murphy’s competing disaggregated model—which included 22 42 Defendant-specific variables, only 28 of which related to the effect of the anti-solicitation 23 agreements, see Murphy Rep., App. 9A, because the Court found that Dr. Murphy’s use of that 24 many variables could also “minimize artificially” the effects of the anti-solicitation agreements by 25 spreading those effects across a wider range of variables. See Apr. 5 Class Cert. Order at 39-40 26 (citing Leamer Reply Rep. ¶ 101). 27 28 Second, the Court rejected Defendants’ criticisms of Dr. Leamer’s choice of a benchmark period. Apr. 5 Class Cert. Order at 40; Mot. to Strike at 13. Defendants argued that, if the 79 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 benchmark period was changed from the two years preceding and the two years following the 2 period during which the anti-solicitation agreements were in effect to only the two years following 3 this period, then the model showed net over-compensation rather than under-compensation. See id. 4 However, the Court noted that Defendants failed to explain why the benchmark period should be 5 limited in this way. Apr. 5 Class Cert. Order at 40. Defendants did not show that the pre-conduct 6 data was not comparable to data from the conduct period and therefore should be excluded. 7 Furthermore, the Court expressed concern that, in altering the benchmark periods, Defendants 8 reduced the total amount of data available regarding the non-conduct periods, which could then 9 result in less accurate results. Id. at 40-41. United States District Court For the Northern District of California 10 Third, the Court rejected Defendants’ argument that the Conduct Regression was flawed 11 because Dr. Leamer failed to include a variable to “control for changes in the value of . . . equity 12 compensation [to employees] over time.” Mot. to Strike at 13. Defendants argued that when Dr. 13 Murphy introduced an equity variable, specifically a variable that tracks changes in the S&P 500, 14 the Conduct Regression yielded much smaller under-compensation for the All-Salaried Employee 15 Class and overcompensation for the Technical Class. See Murphy Rep. ¶ 138. The Court was not 16 persuaded by Defendants’ contention because the equity variable Defendants selected—a variable 17 reflecting changes in the S&P 500—tracked variations in the stock price of hundreds of unrelated 18 companies. See Leamer Reply Rep. ¶ 89. Thus, the fact that including this variable significantly 19 altered the results of Dr. Leamer’s analysis did not persuade the Court that Dr. Leamer’s results 20 should be disregarded or that the Conduct Regression is flawed. 21 In addition to the sensitivity issues discussed above, Defendants argued that the Conduct 22 Regression was flawed because Dr. Leamer failed to account for the fact that compensation for 23 employees within the same firm is correlated. See Mot. to Strike at 16; Murphy Rep. ¶ 126. Dr. 24 Murphy opined that, given this correlation, Dr. Leamer should have clustered the standard errors. 25 See Murphy Rep. ¶ 126 (“A generally accepted method to take into account the fact that 26 observations used to estimate a regression contain[] ‘groups’ of observations that are affected by 27 certain common factors (such as those affecting a particular company or present in a single year) is 28 commonly referred to as ‘clustering’ the standard errors”). For the reasons set forth below, the 80 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Court did not find that Dr. Leamer’s failure to cluster the standard errors provided a sufficient basis 2 to reject the Conduct Regression. Apr. 5 Class Cert. Order at 42. 3 Even assuming that Dr. Leamer should have clustered the standard errors, the fact that 4 when the errors were clustered, the Conduct Regression’s results were not statistically significant at 5 the 95 percent confidence level did not persuade the Court that the regression was unpersuasive. 6 See Cook v. Rockwell Int'l Corp., 580 F. Supp. 2d 1071, 1105 (D. Colo. 2006). Dr. Murphy 7 testified that a model’s results need not necessarily be statistically significant to be reliable. 8 Murphy Depo. at 366:14-20. Further, as explained by Dr. Leamer, adjusting the standard errors 9 was only one way of controlling for correlations between employees. See Leamer Reply Rep. United States District Court For the Northern District of California 10 ¶¶ 76, 78, 82, 83. Another approach would be to include variables to explain the commonalities 11 across firms. See id. ¶ 83. The Court noted that Dr. Leamer had already included one such 12 variable, revenue. See id. ¶¶ 82-83. Thus, the Court concluded that Dr. Leamer’s failure to cluster 13 the standard errors did not provide a sufficient basis to conclude that the Conduct Regression failed 14 to provide a reliable methodology for the purposes of class certification. Apr. 5 Class Cert. Order 15 at 42. 16 Defendants now raise additional arguments not raised in their initial opposition to contend 17 that Dr. Leamer’s methodology cannot be used to show that common questions are likely to 18 predominate with respect to damages. Specifically, renewing their argument that the use of a 19 single conduct variable for all Defendants was inappropriate, 21 Defendants argue that Dr. Leamer’s 20 most recent correlation analyses show that total compensation and changes in total compensation at 21 22 23 24 25 26 27 28 21 Defendants also contend that, because the Court “‘encouraged’ Leamer to address . . . whether additional variables were needed,” Suppl. Opp’n at 23, and Dr. Leamer did not add variables to his Conduct Regression analysis, Dr. Leamer’s model should be rejected now. Defendants argue that this is an issue of even greater importance now that Dr. Leamer opines that compensation within a firm is highly correlated. Suppl. Opp’n at 23. Defendants maintain that, “[w]ithout accounting for the correlation, Leamer’s model cannot reliably determine or measure impact or damages.” Suppl. Opp’n at 23-24 (citing Murphy Suppl. Rep. ¶¶ 68-69). Dr. Leamer contends that he “considered whether to add any variables” but is “not aware of any [he] need[s] to add at the present time.” Leamer Suppl. Reply Rep. ¶ 66. Given that the Court’s acceptance of the Conduct Regression as a means of proving and calculating harm to the Class was not contingent upon Dr. Leamer’s addition of more variables to his analysis, the Court does not reject Dr. Leamer’s model on this basis. 81 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Defendants diverged and sometimes moved in opposite directions. Suppl. Opp’n at 24 (citing 2 Leamer Rep., Tables 1 and 2). Defendants assert that, in light of these divergences, the use of a 3 single conduct variable for all Defendants is inappropriate. As set forth in the April 5 Class 4 Certification Order, Dr. Leamer’s decision to use a single variable in his Conduct Regression was 5 understandable because “the available [compensation] data regarding Defendants’ compensation 6 practices [is] ‘limited.’” Apr. 5 Class Cert. Order at 39-40. Dr. Leamer’s approach of aggregating 7 Defendants’ data and calculating a single conduct variable, rather than using the limited data 8 regarding each Defendant to calculate separate conduct variables, “allowed Dr. Leamer to produce 9 a ‘more coherent, more efficient model.’” Id. The Court is not persuaded that Dr. Leamer’s use of United States District Court For the Northern District of California 10 a single conduct variable prevents the Conduct Regression from serving as a reliable method of 11 determining damages. 12 Finally, Defendants argue that Dr. Leamer’s Conduct Regression is overly sensitive to 13 variable choice. Opp’n at 24. The Court disagrees. Dr. Leamer specifically addresses this concern 14 in his reply report and rebuttal supplemental expert report. See Leamer Reply Rep. ¶¶ 85-97; 15 Leamer Suppl. Rebuttal Rep. ¶ 66 (discussing “the lack of sensitivity of [his] findings to inclusion 16 of alternative external control variables such as firm stock prices and to a different level of 17 aggregation.”). The Court is not persuaded that the Conduct Regression model is so sensitive to 18 variable choice that it cannot be used to satisfy Plaintiffs’ burden at class certification. 19 Because Dr. Leamer’s model is supported by the economic literature (including Dr. 20 Shaw’s), is statistically robust (i.e., insensitive to alternative control variables), and is buttressed by 21 Dr. Leamer’s subsequent analysis, the Court finds that Dr. Leamer’s model is capable of 22 calculating classwide damages. Suppl. Reply at 15. 23 24 4. Conclusion Regarding Predominance This Court’s rigorous analysis shows that common issues are likely to predominate over 25 individual issues. Importantly, this Court’s analysis of predominance involves a “qualitative 26 assessment.” See Butler, 727 F.3d at 801. This qualitative assessment includes some analysis into 27 how this case, should it proceed to trial, would actually be litigated. See In re New Motors, 522 28 F.3d at 20 (“Under the predominance inquiry, a district court must formulate some predication as 82 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 to how specific issues will play out in order to determine whether common or individual issues 2 predominate in a given case.” (internal quotation marks omitted)). 3 As such, this Court notes that there is no dispute that antitrust violation can be shown using 4 exclusively evidence that is common to the entire Technical Class for the reasons discussed above. 5 The Court further finds that antitrust violation is likely to be a central, disputed issue at summary 6 judgment and at trial. Defendants have made quite clear—both through their motions to dismiss 7 and their initial opposition to the motion for class certification—that Defendants will seek to 8 contest the issue of antitrust violation by contending that their agreements had no anti-competitive 9 effect on the market. Specifically, Defendants have stated that they intend to “demonstrate that the United States District Court For the Northern District of California 10 agreements should be evaluated under the rule of reason, were reasonable and lawful under that 11 standard, and could not have conceivably had any adverse effect on compensation in any relevant 12 labor market.” Opp’n at 5, n.1. The Supreme Court has stated that the rule of reason analysis 13 entails “significant costs” and is often “extensive and complex.” See Arizona v. Maricopa Cnty. 14 Med. Soc’y, 457 U.S. 332, 343 (1982). 15 Given the considerable, compelling common proof Plaintiffs have submitted regarding 16 Defendants’ alleged antitrust violation, as well as the parties’ actions indicating that they will 17 vigorously litigate the question of whether Defendants engaged in an antitrust violation, this 18 question is likely to be central to this litigation. At the very least, this aspect of the trial should not 19 be understated. See In re Static Random Access Memory (SRAM) Antitrust Litig., 264 F.R.D. 603, 20 611 (N.D. Cal. 2009) (“Plaintiffs need not show that there will be common proof on each element 21 of the claim. ‘In price-fixing cases, courts repeatedly have held that the existence of the conspiracy 22 is the predominant issue and warrants certification even where significant individual issues are 23 present.’” (quoting Thomas & Thomas Rodmakers, Inc. v. Newport Adhesives & Composites, Inc., 24 209 F.R.D. 159, 167 (C.D. Cal. 2002))); see 6 Newberg on Class Actions § 18.25 (4th ed. 2002) 25 (“[C]ommon liability issues such as conspiracy or monopolization have, almost invariably, been 26 held to predominate over individual issues.”); 7AA Charles Alan Wright, Arthur Miller & Mary 27 Kay Kane, Federal Practice and Procedure, § 1781 (3d ed. 2005) (“whether a conspiracy existed 28 is a common question that is thought to predominate over other issues in the case”); cf. Cordes & 83 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Co. Fin. Servs. v. A.G. Edwards & Sons, Inc., 502 F.3d 91, 108 (2d Cir. 2007) (“Even if the district 2 court concludes that the issue of injury-in-fact presents individual questions, however, it does not 3 necessarily follow that they predominate over common ones and that class action treatment is 4 therefore unwarranted.”). As a result, the voluminous classwide proof of antitrust violation weighs 5 in favor of a finding that common questions predominate. 6 In addition to concluding that common questions will predominate with respect to the central element of antitrust violation, the Court, having conducted a rigorous analysis, also finds 8 that common questions will predominate over individual questions with respect to impact. The 9 extensive documentary evidence suggests that Defendants maintained a formal wage structure and 10 United States District Court For the Northern District of California 7 valued internal equity. This suggests that the anti-solicitation agreements had a structural impact 11 on class members’ compensation. Furthermore, the Court, having taken a hard look at the experts’ 12 reports, concludes that Plaintiffs have presented a methodology that supports a finding that the 13 evidence common to the class will be utilized in demonstrating impact. Finally, the Court finds 14 that Plaintiffs have set forth a methodology for calculating damages on a classwide basis. Thus, 15 following a rigorous analysis, the Court finds that Plaintiffs have satisfied Rule 23(b)(3)’s 16 predominance requirement with respect to all three elements—antitrust violation, impact, and 17 damages. 18 C. 19 Rule 23(b)(3) also tests whether “a class action is superior to other available methods for Rule 23(b)(3): Superiority 20 fairly and efficiently adjudicating the controversy.” Fed. R. Civ. P. 23(b)(3). Under Rule 23(b)(3), 21 the Court must consider four non-exclusive factors in evaluating whether a class action is a 22 superior method of adjudicating plaintiffs’ claims: (1) the interest of each class member in 23 individually controlling the prosecution or defense of separate actions; (2) the extent and nature of 24 any litigation concerning the controversy already commenced by or against the class; (3) the 25 desirability of concentrating the litigation of the claims in the particular forum; and (4) the 26 difficulties likely to be encountered in the management of a class action. Zinser, 253 F.3d at 1190- 27 92. 28 84 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 Plaintiffs state that “[c]lass treatment is by definition superior to thousands of individual 2 claims in an antitrust case where common issues of liability and impact predominate.” Class Cert. 3 Mot. at 23 (citing In re TFT-LCD (Flat Panel) Antitrust Litig., 267 F.R.D. at 314 (“[I]f common 4 questions are found to predominate in an antitrust action . . . the superiority prerequisite of Rule 5 23(b)(3) is satisfied.”)). Plaintiffs contend that Class members’ individual damages, even after 6 mandatory trebling, are insufficiently large to warrant individual litigation. In re TFT-LCD (Flat 7 Panel) Antitrust Litig., 267 F.R.D. at 314-15 (noting that, in antitrust cases, individual damages 8 “are likely to be too small to justify litigation, but a class action would offer those with small 9 claims the opportunity for meaningful redress”). United States District Court For the Northern District of California 10 Plaintiffs further contend that “[c]lass treatment will also be more manageable and efficient 11 than hundreds or thousands of individual actions litigating the same issues with nearly identical 12 proof. . . . Either defendants colluded or they did not; either their conspiracy artificially suppressed 13 their compensation structure or it did not. Any trial here will focus on these questions and the 14 same evidence, whether it involves a single employee or the Class as a whole.” Class Cert. Mot. at 15 23-24. 16 Defendants, however, argue that “[t]he ‘numerous and substantial separate issues’ each 17 class member would have to litigate to ‘establish his or her right to recover individually’ means 18 that ‘class action treatment is not the ‘superior’ method of adjudication.” Opp’n at 25 (citing 19 Zinser, 253 F.3d at 1192). During the hearing on Plaintiffs’ Supplemental Motion for Class 20 Certification, Defendants proposed instead holding multiple “bellwether” trials to accommodate 21 the groups of people impacted. Aug. 8 Tr. at 31-32. In response, counsel for Plaintiffs argued that 22 such an approach would not conserve resources because, “in every single case, the proof of impact 23 would be the opinion that this conduct . . . affected the pay structure of the entire company.” Id. at 24 67-68. The Court agrees. Given that Plaintiffs’ case rises and falls with their common evidence, 25 the Court does not find that conducting numerous bellwether trials, which will effectively be trials 26 27 28 85 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION 1 with Plaintiffs representing subclasses of employees, will ease case management. 22 In fact, this 2 would merely multiply the number of trials with the same issues and evidence. 3 Thus, the Court finds that the Technical Class members’ interests weigh in favor of having 4 this case litigated as a class action. In addition, the nature of Defendants’ alleged overarching 5 conspiracy and the desirability of concentrating the litigation in one proceeding weigh heavily in 6 favor of finding that class treatment is superior to other methods of adjudication of the controversy. 7 See Zinser, 253 F.3d at 1190-92. The Court also finds that questions regarding manageability 8 weigh in favor of finding class treatment superior to other methods of adjudication. Thus, the 9 Court finds that Plaintiffs have satisfied the superiority requirement. United States District Court For the Northern District of California 10 V. CONCLUSION 11 For the reasons set forth above, the Court finds that Plaintiffs have satisfied all of the 12 requirements for Rule 23(a) of the Federal Rules of Civil Procedure, as well as the requirements of 13 Rule 23(b)(3). Accordingly, the Court GRANTS Plaintiffs’ Motion for Class Certification as to the 14 Technical Class and appoints named Plaintiffs as Class Representatives. 15 IT IS SO ORDERED. 16 Dated: October 24, 2013 17 _________________________________ LUCY H. KOH United States District Judge 18 19 20 21 22 22 23 24 25 26 27 28 Defendants further suggest that certifying the Technical Class would violate the Rules Enabling Act because it would prevent Defendants from asserting statutory defenses to which they are entitled. In making this argument, Defendants rely on a section of Dukes in which the Supreme Court held that Rule 23(b)(2), which provides for the certification of a class seeking injunctive relief and not damages, was not an appropriate vehicle for certifying a class of discrimination plaintiffs who sought backpay under Title VII. Opp’n at 25 (citing Dukes, 131 S. Ct. at 2561). The Supreme Court held that the putative class could not be certified in part because the defendant would not be able to litigate Title VII defenses that it may have had against individual class members. Dukes, 131 S. Ct. at 2561. As a result, class certification would have expanded plaintiffs’ substantive rights under Title VII. This Dukes holding applied to a class under Rule 23(b)(2), which provides for only injunctive relief and not for damages; thus, this holding is inapplicable to the instant case. Further, Defendants here have not identified any statutory defenses that Defendants would have against particular class members, nor have Defendants contended that certification would expand Plaintiffs’ rights under the antitrust laws. 86 Case No.: 11-CV-02509-LHK ORDER GRANTING PLAINTIFFS’ SUPPLEMENTAL MOTION FOR CLASS CERTIFICATION