Nitsch v. Dreamworks Animation SKG Inc. et al

Filing 289

ORDER GRANTING-IN-PART AND DENYING-IN-PART 203 PLAINTIFFS' MOTION FOR CLASS CERTIFICATION. Signed by Judge Lucy H. Koh on May 25, 2016. (lhklc1, COURT STAFF) (Filed on 5/25/2016)

1
2
3
4
5
6
7
8

UNITED STATES DISTRICT COURT

9

NORTHERN DISTRICT OF CALIFORNIA

10
SAN JOSE DIVISION

United States District Court
Northern District of California

11
12

ROBERT A. NITSCH, et al.,

Case No. 14-CV-04062-LHK

Plaintiffs,

13

ORDER GRANTING-IN-PART AND
DENYING-IN-PART PLAINTIFFS’
MOTION FOR CLASS
CERTIFICATION

v.

14
15

DREAMWORKS ANIMATION SKG INC.,
et al.,
Defendants.

16

Re: Dkt. No. 203

17
Plaintiffs Robert A. Nitsch, Jr., Georgia Cano, and David Wentworth (collectively,

18
19

“Plaintiffs”), individually and on behalf of a class of all those similarly situated, allege antitrust

20

claims against their former employers, Blue Sky Studios, Inc.; DreamWorks Animation SKG,

21

Inc.; Two Pic MC LLC, formerly known as ImageMovers Digital LLC; Lucasfilm Ltd., LLC;

22

Pixar; Sony Pictures Animation Inc. and Sony Pictures Imageworks, Inc., and The Walt Disney

23

Company (collectively, “Defendants”).1 Plaintiffs allege that Defendants conspired to suppress,

24
25
26
27
28

1

Plaintiffs filed a motion for preliminary approval of class action settlement with Blue Sky
Studios, Inc. ECF Nos. 249, 282. Blue Sky Studios, Inc. thus did not join Defendants’ opposition
to class certification. Plaintiffs also filed a motion for preliminary approval of class action
settlement with Sony Pictures Animation Inc. and Sony Pictures Imageworks, Inc. after
Defendants’ opposition to class certification was filed. See ECF No. 273.
1
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

and actually did suppress, employee compensation to artificially low levels by agreeing not to

2

solicit each other’s employees and by exchanging employee compensation information in

3

violation of Section 1 of the Sherman Act, 15 U.S.C. § 1; California’s Cartwright Act, Cal. Bus. &

4

Prof. Code § 16720; and California’s Unfair Competition Law (“UCL”), Cal. Bus. & Prof. Code

5

§§ 17200 et seq.
Before the Court is Plaintiffs’ Motion for Class Certification. (“Motion” or “Mot.”), ECF

6

No. 203. The Court held a hearing on the Motion on May 6, 2016. See ECF No. 276 (minutes);

8

ECF No. 279 (“May 6 Trans.”). Having considered the parties’ submissions, oral argument, the

9

relevant law, and the record in this case, the Court GRANTS IN PART and DENIES IN PART

10

Plaintiffs’ Motion for Class Certification and CERTIFIES AS MODIFIED Plaintiffs’ proposed

11
United States District Court
Northern District of California

7

class.

12

I. BACKGROUND
This is a consolidated class action brought by former employees alleging antitrust claims

13
14

against their former employers, who are various animation and visual effects studios with

15

principal places of business in California. Second Amended Complaint (“SAC”), ECF No. 121.2

16

Plaintiffs contend that Defendants engaged in a conspiracy to fix and suppress employee

17

compensation and to restrict employee mobility.

18

A. Factual Background

19

1. The Parties

20

Defendants include the following animation and visual effects studios: Blue Sky Studios,

21

Inc. (“Blue Sky”), a Delaware corporation with its principal place of business in Greenwich,

22

Connecticut; DreamWorks Animation SKG, Inc. (“DreamWorks”), a Delaware corporation with

23

its principal place of business in Glendale, California; Two Pic MC LLC, formerly known as

24

ImageMovers Digital LLC (“ImageMovers Digital”), a Delaware corporation with its principal

25
26
27
28

2

Defendant Blue Sky Studios, Inc. has its principal place of business in Greenwich, Connecticut,
but Plaintiffs allege that it is owned by Twentieth Century Fox Film Corporation, which has its
principal place of business in Los Angeles, California. SAC ¶ 22.
2
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

place of business in Burbank, California; Lucasfilm Ltd., LLC (“Lucasfilm”), a California

2

corporation with its principal place of business in San Francisco, California;3 Pixar, a California

3

corporation with its principal place of business in Emeryville, California;4 Sony Pictures

4

Animation, Inc. and Sony Pictures Imageworks, Inc. (collectively, “the Sony Defendants”),

5

California corporations with their principal places of business in Culver City, California; and The

6

Walt Disney Company (“Disney”), a Delaware corporation with its principal place of business in

7

Burbank, California.5 SAC ¶¶ 22–29.
Plaintiffs are artists and engineers who were previously employed by four of the named

8

Defendants. Id. ¶¶ 19–21. Nitsch worked for Sony Picture Imageworks in 2004 and DreamWorks

10

from 2007 to 2011. Id. ¶ 19. Cano worked for Walt Disney Feature Animation from 2004 to 2005,

11
United States District Court
Northern District of California

9

ImageMovers Digital in 2010, and at various other visual effects and animation studios. Id. ¶ 20.

12

Wentworth worked at ImageMovers Digital from 2007 to 2010. Id. ¶ 21. Nitsch is a resident of

13

Massachusetts, and Cano and Wentworth are residents of California. Id. ¶¶ 19–21.

14

2. In re High-Tech Employee Litigation and the Department of Justice investigation

15

There is significant factual overlap between Plaintiffs’ allegations and the related action In

16

re High-Tech Employee Litigation, No. 11-CV-02509-LHK (“High-Tech”), as well as the civil

17

complaints filed by the Department of Justice (“DOJ”) against Pixar, Lucasfilm, and several

18

Silicon Valley technology companies. As the factual history of the related High-Tech action and

19

the DOJ complaints is relevant to the substance of Plaintiffs’ Motion and Defendants’ Opposition,

20

the Court briefly summarizes the factual background of those prior proceedings below.
From 2009 to 2010, the Antitrust Division of the DOJ investigated the employment and

21
22

recruitment practices of various Silicon Valley technology companies, including Adobe Systems,

23
24
25
26
27
28

3

The parties’ papers also refer at points to Industrial Light & Magic (“ILM”). Plaintiffs aver that
ILM is a division of Lucasfilm.
4
According to Plaintiffs, ILM, Lucasfilm, and Pixar have been owned by Defendant The Walt
Disney Company since 2006 (Pixar) and 2012 (ILM and Lucasfilm). SAC ¶¶ 25–26.
5
Disney also “oversees the operations of” Walt Disney Animation Studios, formerly known as
Walt Disney Feature Animation. SAC ¶ 29.
3
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

Inc., Apple, Inc., Google, Inc., Intel Corp., and Intuit, Inc. See In re High-Tech Empl. Litig., 856 F.

2

Supp. 2d 1103, 1109 (N.D. Cal. 2012). In September of 2010, the DOJ then filed civil complaints

3

in the D.C. District Court against the above-mentioned technology companies, in addition to Pixar

4

and Lucasfilm. Id. The DOJ filed its complaint against Adobe, Apple, Google, Intel, Intuit, and

5

Pixar on September 24, 2010. Id. On December 21, 2010, the DOJ filed a separate complaint

6

against Lucasfilm. Id. & n.1. The defendants, including Pixar and Lucasfilm, stipulated to

7

proposed final judgments in which the defendants agreed that the DOJ’s complaints had stated

8

claims under federal antitrust law and agreed to be “enjoined from attempting to enter into,

9

maintaining or enforcing any agreement with any other person or in any way refrain from . . .
soliciting, cold calling, recruiting, or otherwise competing for employees of the other person.” Id.

11
United States District Court
Northern District of California

10

at 1109–10 (quoting Adobe Proposed Final Judgment at 5). The D.C. District Court entered the

12

stipulated proposed final judgments in March and June 2011. Id. at 1110.

13

The High-Tech plaintiffs filed five separate state court actions between May and July 2011.

14

Following removal, transfer to San Jose to the undersigned judge, and consolidation, the High-

15

Tech plaintiffs filed a consolidated amended complaint on September 13, 2011. High-Tech, 856 F.

16

Supp. 2d at 1112–13. In their complaint, the High-Tech plaintiffs alleged antitrust claims against

17

their employers and alleged that the defendants had conspired “to fix and suppress employee

18

compensation and to restrict employee mobility.” Id. at 1108. More specifically, the High-Tech

19

plaintiffs alleged a conspiracy comprised of “an interconnected web of express bilateral

20

agreements.” Id. at 1110. One agreement, the “Do Not Cold Call” agreement, involved one

21

company placing the names of the other company’s employees on a “Do Not Cold Call” list and

22

instructing its recruiters not to cold call the employees of the other company. Id. In addition to the

23

“Do Not Cold Call” agreements, the High-Tech plaintiffs also alleged that Pixar and Lucasfilm,

24

defendants in both High-Tech and the instant action, entered into express, written agreements (1)

25

to not cold call each other’s employees, (2) to notify the other company whenever making an offer

26

to an employee of the other company, and (3) not to engage in “bidding wars.” Id. at 1111.

27
28

4
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1
2
3
4
5
6
7
8
9
10

United States District Court
Northern District of California

11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

3. Alleged Conspiracy in the Instant Action
Although Plaintiffs’ factual allegations have largely been discussed in the Court’s prior
rulings in this case, a brief summary of the allegations is provided below. An in-depth analysis of
the factual record submitted in support of the motion for class certification will be discussed in
Section IV.B, infra.
Here, Plaintiffs allege that Defendants conspired to suppress compensation in two ways.
First, Defendants allegedly entered into a scheme not to actively solicit each other’s employees.
SAC ¶ 42. Second, Defendants allegedly engaged in “collusive discussions in which they
exchanged competitively sensitive compensation information and agreed upon compensation
ranges,” which would artificially limit compensation offered to Defendants’ current and
prospective employees. Id.
a. Anti-Solicitation Scheme
According to Plaintiffs, as in High-Tech, “Defendants agreed not to contact their
coconspirators’ employees to inform them of available positions unless that individual employee
had applied for a job opening on his or her own initiative.” Id. ¶ 43. This solicitation, also known
as “cold calling,” is “a key competitive tool in a properly functioning labor market, especially for
skilled labor.” Id. ¶ 44. Plaintiffs aver that employees of competitor studios represent “one of the
main pools of potential hires,” and that employees of competitor studios that are not actively
searching for new employment are “more likely to be among the most sought after employees.” Id.
Hiring an employee from a competitor studio “can save costs and avoid risks.” Id. Absent active
solicitation, these employees are also difficult to reach. Id. Defendants’ anti-solicitation scheme
also allegedly included “notifying each other when an employee of one Defendant applied for a
position with another Defendant, and agreeing to limit counteroffers in such situations.” Id. ¶ 45.
Moreover, Defendants allegedly “often refrained from hiring other Defendants’ employees at all
without the permission of the current employer,” and would sometimes decline to make offers of
employment to an unemployed prospective hire if that individual had an outstanding offer from
another Defendant. Id. ¶ 46.
5
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

According to Plaintiffs, “the roots of the conspiracy reach back to the mid-1980s,” when

1
2

George Lucas, the former Lucasfilm Chairman of the Board and CEO, sold Lucasfilm’s

3

“computer division” to Steve Jobs, who had recently left Apple. SAC ¶ 47. Jobs named his new

4

company Pixar. Id. Pixar’s President, Ed Catmull, Lucas, and “other senior executives,

5

subsequently reached an agreement to restrain their competition for the skilled labor that worked

6

for the two companies.” Id. Pixar and Lucasfilm allegedly agreed to the following terms: (1) not to

7

cold call each other’s employees; (2) to notify each other when making an offer to the other

8

company’s employee; and (3) that any offer by the other company would be “final,” i.e., neither

9

Pixar nor Lucasfilm would engage in counteroffers. Id. ¶¶ 47–51 (citing internal Pixar email sent

10

on January 16, 2006).
Plaintiffs further allege that while the conspiracy originated with Pixar and Lucasfilm,

United States District Court
Northern District of California

11
12

Catmull brought additional studios into the fold. Id. ¶ 52. According to Plaintiffs, Blue Sky,

13

DreamWorks, ImageMovers Digital,6 the Sony Defendants, and Walt Disney Animation Studios

14

all became part of the anti-solicitation conspiracy during the mid-2000s and agreed not to directly

15

recruit each other’s employees. Id. ¶¶ 53–79.

16

b. Compensation Ranges
In addition to the anti-solicitation scheme, Plaintiffs further allege that Defendants

17
18

“directly communicated and met regularly to discuss and agree upon compensation ranges.” Id.

19

¶ 86 (citing March 28, 2007 email from Pixar’s Vice President of Human Resources, Lori

20

McAdams). According to Plaintiffs, Defendants met at least once a year in California at meetings

21

organized by the Croner Company, a third party that apparently collects industry-specific salary

22

information.
At the official meetings, Defendants “set the parameters of a compensation survey” among

23
24

participating animation studios. Id. ¶ 87. The survey, known as the “Croner Survey,” “provides

25
26
27
28

6

Plaintiffs dismissed a separate Defendant, ImageMovers LLC, without prejudice pursuant to a
tolling agreement on January 14, 2015. ECF No. 83. The dismissal of ImageMovers LLC did not
affect ImageMovers Digital.
6
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

wage and salary ranges for the studios’ technical or artistic positions, broken down by position and

2

experience level.” Id. The purpose of the meetings and the Croner Survey was for Defendants to

3

“confirm or adjust [their] salary ranges.” Id. Senior human resources and recruiting personnel

4

from DreamWorks, Pixar, Lucasfilm, Disney, ImageMovers, the Sony Defendants, Blue Sky, and

5

Digital Domain7 attended these survey meetings, in addition to other studios. Id. ¶ 88. Defendants

6

also requested “custom cuts” of the survey information collected by the Croner Company. Id. ¶ 93.

7

These custom cuts allegedly included compensation data limited to a “special subset of Croner

8

Survey participants, namely Blue Sky, DreamWorks, Lucasfilm, Sony, and Pixar.” Id.
Plaintiffs aver that Defendants used the Croner meetings to “go further than their matching

9

of job positions across companies; they discussed, agreed upon and set wage and salary ranges

11
United States District Court
Northern District of California

10

during meals, drinks and other social gatherings that they held outside of the official Croner

12

meetings.” Id. ¶ 89. Defendants’ human resources and recruiting personnel also allegedly held

13

“side” meetings at the Siggraph conference, a major visual effects industry conference, which

14

senior personnel from Blue Sky, Pixar, DreamWorks, Lucasfilm, and Sony Picture ImageWorks

15

attended. Id. ¶ 91.
Plaintiffs further allege that Defendants regularly emailed each other with specific salary

16
17

ranges. Plaintiffs contend that Defendants’ “collusive compensation setting was not limited to

18

wages and salaries, but extended to other benefits and terms of employment.” Id. ¶ 97.
According to Plaintiffs, Defendants’ communications regarding salary ranges were not

19
20

limited to bilateral “one-off” exchanges, but rather Defendants would “openly email[] each other

21

in large groups with competitively sensitive confidential current and future compensation

22

information.” Id. ¶ 102.
Defendants’ human resources and recruiting personnel also allegedly regularly

23
24

communicated via telephone. Id. ¶ 113. As Plaintiffs describe it, the Croner survey meetings, side

25
26
27
28

7

Plaintiff also dismissed former defendant Digital Domain 3.0 without prejudice pursuant to a
tolling agreement. See SAC at 19 n.3.
7
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

meetings, emails, and telephone calls “provided the means and opportunities for Defendants to

2

collude and to implement and enforce the conspiracy to suppress workers’ compensation.” Id.

3

¶ 114. According to Plaintiffs, executives such as Pixar’s Lori McAdams, “knew that such

4

conversations were inappropriate.” Id. ¶ 5, 110–11.

5

Plaintiffs further allege that while press reports in 2009 noted that the DOJ was

6

investigating anti-solicitation agreements among high-tech companies, including Google and

7

Apple, there was no indication that the DOJ was also investigating Pixar, Lucasfilm, or any other

8

animation company. Id. ¶ 119. Plaintiffs aver that September 17, 2010 marked the first news story

9

naming Pixar as a company under investigation, but that there was no public disclosure that any
other Defendant in the instant action was part of the conspiracy. Id. ¶¶ 119, 184. According to

11
United States District Court
Northern District of California

10

Plaintiffs, Lucasfilm was implicated in the Pixar investigation in December 2010, but until this

12

Court unsealed certain filings in the High-Tech case, there was no public information that the other

13

Defendants in this action had engaged in similar conduct. Id. Plaintiffs also cite the absence of

14

news coverage as proof that Plaintiffs had no way of discovering the conspiracy, as even industry

15

journalists were “unable to discover and explore the conspiracy.” Id. ¶ 186.

16
17

c. Fraudulent Concealment
In their SAC, Plaintiffs allege that Defendants fraudulently concealed the conspiracy and

18

therefore prevented the Plaintiffs from filing their claims on time. Plaintiffs allege that Defendants

19

(1) took affirmative steps to keep their conspiracy a secret; (2) affirmatively misled class members

20

by claiming that compensation and recruiting was determined by factors other than the alleged

21

conspiracy; and (3) took affirmative steps to mislead class members about the conspiracy during

22

the High-Tech litigation.

23
24

i.

Affirmative steps to keep their conspiracy a secret

Plaintiffs aver that Defendants carried out their conspiracy “in a manner specifically

25

designed to avoid detection.” Id. ¶ 136. Plaintiffs claim that Defendants limited meetings to top

26

executives and human resources employees, “avoided discussing the agreements in written

27

documents,” and “avoid[ed] unnecessarily creating evidence that might alert Plaintiffs . . . to the

28

8
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

conspiracy’s existence.” Id. Defendants also allegedly “opted for in-person meetings” when

2

possible, instead of communication via email. Id. ¶ 143.

3
4

ii.

Pretextual statements regarding compensation and recruiting

Plaintiffs further allege that Defendants “routinely provided pretextual, incomplete or

5

materially false and misleading explanations for compensation decisions and recruiting and

6

retention practices.” Id. ¶ 145. For example, Defendants’ recruiting websites and brochures state

7

that they provide “competitive salar[ies]” or “competitive compensation,” which, according to

8

Plaintiffs, “hid[es] . . . the fact that the competition that normally exists among rival employers

9

had been restrained by collusion.” Id. ¶ 146.

10

Plaintiffs also allege that Defendants’ own codes of conduct contained statements that

United States District Court
Northern District of California

11

“misrepresented the truth about the conspiracy,” id. ¶ 151, that Defendants’ public filings with the

12

Securities and Exchange Commission (“SEC”) were misleading, id. ¶ 159, and that Defendants

13

made pretextual and misleading statements regarding recruiting and retention, id. ¶ 166.

14

According to Plaintiffs, Defendants “misrepresented the steps they took to retain employees or

15

attract talent,” including Lucasfilm’s statements on its recruiting website that it was “continually

16

on the lookout for talent,” despite having agreed not to solicit or cold call employees of its

17

competitors. Id. ¶ 166.

18

iii.

Misleading statements during the High-Tech litigation

19

Plaintiffs further allege that Defendants Pixar and Lucasfilm made affirmative

20

misrepresentations to Plaintiffs and class members at the outset of the High-Tech litigation. Id.

21

¶ 171. According to Plaintiffs, Pixar and Lucasfilm misleadingly denied that the anti-solicitation

22

agreement “was created with the intent and effect of eliminating ‘bidding wars,’” whereby a

23

prospective employee could increase her total compensation by leveraging offers from either

24

Defendant. Id. ¶ 172. Plaintiffs contend that this denial “affirmatively deceiv[ed]” Plaintiffs and

25

class members “as to the purpose of the agreement.” Id. Moreover, Plaintiffs point to Pixar’s and

26

Lucasfilm’s apparently misleading statements regarding the scope of their agreement. Id. ¶ 173. In

27

addition to these representations, Plaintiffs also contend that Pixar and Lucasfilm denied under

28

9
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

oath that Pixar or Lucasfilm had “conspired with any entities beyond those named by the DOJ.”

2

Id. ¶ 174.

3

Finally, Plaintiffs allege that Pixar and Lucasfilm “took steps to conceal documents

4

revealing the true scope of their conspiracy by designating all depositions, declarations and most

5

documents in High-Tech ‘attorneys’ eyes only,’” thus preventing class members from examining

6

these documents until this Court unsealed the documents in 2013. Id. ¶ 179. According to

7

Plaintiffs, Pixar and Lucasfilm “made the affirmative decision” to file such documents under seal

8

or sought sealing even “when such requests were unjustified,” and that the true purpose of Pixar’s

9

and Lucasfilm’s actions was to “conceal the documents.” Id. ¶ 180.
4. Claims

11
United States District Court
Northern District of California

10

Plaintiffs’ SAC asserts three claims for relief under the following statutes: (1) Section 1 of

12

the Sherman Act, 15 U.S.C. § 1; (2) California’s Cartwright Act, Cal. Bus. & Prof. Code § 16720;

13

and (3) California’s Unfair Competition Law (“UCL”), Cal. Bus. & Prof. Code §§ 17200 et seq.

14

SAC ¶¶ 205–18. Plaintiffs seek treble damages, pre- and post-judgment interest, attorney’s fees

15

and expenses, and a permanent injunction. Id. ¶ 219.

16
17

B. Procedural Background
In light of the relationship between the instant case and the High-Tech case, the Court

18

summarizes the relevant procedural history of the High-Tech case in addition to the procedural

19

history of the instant case.

20

1. High-Tech Procedural Background

21

The High-Tech defendants removed the first state court action on May 23, 2011. No. 11-

22

2509, ECF No. 1. The last state-court action in the High-Tech litigation was removed on July 19,

23

2011. No. 11-2509, ECF No. 41. After reassignment of the cases to the undersigned judge, a First

24

Consolidated Amended Complaint was filed on September 13, 2011. No. 11-2509, ECF No. 65.

25

On April 18, 2012, the Court granted in part and denied in part the High-Tech defendants’ motions

26

to dismiss. No. 11-2509, ECF No. 119. On April 5, 2013, the Court granted in part and denied in

27

part the High-Tech plaintiffs’ motion for class certification with leave to amend. No. 11-2509,

28

10
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

ECF No. 382. The High-Tech plaintiffs filed a supplemental motion for class certification on May

2

10, 2013, which the Court granted on October 24, 2013. No. 11-2509, ECF No. 531. On

3

November 7, 2013, the High-Tech defendants filed a Rule 23(f) petition before the Ninth Circuit,

4

requesting permission to appeal this Court’s October 24, 2013 class certification order. In re High-

5

Tech Empl. Antitrust Litig., No. 13-80223, ECF No. 1 (9th Cir. Nov. 7, 2013). The Ninth Circuit

6

denied the defendants’ petition on January 14, 2014. Id., ECF No. 18.

7

In the interim, three of the High-Tech defendants—Intuit, Lucasfilm, and Pixar—reached a

8

settlement with the High-Tech plaintiffs. On October 30, 2013, the Court granted preliminary

9

approval as to that settlement. No. 11-2509, ECF No. 540. The Court granted final approval on
May 16, 2014. No. 11-2509, ECF No. 915. The Court entered a final judgment with regard to

11
United States District Court
Northern District of California

10

Lucasfilm, Pixar, and Intuit on June 9, 2014 and an amended final judgment on June 20, 2014. No.

12

11-2509, ECF Nos. 936, 947.

13

The remaining High-Tech defendants—Adobe, Apple, Google, and Intel—filed individual

14

motions for summary judgment, a joint motion for summary judgment, and a joint motion to strike

15

certain expert testimony on January 9, 2014. No. 11-2509, ECF Nos. 554 (Intel), 556 and 557

16

(joint motions), 560 (Adobe), 561 (Apple), 564 (Google). The Court denied the High-Tech

17

defendants’ individual motions for summary judgment on March 28, 2014. No. 11-2509, ECF No.

18

771. On April 4, 2014, the Court granted in part and denied in part the High-Tech defendants’

19

joint motion to strike, and denied the defendants’ joint motion for summary judgment. No. 11-

20

2509, ECF No. 788.

21

On May 22, 2014, the High-Tech plaintiffs filed a motion for preliminary approval of class

22

action settlement as to the remaining defendants. No. 11-2509, ECF No. 920. On August 8, 2014,

23

the Court denied the High-Tech plaintiffs’ motion for preliminary approval and concluded that the

24

proposed settlement, which included a settlement fund of $324.5 million, did not fall “within the

25

range of reasonableness.” No. 11-2509, ECF No. 974 at 30. On September 4, 2014, the High-Tech

26

defendants filed a petition for a writ of mandamus with the Ninth Circuit. In re Adobe Sys., Inc., et

27

al., No. 14-72745, ECF No. 1 (9th Cir. Sept. 4, 2014). On September 22, 2014, the Ninth Circuit

28

11
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

found that the petition “raises issues that warrant a response,” and ordered briefing. Id., ECF No.

2

2. On January 13, 2015, the High-Tech defendants filed correspondence with the Ninth Circuit

3

referring to a new proposed settlement agreement. Id., ECF No. 21. On January 30, 2015, the

4

defendants filed an unopposed motion to dismiss the petition, which the Ninth Circuit granted on

5

February 2, 2015. Id., ECF Nos. 23, 24.

6

On January 15, 2015, the High-Tech plaintiffs filed a motion for preliminary approval of

7

class action settlement as to the remaining defendants. No. 11-2509, ECF No. 1032. In this second

8

proposed class action settlement, the parties had reached a settlement amount exceeding the

9

previously rejected settlement by approximately $90.5 million. Id. at 1. Following a fairness
hearing on March 2, 2015, the Court granted preliminary approval to the January 2015 settlement

11
United States District Court
Northern District of California

10

agreement on March 3, 2015. No. 11-2509, ECF Nos. 1051, 1054. The Court held a final approval

12

hearing on July 9, 2015. No. 11-2509, ECF No. 1096. On September 2, 2015, the Court granted

13

final approval of the class action settlement and entered final judgment with regard to Adobe,

14

Apple, Google, and Intuit. No. 11-2509, ECF Nos. 1111, 1113.

15

2. Procedural Background in the Instant Action

16

Plaintiff Nitsch filed the first complaint against all Defendants except Blue Sky on

17

September 8, 2014. ECF No. 1. The Court related Nitsch’s action to In re High-Tech Employee

18

Antitrust Litigation, No. 11-2509, on September 23, 2014. ECF No. 12. Plaintiff Cano filed the

19

second complaint against all Defendants on September 17, 2014, which the Court related to High-

20

Tech on October 7, 2014. See Case No. 14-4203, ECF Nos. 1, 9. Plaintiff Wentworth filed the

21

third complaint against all Defendants except Blue Sky on October 2, 2014, which the Court

22

related to High-Tech on October 28, 2014. See Case No. 14-4422, ECF Nos. 1, 26. On November

23

5, 2014, the Court granted Plaintiffs’ motion to consolidate the above-mentioned three cases into a

24

single action, In re Animation Workers Antitrust Litigation. See Case No. 14-4062, ECF No. 38.

25

Pursuant to the Court’s November 6, 2014 case management order, ECF No. 39, Plaintiffs

26

filed their first consolidated amended complaint on December 2, 2014. ECF No. 63. On January 9,

27

2015, Defendants filed a joint motion to dismiss. ECF Nos. 75. Plaintiffs filed a timely opposition,

28

12
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

ECF No. 97, and Defendants replied, ECF No. 100. On April 3, 2015, the Court granted

2

Defendants’ motion to dismiss. In re Animation Workers Antitrust Litig., 87 F. Supp. 3d 1195

3

(N.D. Cal. Apr. 3, 2015). The Court found that Plaintiffs’ claims were time barred under the

4

statute of limitations, and that Plaintiffs had failed to adequately plead a “continuing violations”

5

theory or a “fraudulent concealment” theory to toll the statute of limitations. See id. at 1212,

6

1217–18. The dismissal was without prejudice, as the Court determined that Plaintiffs might be

7

able to allege sufficient facts to support their continuing violations or fraudulent concealment

8

theories. Id.at 1218.
On May 15, 2015, Plaintiffs filed their Second Amended Complaint. ECF No. 121. Six

10

days later, Defendants filed a joint motion to dismiss the SAC. ECF No. 126. Plaintiffs filed a

11
United States District Court
Northern District of California

9

timely opposition, ECF No. 132, and Defendants replied, ECF No. 137. On August 20, 2015, the

12

Court denied Defendants’ motion to dismiss the SAC. In re Animation Workers Antitrust Litig.,

13

123 F. Supp. 3d 1175, 1207 (N.D. Cal. 2015). In its order denying Defendants’ motion to dismiss,

14

the Court found that the SAC was not barred by the statute of limitations because the SAC

15

sufficiently alleged that Defendants had fraudulently concealed the alleged conspiracy. Id.

16

Plaintiffs filed the instant motion for class certification on February 1, 2016. ECF No. 203

17

(“Mot.”). Plaintiffs filed errata to Plaintiffs’ Motion on February 10, 2016. ECF No. 214. The

18

Court’s references to Plaintiffs’ Motion refer to the corrected brief. Plaintiffs attached as an

19

exhibit the expert report of Dr. Orley C. Ashenfelter, Ph.D. in support of class certification. ECF

20

No. 215-7 (“Ashenfelter Report”). Plaintiffs filed errata to the Ashenfelter Report on March 2,

21

2016. ECF No. 224. The Court’s references to the Ashenfelter Report refer to the corrected report.

22

Defendants DreamWorks, Disney, Lucasfilm, Pixar, Two Pic (ImageMovers Digital), Sony

23

Pictures Animation, and Sony Pictures Imageworks filed an opposition on March 24, 2016. ECF

24

No. 239-1. Defendants filed errata to their opposition and a corrected opposition brief on March

25

25, 2016. ECF No. 242 (“Opp.”). The Court’s references to Defendants’ Opposition refer to the

26

corrected brief. Defendants attached as an exhibit the expert report of Dr. Michael C. Keeley,

27

Ph.D. in opposition to class certification. ECF No. 241 (“Keeley Report”). Defendants filed errata

28

13
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

to the Keeley Report on April 4, 2016. ECF No. 257. The Court’s references to the Keeley Report

2

refer to the corrected report. Plaintiffs filed a reply on April 14, 2016. ECF No. 262. Plaintiffs

3

attached as an exhibit a second expert report of Dr. Ashenfelter in support of class certification.

4

ECF No. 265 (“Ashenfelter Reply Report” or “Reply Report”). On April 25, 2016, Plaintiffs filed

5

errata to their reply and a corrected reply brief. ECF No. 267-1 (“Reply”). The Court’s references

6

to Plaintiffs’ Reply refer to the corrected brief. Plaintiffs additionally filed errata to the

7

Ashenfelter Reply Report on April 25, 2016. ECF No. 268. The Court’s references to the

8

Ashenfelter Reply Report refer to the corrected report.

9

On April 29, 2016, the Court filed an order requesting supplemental briefing to address
specific questions posed by the Court. ECF No. 270. The parties filed supplemental briefs in

11
United States District Court
Northern District of California

10

accordance with the Court’s April 29, 2016 Order on May 4, 2016. ECF No. 274 (“Defendants’

12

Supp. Br.”); ECF No. 275 (“Plaintiffs’ Supp. Br.”).

13

Defendant Blue Sky did not join Defendants’ opposition to class certification. Instead,

14

Plaintiffs filed a joint motion for preliminary approval of class action settlement with Blue Sky on

15

March 31, 2016, ECF No. 249, and an amended motion for preliminary approval of class action

16

settlement on May 11, 2016, ECF No. 282.

17

After Defendants’ opposition to class certification was filed, Plaintiffs filed a motion for

18

preliminary approval of class action settlement with the Sony Defendants on May 3, 2016. ECF

19

No. 273 at 4. As part of the settlement agreement, the Sony Defendants agreed not to cooperate

20

with the remaining Defendants in opposing Plaintiffs’ motion for class certification. See id.

21

Accordingly, the only Defendants who continue to oppose class certification are DreamWorks,

22

Two Pic (ImageMovers Digital), Lucasfilm, Pixar, and Disney.

23

II. PROPOSED CLASS DEFINITION

24

In their Motion for Class Certification, Plaintiffs seek to represent the following class:

25
26
27
28

14
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

All animation and visual effects employees employed by defendants
in the United States who held any of the jobs listed in Ashenfelter
Report Appendix C during the following time periods: Pixar (20012010), Lucasfilm Ltd., LLC (2001-2010), DreamWorks Animation
SKG, Inc. (2003-2010), The Walt Disney Company (2004-2010),
Sony Pictures Animation, Inc. and Sony Pictures Imageworks, Inc.
(2004-2010), Blue Sky Studios, Inc. (2005-2010) and Two Pic MC
LLC f/k/a ImageMovers Digital LLC (2007-2010). Excluded from
the Class are senior executives, members of the board of directors,
and persons employed to perform office operations or administrative
tasks.

2
3
4
5
6
7
8
9

Mot. at v.
Plaintiffs’ proposed class definition amends the class definition from the definition set
forth in the SAC in the following ways: (1) the proposed class definition specifies that the class
period ends in 2010; (2) the proposed class definition adds the years 2001–2003 to the class period

11
United States District Court
Northern District of California

10

for Pixar and Lucasfilm; (3) the proposed class definition adds the year 2003 to the class period

12

for DreamWorks; (4) the proposed definition omits the year 2004 for BlueSky; (5) the proposed

13

class definition omits the years 2004–2006 for ImageMovers Digital; and (6) the proposed class

14

definition incorporates a list of job titles compiled by Plaintiffs’ expert, Dr. Ashenfelter.

15

At the hearing on Plaintiffs’ Motion, Plaintiffs indicated that the class definition should be

16

further amended to reference Amended Appendix C to the Ashenfelter Reply Report. May 6

17

Trans. at 22:19–21 (Plaintiffs’ counsel asking “[i]f the Court does certify the case, that the Court

18

certifies the case in reference to Appendix C of the Reply Report”). Plaintiffs estimate that this

19

proposed class includes 10,042 individuals. Plaintiffs’ Supp. Br. at 1. Defendants estimate that the

20

proposed class includes approximately 10,100 individuals. May 6 Trans. at 9:14–10:3.

21

III. LEGAL STANDARD

22

Class actions are governed by Rule 23 of the Federal Rules of Civil Procedure. Rule 23

23

does not set forth a mere pleading standard. To obtain class certification, Plaintiffs bear the burden

24

of showing that they have met each of the four requirements of Rule 23(a) and at least one

25

subsection of Rule 23(b). Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1186, amended by

26

273 F.3d 1266 (9th Cir. 2001). “A party seeking class certification must affirmatively demonstrate

27

. . . compliance with the Rule[.]” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350 (2011).

28

15
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

Rule 23(a) provides that a district court may certify a class only if: “(1) the class is so

1
2

numerous that joinder of all members is impracticable; (2) there are questions of law or fact

3

common to the class; (3) the claims or defenses of the representative parties are typical of the

4

claims or defenses of the class; and (4) the representative parties will fairly and adequately protect

5

the interests of the class.” Fed. R. Civ. P. 23(a). That is, the class must satisfy the requirements of

6

numerosity, commonality, typicality, and adequacy of representation to maintain a class action.

7

Mazza v. Am. Honda Motor Co., Inc., 666 F.3d 581, 588 (9th Cir. 2012). Further, while Rule 23(a)

8

is silent as to whether the class must be ascertainable, courts have held that the Rule implies this

9

requirement as well. See, e.g., In re Yahoo Mail Litig., 308 F.R.D. 577, 596 (N.D. Cal. 2015)
(party seeking class certification under Rule 23(b)(3) “must demonstrate that an identifiable and

11
United States District Court
Northern District of California

10

ascertainable class exists”); Herrera v. LCS Fin. Servs. Corp., 274 F.R.D. 666, 672 (N.D. Cal.

12

2011).

13

If all four prerequisites of Rule 23(a) are satisfied, the Court must also find that Plaintiffs

14

“satisfy through evidentiary proof” at least one of the three subsections of Rule 23(b). Comcast

15

Corp. v. Behrend, 133 S. Ct. 1426, 1432 (2013). Rule 23(b) sets forth three general types of class

16

actions. See Fed. R. Civ. P. 23(b)(1)–(b)(3). Of these types, Plaintiffs seek certification under Rule

17

23(b)(3). A class may be certified under Rule 23(b)(3) if a court finds that “questions of law or

18

fact common to class members predominate over any questions affecting only individual

19

members, and that a class action is superior to other available methods for fairly and efficiently

20

adjudicating the controversy.” Fed. R. Civ. P. 23(b)(3).

21

“[A] court’s class-certification analysis must be ‘rigorous’ and may ‘entail some overlap

22

with the merits of the plaintiff’s underlying claim[.]’” Amgen Inc. v. Conn. Ret. Plans & Trust

23

Funds, 133 S. Ct. 1184, 1194 (2013) (quoting Dukes, 564 U.S. at 351); see also Mazza, 666 F.3d

24

at 588 (“‘Before certifying a class, the trial court must conduct a ‘rigorous analysis’ to determine

25

whether the party seeking certification has met the prerequisites of Rule 23.’” (quoting Zinser, 253

26

F.3d at 1186)). This “rigorous” analysis applies to both Rule 23(a) and Rule 23(b). Comcast, 133

27

S. Ct. at 1432 (discussing how Congress included “addition[al] . . . procedural safeguards for

28

16
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

(b)(3) class members beyond those provided for (b)(1) or (b)(2) class members (e.g., an

2

opportunity to opt out)” and how a court has a “duty to take a ‘close look’ at whether common

3

questions predominate over individual ones”).

4

Nevertheless, “Rule 23 grants courts no license to engage in free-ranging merits inquiries

5

at the certification stage.” Amgen, 133 S.Ct. at 1194–95. “Merits questions may be considered to

6

the extent—but only to the extent—that they are relevant to determining whether the Rule 23

7

prerequisites for class certification are satisfied.” Id. at 1195. If a court concludes that the moving

8

party has met its burden of proof, then the court has broad discretion to certify the class. Zinser,

9

253 F.3d at 1186.

10

IV. DISCUSSION

United States District Court
Northern District of California

11

Plaintiffs allege that Defendants entered into a conspiracy to “restrain competition for labor

12

and reduce compensation class-wide.” Mot. at 1; SAC ¶ 42. According to Plaintiffs, Defendants’

13

agreements restrained trade and commerce, and the agreements were thus per se unlawful under

14

Section 1 of the Sherman Act. SAC ¶¶ 205–08; Mot. at 6 n.3; see 15 U.S.C. § 1 (“Every contract,

15

combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce

16

among the several States, or with foreign nations, is declared to be illegal.”). Plaintiffs seek to

17

recover the damages caused by Defendants’ alleged antitrust violation on behalf of themselves and

18

the members of the proposed class under Federal Rule of Civil Procedure 23(b)(3).

19

In order to proceed as a class action, as noted above, Plaintiffs must satisfy the

20

requirements of Federal Rule of Civil Procedure 23(a) as well as the requirements of Rule

21

23(b)(3). The Court begins by addressing the Rule 23(a) requirements and then addresses Rule

22

23(b)(3).

23

A. Rule 23(a) Requirements

24

1. Numerosity

25

Pursuant to Rule 23(a)(1), Plaintiffs must show that “the class is so numerous that joinder

26

of all members is impracticable.” Fed. R. Civ. P. 23(a)(1). Plaintiffs need not state the exact

27

number of potential class members, nor is there a bright-line minimum threshold requirement. In

28

17
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

re Rubber Chems. Antitrust Litig., 232 F.R.D. 346, 350–51 (N.D. Cal. 2005). Rather, the Court

2

must examine the specific facts of each case. Gen. Tel. Co. of Nw., Inc. v. EEOC, 446 U.S. 318,

3

330 (1980).

4

In the instant case, Defendants do not contest that numerosity is satisfied. May 6 Trans. at

5

49:6–7. There are approximately 10,042 class members. Plaintiffs’ Supp. Br. at 1. This is

6

sufficient to satisfy the numerosity requirement. See In re Beer Distrib. Antitrust Litig., 188 F.R.D.

7

557, 562 (N.D. Cal. 1999) (25 class members satisfied numerosity requirement).

8

2. Commonality

9

Rule 23(a)(2) requires that there be “questions of law or fact common to the class.” Dukes,
564 U.S. at 349. To satisfy the commonality requirement, Plaintiffs must show that the class

11
United States District Court
Northern District of California

10

members have suffered “the same injury,” meaning that class members’ claims must “depend

12

upon a common contention” of such a nature that “determination of its truth or falsity will resolve

13

an issue that is central to the validity of each [claim] in one stroke.” Id. at 350 (quotation marks

14

and citation omitted). Plaintiffs must demonstrate not merely the existence of a common question,

15

but rather “the capacity of a classwide proceeding to generate common answers apt to drive the

16

resolution of the litigation.” Id. (quotation marks omitted) (emphasis in original). Nevertheless,

17

“for purposes of Rule 23(a)(2), even a single common question will do.” Id. at 359 (alteration and

18

quotation marks omitted).

19

“Where an antitrust conspiracy has been alleged, courts have consistently held that ‘the

20

very nature of a conspiracy antitrust action compels a finding that common questions of law and

21

fact exist.’” In re TFT-LCD (Flat Panel) Antitrust Litig. (“In re TFT-LCD II”), 267 F.R.D. 583,

22

593 (N.D. Cal. 2010), amended in part by No. 07-MDL-1827-SI, 2011 WL 3268649 (N.D. Cal.

23

July 28, 2011) (quoting In re Dynamic Random Access Memory (DRAM) Antitrust Litig., No. 02-

24

MDL-1486-PHJ, 2006 WL 1530166, at *3 (N.D. Cal. June 5, 2006)). Antitrust liability alone

25

constitutes a common question that “will resolve an issue that is central to the validity” of each

26

class member’s claim “in one stroke,” Dukes, 564 U.S. at 350, “because proof of an alleged

27

conspiracy will focus on defendants’ conduct and not on the conduct of individual class

28

18
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

members.” In re TFT-LCD (Flat Panel) Antitrust Litig.(“In re TFT-LCD I”), 267 F.R.D. 291, 310

2

(N.D. Cal. 2010) (citing cases), abrogated on other grounds in In re ATM Fee Antitrust Litig., 686

3

F.3d 741, 755 & n. 7 (9th Cir. 2012).

4

Here, Plaintiffs bring claims for antitrust liability, and Defendants do not contest that the

5

allegations of antitrust conspiracy present common legal and factual issues. See generally Opp.;

6

May 6 Trans. at 49–50 (Defendants’ counsel stating that Defendants are challenging only

7

ascertainability, typicality, adequacy, predominance, and superiority). Thus, as in High-Tech, “the

8

adjudication of Defendants’ alleged antitrust violation will turn on overwhelmingly common legal

9

and factual issues.” See In re High-Tech. Emp. Antitrust Litig., 985 F. Supp. 2d 1167, 1180 (N.D.

10

Cal. 2013). Accordingly, Plaintiffs have satisfied the commonality requirement.

United States District Court
Northern District of California

11

3. Typicality

12

Under the “permissive standards” of Rule 23(a)(3), “representative claims are ‘typical’ if

13

they are reasonably co-extensive with those of absent class members; they need not be

14

substantially identical.” Hanlon v. Chrysler Corp., 150 F.3d 1011, 1020 (9th Cir. 1998); accord

15

Staton v. Boeing Inc., 327 F.3d 938, 957 (9th Cir. 2003). “The test of typicality is whether other

16

members have the same or similar injury, whether the action is based on conduct which is not

17

unique to the named plaintiffs, and whether other class members have been injured by the same

18

course of conduct.” Hanon v. Dataproducts Corp., 976 F.2d 497, 508 (9th Cir. 1992) (quotation

19

marks omitted). The purpose of the typicality requirement is to assure that the interests of the

20

named representative align with the interests of the class. See Ellis v. Costco Wholesale Corp., 657

21

F.3d 970, 984–85 (9th Cir. 2011). In antitrust cases, “typicality usually ‘will be established by

22

plaintiffs and all class members alleging the same antitrust violations by defendants.’” Pecover v.

23

Elec. Arts, Inc., No. 08-CV-2820-VRW, 2010 WL 8742757, at *11 (N.D. Cal. Dec. 21, 2010)

24

(quoting In re Playmobil Antitrust Litig., 35 F. Supp. 2d 231, 241 (E.D.N.Y. 1998)).

25

In this case, all class members, regardless of their individual employers, allege the same

26

injuries arising from common conduct: suppression of compensation due to Defendants’ anti-

27

solicitation agreements and collusive compensation agreements. See Mot. at 4 (“Plaintiffs have

28

19
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

alleged the same antitrust violation as to every class member[.]”). This allegation is sufficient to

2

satisfy the typicality requirement. See High-Tech, 985 F. Supp. 2d at 1181.

3

Defendants argue in passing in a single sentence with a citation to an inapposite case that
Plaintiffs’ claims are not typical of the class because some class members have arbitration or

5

release agreements with some Defendants, and the named Plaintiffs were not party to the same

6

agreements. Opp. at 11. Defendants cite this Court’s decision in Schulken v. Wash. Mut. Bank, No.

7

09-CV-02708-LHK, 2012 WL 28099, at *12 (N.D. Cal. Jan 5, 2012). In the portion of Schulken

8

upon which Defendants rely, the Court found that a named plaintiff’s breach of contract claim was

9

not typical of the class where class members’ claims (not the defendant’s affirmative defenses)

10

were premised on materially differing contracts. Id. Schulken does not apply in the instant case,

11
United States District Court
Northern District of California

4

however, because “defenses that may bar recovery for some members of the putative class, but

12

that are not applicable to the class representative do not render a class representative atypical

13

under Rule 23.” Barnes v. AT&T Pension Benefit Plan-Nonbargained Program, 270 F.R.D. 488,

14

494 (N.D. Cal. 2010), modified by 273 F.R.D. 562 (N.D. Cal. 2011); see also In re Live Concert

15

Antitrust Litig., 247 F.R.D. 98, 117 (C.D. Cal. 2007) (holding that the fact that “some potential

16

class members are unlikely to recover because of a unique defense” is “not relevant to typicality”);

17

In re Scientific-Atlanta, Inc. Sec. Litig., 571 F. Supp. 2d 1315, 1327 (N.D. Ga. 2007) (holding that

18

typicality was satisfied even though “certain class members may be subject to unique reliance-

19

based defenses separate and apart from those of the class representatives”); Winkler v. DTE, Inc.,

20

205 F.R.D. 235, 241–42 (D. Ariz. 2001) (holding that typicality was satisfied despite the

21

defendant’s argument that “it has valid defenses and counterclaims it may assert against some

22

class members but not the named representatives”).

23

By contrast, affirmative defenses may pose a bar to typicality “where a putative class

24

representative is subject to unique defenses which threaten to become the focus of the litigation.”

25

Hanon, 976 F.2d at 508. In such a case, “class certification should not be granted if there is a

26

danger that absent class members will suffer if their representative is preoccupied with defenses

27

unique to it.” Id. (quotation marks omitted). That concern is absent where, as Defendants argue in

28

20
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

the instant case, there may be defenses unique to some class members other than the class

2

representatives. Notably, Defendants do not contend that typicality is defeated here based on any

3

unique defenses faced by the named Plaintiffs that “threaten to become the focus of the litigation.”

4

Id.; Opp. at 11. Thus, the fact that Defendants may have affirmative defenses against some absent

5

class members does not affect the Court’s typicality analysis.
In the instant case, all class members were injured by the same alleged antitrust conspiracy

6
7

and incurred the same alleged injury—suppressed compensation caused by Defendants’ single

8

antitrust conspiracy. This is all that is required to show typicality. Hanon, 976 F.2d at 508; see

9

Newberg on Class Actions § 3:45 (5th ed. 2011) (“[T]ypicality will generally not be defeated by
allegations that the proposed class representative has released the defendants from the claims

11
United States District Court
Northern District of California

10

asserted[.]”).

12

4. Adequacy

13

Legal adequacy of a class representative under Rule 23(a)(4) turns on two inquiries:

14

(1) whether named Plaintiffs and their counsel have “any conflicts of interest with other class

15

members,” and (2) whether named Plaintiffs and their counsel will “prosecute the action

16

vigorously on behalf of the class.” Hanlon, 150 F.3d at 1020.
Defendants argue in passing in the same single sentence that references typicality that

17
18

some class members have arbitration or release agreements with some Defendants, and the named

19

Plaintiffs are not adequate representatives to challenge those agreements. Opp. at 11. Defendants

20

offer no reason why Plaintiffs’ interests would conflict with those of absent class members,

21

particularly given that a named Plaintiff has an arbitration agreement with DreamWorks, the only

22

Defendant in this case asserting arbitration agreements as a defense.8 Indeed, not only does named

23

Plaintiff Nitsch have an arbitration agreement with DreamWorks, but Nitsch also has already

24

litigated before this Court whether Nitsch’s claims are barred by his arbitration agreement with

25
26
27
28

8

The Sony Defendants previously asserted defenses based on arbitration agreements, but the Sony
Defendants have now settled with Plaintiffs. ECF No. 273; see also Defendants’ Supp. Br. at 5
(noting that only DreamWorks and the Sony Defendants had asserted arbitration agreements).
21
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

DreamWorks. See ECF No. 116 (order regarding arbitration agreement); Defendants’ Supp. Br. at

2

5.

3

Moreover, a named plaintiff is not rendered inadequate merely because he or she is not

4

subject to every affirmative defense that a defendant may assert against particular absent class

5

members. Barnes, 270 F.R.D. at 495 (holding that “the potential existence of [affirmative]

6

defenses against absent class members does not, standing alone, make [the named plaintiff]

7

inadequate”); see also Boyd v. Bank of Am. Corp., 300 F.R.D. 431, 439 (C.D. Cal. 2014) (noting

8

that “there is no authority for the proposition that an affirmative defense, which may affect some

9

members of the class, creates a conflict that otherwise defeats the adequacy of a proposed class

10

United States District Court
Northern District of California

11

representative”) (quotation marks omitted).
Plaintiffs and members of the proposed class share an interest in proving that Defendants’

12

conduct violated the antitrust laws and suppressed their compensation, and Plaintiffs have

13

diligently litigated this case. Additionally, the parties have not identified any conflicts of interests

14

the Plaintiffs have with class members. Therefore, the Court finds that the adequacy requirement is

15

satisfied with respect to the named Plaintiffs.

16

In addition, Federal Rule of Civil Procedure 23(g)(4) requires that “[c]lass counsel must

17

fairly and adequately represent the interests of the class.” Here, Defendants do not challenge the

18

adequacy of class counsel, whom the Court previously appointed as interim class counsel after

19

concluding that the requirements articulated in Rule 23(g) were satisfied. See ECF No. 33

20

(unopposed motion detailing counsels’ experience and seeking appointment as interim class

21

counsel); ECF No. 54 (order granting same). Therefore, the Court also finds that the adequacy

22

requirement is satisfied with respect to class counsel.

23

5. Ascertainability and Job Titles Included in the Class

24

In addition to the four requirements explicitly provided in Rule 23(a), courts have held that

25

Rule 23(a) also implicitly requires that the class be ascertainable. See, e.g., In re Yahoo Mail

26

Litig., 308 F.R.D. at 596; Herrera, 274 F.R.D. at 672. A class definition is sufficient if the

27

description of the class is “definite enough so that it is administratively feasible for the court to

28

22
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

ascertain whether an individual is a member.” O’Connor v. Boeing N. Am. Inc., 184 F.R.D. 311,

2

319 (C.D. Cal. 1998) (internal citation omitted). In addition, “the court must be able to [determine

3

that] class members are included or excluded from the class by reference to objective criteria.” 5

4

Moore’s Federal Practice, § 23.21[3] (3d ed. 1997).

5

In the instant case, the proposed class definition explicitly incorporates the list of job titles

6

in Appendix C to the Ashenfelter Report. Mot. at v. Dr. Ashenfelter created Appendix C by

7

matching individuals whose employee data was included in the Croner Survey to job titles at each

8

Defendant. Ashenfelter Report, Appendix C; May 6 Trans. at 26:6–27:12 (Plaintiffs’ counsel

9

explaining the process by which Dr. Ashenfelter created Appendix C). As noted above, the Croner
Survey was an annual compensation survey among animation studios, organized by Defendants,

11
United States District Court
Northern District of California

10

that provided Defendants with “wage and salary ranges for the studios’ technical or artistic

12

positions, broken down by position and experience level.” SAC ¶ 87. Plaintiffs allege that

13

Defendants used the Croner Survey to “confirm or adjust [their] salary ranges” as part of the

14

antitrust conspiracy. Id.

15

After identifying job titles included in the Croner Survey, Dr. Ashenfelter then used an

16

algorithm to help identify additional job titles related to those included in the Croner Survey, and

17

the resulting list of job titles represents a combination of job title matches to the Croner Survey

18

and Dr. Ashenfelter’s judgment about what additional job titles at each Defendant were

19

“animation and visual effects employees.” Ashenfelter Reply Report ¶¶ 138–141; Ashenfelter

20

Report, Appendix C; May 6 Trans. at 26:6–27:12.

21

After filing the Ashenfelter Report, Plaintiffs learned that the employee data for

22

DreamWorks did not include an indicator for independent contractors, such that some job titles

23

held exclusively by independent contractors at DreamWorks, such as actors, had improperly been

24

included in Appendix C. Reply at 9; Defendants’ Supp. Br. at 1. Defendants contend that inclusion

25

of independent contractors undermines Plaintiffs’ identification of class members because the

26

class is limited to “employees” and independent contractors are not employees. Opp. at 12. Dr.

27

Ashenfelter accordingly revised the list of job titles to remove the job titles held exclusively by

28

23
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

independent contractors. Reply at 9. Dr. Ashenfelter included the revised list as Amended

2

Appendix C to the Ashenfelter Reply Report. Id.; Ashenfelter Reply Report, Amended Appendix

3

C. At the hearing, Plaintiffs agreed that the proposed class definition should be narrowed to

4

incorporate only the job titles in Amended Appendix C to the Ashenfelter Reply Report. May 6

5

Trans. at 22:19–22.
Defendants argue that Dr. Ashenfelter’s selection of job titles for inclusion in Amended

7

Appendix C is unreliable. Opp. at 12–13; Defendants’ Supp. Br. at 1–2; May 6 Trans. at 35:11–

8

37:10. Defendants criticize Dr. Ashenfelter’s selection process for relying in part upon Dr.

9

Ashenfelter’s judgment instead of adhering exclusively to an objectively replicable process. Opp.

10

at 12; Defendants’ Supp. Br. at 1–2. Defendants further argue that the resulting list of job titles is

11
United States District Court
Northern District of California

6

over-inclusive, as evidenced by Dr. Ashenfelter’s inclusion of independent contractor job titles

12

such as “ACTOR” in his original Appendix C. Opp. at 12–13; Defendants’ Supp. Br. at 1–2.

13

Defendants contend that the unreliability of Amended Appendix C makes the proposed class

14

unascertainable and the expert testimony of Dr. Ashenfelter inadmissible. May 6 Trans. at 35:10–

15

12.

16

Defendants’ criticism of Dr. Ashenfelter’s selection of job titles does not defeat

17

ascertainability. The proposed class definition explicitly incorporates Dr. Ashenfelter’s list of job

18

titles and specifies for each Defendant which years are included in the class period. Mot. at v. The

19

identity of the individuals who held those job titles during the class period is readily ascertainable.

20

The list of job titles in Amended Appendix C is an “objective criteria” by which to evaluate

21

whether an individual belongs in the class. 5 Moore’s Federal Practice, § 23.21[3] (3d ed. 1997);

22

see also High-Tech, 985 F. Supp. 2d at 1182 (finding ascertainable a class of technical workers

23

where the plaintiffs’ expert provided a list of job titles for inclusion in the class).

24

Second, as to Defendants’ Daubert challenge based on Dr. Ashenfelter’s selection of job

25

titles, the Court concludes that Defendants’ criticisms go to weight, not admissibility. Federal Rule

26

of Evidence 702 allows admission of “scientific, technical, or other specialized knowledge” by a

27

qualified expert if it will “help the trier of fact to understand the evidence or to determine a fact in

28

24
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

issue.” Expert testimony is admissible pursuant to Rule 702 if it is both relevant and reliable.

2

Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 589 (1993). An expert witness may provide

3

opinion testimony if: (1) the testimony is based upon sufficient facts or data; (2) the testimony is

4

the product of reliable principles and methods; and (3) the expert has reliably applied the

5

principles and methods to the facts of the case. Fed. R. Evid. 702.

6

In the instant case, Dr. Ashenfelter’s selection of job titles for inclusion in Amended

7

Appendix C was based primarily upon Defendants’ employment data and the Croner Survey.

8

Ashenfelter Report, Appendix C; May 6 Trans. at 26:6–27:12. His selection process involved a

9

combination of matching job titles to the Croner Survey, the assistance of an algorithm, and his
own judgment about what job titles to include. Dr. Ashenfelter’s use of his judgment does not bar

11
United States District Court
Northern District of California

10

the use of the resulting list of job titles to define the class because, as this Court recognized in

12

High-Tech, an expert’s exercise of judgment can be an appropriate method of identifying job titles

13

to include in a class of employees. See High-Tech, 985 F. Supp. 2d at 1182 (approving the

14

plaintiffs’ expert’s list of job titles for inclusion of the class where the list was constructed based

15

upon the expert’s judgment). Although unlike in High-Tech Plaintiffs here did not employ an

16

additional expert to evaluate the list of job titles, Dr. Ashenfelter had the benefit of the Croner

17

Survey, which lists job titles for which Defendants exchanged compensation information, to guide

18

his selection of job titles for Amended Appendix C. Ashenfelter Report Appendix C ¶ 3; May 6

19

Trans. at 26:6-20. Only Lucasfilm and Pixar used the Croner Survey in High-Tech, so the High-

20

Tech Plaintiffs did not use the Croner Survey to define the class. See 985 F. Supp. 2d at 1182;

21

May 6 Trans. at 28:9–18. Thus, the Court concludes that Dr. Ashenfelter’s reliance upon the

22

Croner Survey to guide his exercise of judgment makes his method of selecting job titles for

23

inclusion sufficiently reliable.

24

Furthermore, Dr. Ashenfelter provided an Amended Appendix C addressing the primary

25

criticisms of the job selection list raised in Defendants’ Opposition, i.e., that Dr. Ashenfelter

26

included independent contractors including actors. Dr. Ashenfelter repeated all of the analyses in

27

the Ashenfelter Report using the list of job titles from Amended Appendix C, which excludes

28

25
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

independent contractors including actors. See Ashenfelter Reply Report, Appendix B. The results

2

of Dr. Ashenfelter’s analyses did not change significantly in response to the change in the list of

3

job titles. Compare Ashenfelter Report (analysis using Appendix C to Ashenfelter Report) to

4

Ashenfelter Reply Report, Appendix B (analysis using Amended Appendix C to Ashenfelter

5

Reply Report). That fact undercuts Defendants’ contention that a dispute over the exact job titles

6

Dr. Ashenfelter chose to include undermines the fundamental methodological reliability of Dr.

7

Ashenfelter’s report’s ability to show classwide antitrust violation, impact, and damages.

8

Accordingly, Dr. Ashenfelter’s selection of job titles is sufficiently reliable to survive a Daubert

9

challenge.

10

In summary, the Court finds that Plaintiffs’ proposed class definition referencing Amended

United States District Court
Northern District of California

11

Appendix C is ascertainable, and Dr. Ashenfelter’s report using Amended Appendix C is

12

admissible. For the reasons stated above, the Court finds that Plaintiffs have satisfied the

13

requirements set forth by Rule 23(a).

14
15

B. Rule 23(b)(3): Predominance
Plaintiffs also contend that their proposed class satisfies the requirements of Rule 23(b)(3).

16

Defendants, however, argue that Plaintiffs’ proposed class does not satisfy Rule 23(b)(3)’s

17

predominance requirement because (1) antitrust impact; (2) antitrust damages; (3) fraudulent

18

concealment; and (4) issues regarding arbitration or release of claims agreements cannot be proven

19

on a classwide basis. Opp. at 2–11, 13–25. For the reasons discussed below, the Court finds that

20

questions common to the class are likely to predominate over any individual questions.

21

1. Principles Governing the Predominance Analysis

22

The predominance analysis focuses on “the legal or factual questions that qualify each

23

class member’s case as a genuine controversy” to determine “whether proposed classes are

24

sufficiently cohesive to warrant adjudication by representation.” Amchem Prods., 521 U.S. at 623;

25

see also Fed. R. Civ. P. 23(b)(3) (to certify a class, the court must find that “questions of law or

26

fact common to class members predominate over any questions affecting only individual

27

members”).

28

26
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

“Considering whether questions of law or fact common to class members predominate

2

begins . . . with the elements of the underlying cause of action.” Erica P. John Fund, Inc. v.

3

Halliburton Co., 563 U.S. 804, 809 (2011) (quotation marks omitted). A court must analyze these

4

elements to “determine which are subject to common proof and which are subject to

5

individualized proof.” In re TFT-LCD I, 267 F.R.D. at 310–11.

6

In this case, Plaintiffs allege violations of Section 1 of the Sherman Act, California’s

7

Cartwright Act, and California’s UCL. See SAC ¶¶ 205–218; Mot. at 6 n.3. All three claims are

8

based upon the same alleged antitrust behavior. See generally SAC. “[T]o establish an antitrust

9

claim, plaintiffs typically must prove (1) a violation of antitrust laws, (2) an injury they suffered as
a result of that violation, and (3) an estimated measure of damages.” In re New Motor Vehicles

11
United States District Court
Northern District of California

10

Canadian Export Antitrust Litig. (“In re New Motors”), 522 F.3d 6, 19 n.18 (1st Cir. 2008); see

12

also High-Tech, 985 F. Supp. 2d at 1183.

13
14
15
16
17
18
19
20
21
22
23
24
25
26

In its order granting class certification in High-Tech, this Court identified five principles
that guide the Court’s predominance inquiry:
First, and most importantly, the critical question that this Court must answer is
whether common questions predominate over individual questions. Amgen, 133 S.
Ct. at 1191. In essence, this Court must determine whether common evidence and
common methodology could be used to prove the elements of the underlying cause
of action. Id. Second, in answering this question, this Court must conduct a
“rigorous” analysis. Comcast Corp., 133 S. Ct. at 1432. This analysis may overlap
with the merits, but the inquiry cannot require Plaintiffs to prove elements of their
substantive case at the class certification stage. Amgen, 133 S. Ct. at 1194. Third,
this Court must determine not only the admissibility of expert evidence that forms
the basis of the methodology that demonstrates whether common questions
predominate. Ellis, 657 F.3d at 982. Rather, this Court must also determine whether
that expert evidence is persuasive, which may require the Court to resolve
methodological disputes. Id.; see also In re Rail Freight Fuel Surcharge Antitrust
Litig., 725 F.3d [244, 255 (D.C. Cir. 2013)]. Fourth, the predominance inquiry is
not a mechanical inquiry of “bean counting” to determine whether there are more
individual questions than common questions. Butler [v. Sears, Roebuck & Co., 727
F.3d 796, 801 (7th Cir. 2013)]. Instead, the inquiry contemplates a qualitative
assessment, which includes a hard look at the soundness of statistical models. Id.;
In re Rail Freight Fuel Surcharge Antitrust Litig., 725 F.3d at 255. Fifth, Plaintiffs
are not required to show that each element of the underlying cause of action is
susceptible to classwide proof. Amgen, 133 S. Ct. at 1196. Rather, they need only
show that common questions will predominate with respect to their case as a
whole. Id.

27
28

27
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

High-Tech, 985 F. Supp. 2d at 1186–87. As noted above, the Ninth Circuit denied the High-Tech

2

defendants’ petition for 23(f) review of the Court’s order applying the foregoing framework to

3

grant class certification. See In re High-Tech Empl. Antitrust Litig., No. 13-80223, ECF No. 18

4

(9th Cir. Jan. 14, 2014).

5

Plaintiffs argue that common questions will predominate all three elements of the antitrust

6

causes of action: (1) antitrust violation, (2) antitrust impact, and (3) damages. Mot. at 5–25; Reply

7

at 9–15. Plaintiffs assert that the same evidence that will predominate the antitrust causes of action

8

will also predominate the elements of the existence, impact, and damages of Defendants’

9

conspiracy, as necessary to prove Plaintiffs’ UCL cause of action. Mot. at 6 n.3. Defendants do not

10

United States District Court
Northern District of California

11

separately contest any elements of Plaintiffs’ UCL cause of action.
At the hearing, Defendants conceded that classwide evidence will predominate

12

determination of the first element, the existence of an antitrust violation. May 6 Trans. at 50:20–

13

25. Thus, Defendants dispute the latter two elements of Plaintiffs’ antitrust causes of action. In

14

addition, Defendants raise two additional challenges to predominance: (1) that overwhelming

15

individual inquiries with regard to fraudulent concealment will be required, because Plaintiffs’

16

claims are otherwise barred by the statute of limitations, see In re Animation Workers Antitrust

17

Litig., 123 F. Supp. 3d at 1193–94; and (2) that overwhelming individual inquiries will be required

18

due to arbitration agreements and releases signed by some class members.

19
20

The Court first addresses each of the three antitrust elements. The Court then addresses the
statute of limitations and arbitration/release of claims arguments.

21

2. Antitrust Violation

22

To prevail on a cause of action for violation of Section 1 of the Sherman Act, a plaintiff

23

must show that: “(1) there was an agreement, conspiracy, or combination between two or more

24

entities; (2) the agreement was an unreasonable restraint of trade under either a per se or rule of

25

reason analysis; and (3) the restraint affected interstate commerce.” Am. Ad Mgmt., Inc. v. GTE

26

Corp., 92 F.3d 781, 784 (9th Cir. 1996); see also Tanaka v. Univ. of S. Cal., 252 F.3d 1059, 1062

27

(9th Cir. 2001).

28

28
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

Plaintiffs allege that “Defendants conspired to suppress compensation by agreeing not to

2

solicit each other’s employees, to take special procedures when contacted by each other’s

3

employees, and to coordinate compensation policies through direct, collusive communications.”

4

Mot. at 6. Accordingly, Plaintiffs contend that proving the existence of this conspiracy “will be the

5

main issue at trial and will be established through common evidence.” Id.

6

In support of Plaintiffs’ allegations, Plaintiffs have set forth copious common evidence in

7

the form of Defendants’ internal work documents, deposition transcripts, and email exchanges

8

between Defendants’ directors, officers, and senior managers, all of which support Plaintiffs’

9

allegations that Defendants entered into express agreements not to compete for one another’s

10

United States District Court
Northern District of California

11

employees and to coordinate compensation policies.
Although Defendants have since conceded that the alleged antitrust violations will turn on

12

common legal and factual issues, see May 6 Trans. at 50:20–25, the Court reviews the evidence of

13

alleged antitrust violations because such evidence forms the basis of the Court’s analysis of

14

Plaintiffs’ theories of antitrust impact and damages.

15
16

a. Anti-Solicitation Agreements
Plaintiffs allege that the conspiracy began in the 1980s with Pixar and Lucasfilm entering a

17

“gentleman’s agreement” because the two companies “want[ed] to avoid bidding wars” over

18

employees. ECF Nos. 204–209 (“Talge Decl.”), Exh. 49. George Lucas, the former Lucasfilm

19

Chairman of the Board and CEO, testified that Lucasfilm had a policy that “we would not actively

20

go out and recruit from other companies.” Talge Decl., Exh. 2 at 74:19–19. According to Lucas,

21

Lucasfilm “had a general policy, because we were out to promote other digital companies and help

22

them, that the—we weren’t going to try to recruit people from them. . . . It’s not a normal

23

industrial competitive situation.” Id. at 51:18–21, 52:5–6.

24

Pixar and Lucasfilm additionally set in place strict procedures each company would follow

25

if contacted by an employee of the other company seeking to change employment. Plaintiffs have

26

produced an internal Pixar document explaining that Pixar would “never counter if the candidate

27

comes back to us with a better offer from Lucasfilm.” Talge Decl., Exh. 8. Internal emails within

28

29
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

Lucasfilm’s human resources department likewise confirm that Lucasfilm abided by these

2

procedures and had “actually canceled offers to people that Pixar said were ‘essential.’” Talge

3

Decl., Exh. 50.

4

The documentary evidence tends to show that DreamWorks joined the conspiracy as early
as 2003. For example, Pixar’s President Ed Catmull testified that at some point Steve Jobs

6

discussed with DreamWorks’ CEO Jeffrey Katzenberg having Pixar and DreamWorks enter an

7

agreement related to employee recruitment. Talge Decl., Exh. 1 at 55:6–21; 56:11–14. A 2003

8

email from Pixar executive Mary Conlin to Jobs confirmed that at that time Pixar had an

9

“agreement with Dreamwork [sic] to not poach their people.” Talge Decl., Exh. 135. In 2004,

10

Catmull reported to Jobs that Pixar’s “no raid arrangement” with DreamWorks “ha[d] worked

11
United States District Court
Northern District of California

5

quite well.” Talge Decl., Exh. 10. Also in 2004, DreamWorks’ Head of Human Resources, Kathy

12

Mandato, confirmed to Pixar’s Vice President of Human Resources, Lori McAdams, that

13

DreamWorks and Pixar had an agreement to notify each other when an offer was made to the

14

other’s employees. Talge Decl., Exh. 11. Similarly, a list of Lucasfilm’s “Gentleman’s

15

Agreements” from 2006 states that, in addition to the anti-solicitation agreement with Pixar,

16

Lucasfilm would not “recuit actively or passively from Dreamworks . . . for ANY positions.”

17

Talge Decl., Exh. 50.

18

By 2004, the alleged conspiracy grew to encompass Disney and the Sony Defendants as

19

well as Pixar, Lucasfilm, and DreamWorks. A 2004 email within Pixar’s human resources

20

department stated that Pixar had anti-solicitation agreements with DreamWorks, Disney, and ILM

21

(a division of Lucasfilm). Talge Decl., Exh. 11. Another 2004 email within Pixar’s human

22

resources department likewise confirmed that Pixar had “a gentlemen’s agreement” regarding

23

recruiting with ILM and Disney. Talge Decl., Exh. 12. In 2007, following Disney’s 2006 purchase

24

of Pixar, Catmull wrote to Disney’s Chairman Dick Cook describing how the animation studios in

25

Northern California “have conscientiously avoided raiding each other” because recruiting from

26

other studios “seriously messes up the pay structure.” Talge Decl., Exh. 7. Cook responded to

27

Catmull that he agreed and that Disney would treat Pixar’s employees as “off limits.” Id.

28

30
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

As to the Sony Defendants, in 2004, Catmull sent an email to Jobs stating that “We don’t

2

have a no raid arrangement with Sony. . . . I probably should go down and meet with Sandy and

3

Penney and Sony to reach some agreement. Our people are become [sic] really desirable and we

4

need to nip this in the bud.” Talge Decl., Exh. 10. Then, as memorialized by another Pixar

5

employee in a 2005 email, Catmull met with Sony in 2004 “and asked [Sony] to quit calling all

6

[Pixar’s] employees.” Talge Decl., Exh. 13. Subsequently, Pixar’s Lori McAdams stated that she

7

would contact Sony about Pixar’s anti-solicitation agreement with Sony to “make sure they’re still

8

honoring it as they may have had turnover in their Recruiting team.” Talge Decl., Exh. 16.

9

The evidence also indicates that, by 2005, Blue Sky joined the conspiracy. In 2005, an
internal Pixar human resources email described Pixar’s agreement not to hire Disney employees

11
United States District Court
Northern District of California

10

without written permission from Disney and “gentleman’s agreements not to directly

12

solicit/poach” from Lucasfilm, Sony, or Blue Sky. Talge Decl., Exh. 15. The email further

13

indicated that the “gentleman’s agreements” were “mutual,” such that the other Defendants

14

likewise had agreed not to recruit from Pixar. Id. George Lucas testified in his deposition that

15

Lucasfilm had a policy “not to recruit” from Blue Sky. Talge Decl., Exh. 2 at 75:4–9.

16

Furthermore, a 2008 internal email from Blue Sky’s Human Resources Director Linda Zazza

17

indicated that Zazza would check with employees who left Blue Sky to verify whether they had

18

been approached by Pixar or other studios. Talge Decl., Exh. 115.

19

In 2007, Pixar’s Ed Catmull spoke with Steve Starkey, one of the founders of

20

ImageMovers, to discuss “how important it is that we not have a hiring war.” Talge Decl., Exh.

21

18. In the same conversation, Starkey said that he had agreed with George Lucas that

22

ImageMovers would not recruit from ILM. Id. A 2009 internal ILM email confirmed that ILM had

23

“a gentlemen’s agreement with IMD that we cannot recruit people from their studio.” Talge Decl.,

24

Exh. 51. Additionally, an internal “Competitors List” maintained by Pixar described Pixar as

25

maintaining anti-solicitation agreements not to recruit from Blue Sky, DreamWorks,

26

ImageMovers Digital, Disney, and Sony. Talge Decl., Exh. 20.

27
28

b. Explicit Collusion on Compensation Policies
31
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

In addition to the documentary evidence that Defendants agreed not to recruit from each

2

other, the documentary evidence supports Plaintiffs’ allegations that Defendants colluded on

3

compensation policies through industry surveys including the Croner Survey, annual closed-door

4

in-person meetings, and emails.

5

The documentary evidence indicates that Defendants used industry surveys to collude to

6

keep compensation low. In 2004, Lucasfilm’s President Jim Morris sent an email to Pixar’s Ed

7

Catmull inviting Pixar to participate in a salary survey, paid for by Lucasfilm. Talge Decl., Exh. 6.

8

Morris wrote that he was inviting Pixar to participate because he knew Catmull was “adamant

9

about keeping a lid on rising labor costs, so I thought it might be something you’d want to be
involved with.” Id. In 2006, Pixar’s Vice President of Human Resources Lori McAdams emailed

11
United States District Court
Northern District of California

10

the Croner Company to discuss the timing of the Croner Survey in relation to the Siggraph

12

computer graphics conference at which Defendants conducted “heavy recruiting.” Talge Decl.,

13

Exh. 21. McAdams emphasized to Croner that “[h]aving updated survey data prior to that event is

14

always important so we can each confirm or adjust our salary ranges” based on the results of the

15

Croner Survey. Id. McAdams additionally suggested Croner contact DreamWorks and Sony to

16

discuss the timing of the Croner Survey. Id.

17

Furthermore, at official industry survey meetings, Defendants’ human resources directors

18

would hold their own private meetings to discuss compensation policies in person. Attendance at

19

these meetings was limited. In 2006, DreamWorks’ Head of Human Resources Kathy Mandato

20

organized one such “intimate” dinner with attendees from only DreamWorks, Disney, Pixar, Blue

21

Sky, Sony, and Lucasfilm. Talge Decl., Exhs. 75, 116. At these meetings, Defendants would

22

discuss specific details of their compensation policies. For example, the meeting notes from a

23

human resources dinner in 2006 with participants from DreamWorks, Disney, Pixar, Sony, and

24

Lucasfilm indicate that compensation policies had been discussed in significant detail, and that

25

one topic of discussion had been “Salary Increases.” Talge Decl., Exh. 76. In particular, the notes

26

state that “[a]ll of the companies give out 4% increases on average.” Id. Additionally, in January

27

2007, Pixar’s Lori McAdams emailed human resources directors at Lucasfilm, Sony, Disney,

28

32
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

DreamWorks, and Blue Sky to share the agenda for an upcoming meeting between the human

2

resources directors. Talge Decl., Exh. 54. Topics for discussion at the meeting included “Long

3

Term Incentive programs,” “General hiring plans for 2007,” and “Health plans and how you’re all

4

managing the escalating costs.” Id. Similarly, in December 2007 McAdams emailed Disney Senior

5

Vice President of Human Resources Marjorie Randolph explaining that, at the upcoming Croner

6

Survey planning meeting, “Friday is HR Directors networking day (we get the group together

7

twice a year to talk/visit/compare/benchmark stuff).” Talge, Decl. Exh. 24.

8
9

The documentary evidence further demonstrates that Defendants exchanged extensive
compensation information outside of in-person meetings. A 2006 email from Pixar’s McAdams to
her counterparts at DreamWorks, Sony, ILM, Disney, and Blue Sky asked for their companies’

11
United States District Court
Northern District of California

10

budgets for salary increases for 2007 and informed them that Pixar’s budget was “4% but we may

12

manage it to closer to 3% on average.” Talge Decl., Exh. 57. Sony’s Sharon Berlin responded that

13

Sony was “doing the same, 4% but trying to manage to 3% when we can.” Id. Similarly, an

14

internal email sent by McAdams explained that Pixar generally would “share salary ranges with

15

other employers/HR folks who write or call and ask us about how we compensate positions that

16

are the same as theirs,” particularly with other employers in the animation industry. Talge Decl.,

17

Exh. 31. Furthermore, a Lucasfilm “Pay For Performance Executive Review” dated December

18

2006 demonstrates the level of detail at which Defendants were exchanging compensation

19

information. Talge Decl., Exh. 56. This Lucasfilm Executive Review includes the exact merit

20

budgets for 2005 and 2006 for Disney, Pixar, Sony, and DreamWorks. Id.

21

Moreover, Plaintiffs have produced evidence that Defendants began changing their

22

behavior following the DOJ investigations because Defendants recognized that these exchanges of

23

compensation information through in-person meetings and emails were collusive. After Pixar and

24

Lucasfilm came under DOJ scrutiny, Defendants began curtailing their in-person meetings and

25

email exchanges. In November 2009, Pixar’s Lori McAdams sent an email to Sony and Disney

26

explaining that McAdams was “not planning to attend the Croner meeting” in January 2010. Talge

27

Decl., Exh. 99. McAdams stated that it was “time for me to let go and bring a close to that era. I

28

33
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

have a capable team who can/should represent us and without some of the other cronies it doesn’t

2

make sense.” Id. Then in 2010, McAdams sent an email to Kathy Mandato, formerly of

3

DreamWorks, explaining that “[s]ince the DOJ rained on our parade of getting together with other

4

companies, there’s not really a good reason for me to go” to the annual Siggraph conference.

5

Talge Decl., Exh. 36. In 2011, Pixar Senior Recruiter Dawn Haagstad wrote to Sony Pictures

6

Animation’s Director of Recruiting, Jana Day, explaining that Pixar could no longer respond to

7

requests for information about compensation ranges. Talge Decl., Exh. 37. Haagstad explained

8

that, as a result of the Consent Decree Pixar signed with the DOJ, “the prospect of any direct

9

communication with another company about salary ranges makes our lawyers nervous.” Id.

10

Thus, the documentary evidence supports Plaintiffs’ contention that common legal and

United States District Court
Northern District of California

11

factual issues will predominate as to whether Defendants maintained a conspiracy not to recruit

12

each other’s employees and to improperly share compensation information through industry

13

surveys, in-person meetings, and email exchanges.

14

3. Antitrust Impact

15

Having found that common questions will predominate with respect to the first element,

16

antitrust violation, the Court now turns to the second element, antitrust impact. “Antitrust

17

‘impact’—also referred to as antitrust injury—is the ‘fact of damage’ that results from a violation

18

of the antitrust laws.” In re DRAM Antitrust Litig., 2006 WL 1530166, at *7. “It is the causal link

19

between the antitrust violation and the damages sought by plaintiffs.” In re New Motors, 522 F.3d

20

at 19 n.18.

21

Ultimately, the Court is not tasked at this phase with determining whether Plaintiffs will

22

prevail on these theories. Rather, the question is more narrow: whether Plaintiffs have presented a

23

sufficiently reliable theory to demonstrate that common evidence can be used to demonstrate

24

impact. Based on the extensive documentary evidence, economic theory, data, and expert

25

statistical modeling, Plaintiffs’ methodology demonstrates that common issues are likely to

26

predominate over individual issues. For the reasons discussed below, Defendants’ attempts to

27

identify flaws that would undermine Plaintiffs’ entire methodology are unavailing.

28

34
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

a. Overview of Antitrust Impact Evidence

1
2
3
4
5
6
7
8
9
10

United States District Court
Northern District of California

11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

Before describing the specific evidence marshalled by Plaintiffs to show that antitrust
impact may be proven by classwide evidence, the Court begins by providing an overview of
Plaintiffs’ theory for how classwide evidence can demonstrate antitrust impact. Plaintiffs’
argument for classwide antitrust impact proceeds in two stages. First, Plaintiffs present evidence
that Defendants’ anti-solicitation agreements and collusion over compensation policies would
have had the effect of directly suppressing compensation for some class members. Second,
Plaintiffs present evidence that because of the ways in which Defendants determined
compensation for employees generally, including the use of formal compensation structures that
are not inherently collusive, the collusive suppression of compensation for certain class members
would have spread throughout the class and suppressed compensation to anti-competitive levels
classwide.
Plaintiffs’ documentary and expert evidence that Defendants’ antitrust violation would
have directly suppressed compensation for some class members focuses on demonstrating that
cold calling and compensation collusion have the effect of reducing overall compensation. As in
High-Tech, Plaintiffs note that cold calling, a recruitment tool that Defendants viewed as likely to
yield the most valuable recruits, has the effect of spreading information about salaries and benefits
from recruiters of one firm to employees of another. Ashenfelter Report ¶¶ 54–59. Such
information could then spread to other employees within a firm and beyond, leading to widespread
increases in employee compensation across the labor market due to increased access to
information. Id. By agreeing not to cold call each other’s employees, Defendants restricted this
flow of information among all class members. Id. Moreover, Plaintiffs allege that Defendants
explicitly colluded to suppress compensation by exchanging compensation information, including
specific compensation for certain job titles, through industry surveys, private meetings between
Defendants’ human resources directors, and email exchanges. Mot. at 9–13. Thus Plaintiffs argue
that, by agreeing not to cold call each other’s employees and by colluding on compensation,
Defendants directly suppressed the compensation of class members.
35
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

To demonstrate that this compensation suppression would have spread throughout the
class, Plaintiffs present documentary and expert evidence that Defendants had compensation

3

structures that prioritized “internal” and “external” equity. Internal equity seeks to ensure that

4

individuals performing similar jobs are compensated on a similar level. Ashenfelter Decl. ¶ 85. As

5

evidence of internal equity, Plaintiffs contend that Defendants had company-wide compensation

6

structures, which organize employees by job titles whose compensation ranges were evaluated by

7

reference to all other job titles within each company. Mot. at 18; Ashenfelter Report at 79–87.

8

Because of Defendants’ desire to maintain internal equity between employees, the upward

9

pressure that cold calls placed on the salaries of individual employees who would have received

10

the calls would have also affected other employees who were part of the same salary structure.

11
United States District Court
Northern District of California

2

Similarly, Defendants’ concern for internal equity would have caused the suppression of

12

compensation through Defendants’ explicit collusion over compensation policies to impact

13

employees beyond those included in the Croner Survey and Defendants’ discussions. Thus,

14

suppression of individual employees’ salaries would affect other employees who were in a similar

15

position and suppression of compensation ranges for specific job titles would affect the

16

compensation ranges for comparable job titles within each Defendant.

17

Whereas internal equity seeks to equalize compensation to similar employees within a

18

single Defendant, external equity seeks to ensure that individuals performing similar work across

19

Defendants are compensated similarly. Ashenfelter Decl. ¶ 88. Plaintiffs argue that Defendants

20

sought to maintain external equity by benchmarking compensation for employees based on what

21

comparable employees made at the other Defendants. Mot. at 9–13. Plaintiffs argue that

22

Defendants obtained information about compensation at other Defendants through industry

23

surveys such as the Croner Survey, in-person meetings, and emails regarding compensation. Id.

24

Thus, synthesizing Plaintiffs’ theories, Plaintiffs’ argument for antitrust impact proceeds as

25

follows: Plaintiffs argue that Defendants’ alleged antitrust violations would have directly

26

suppressed compensation for some class members. Then, as a result of Defendants’ emphasis on

27

internal equity, compensation suppression would have spread beyond the employees directly

28

36
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

affected by the antitrust violations to impact all class members within each Defendant. At the same

2

time, Defendants’ goal of maintaining external equity would have spread the effects of

3

compensation suppression between Defendants. This is the same approach to showing classwide

4

antitrust impact approved by this Court in High-Tech. See High-Tech, 985 F. Supp. 2d at 1206

5

(describing the approach to demonstrating classwide antitrust impact by showing direct impact to

6

some class members combined with evidence of internal and external equity).

7

Against that overarching framework of Plaintiffs’ theory of antitrust impact, the Court

8

considers below the substantial evidence that Plaintiffs argue demonstrates that common questions

9

will predominate over individual questions in determining the impact of the alleged antitrust

10

United States District Court
Northern District of California

11
12

violations.
b. Documentary Evidence
The Court begins its discussion of the thousands of pages of documents submitted by

13

Plaintiffs with the documentary evidence on the importance of cold calling as a recruitment tool

14

and the effect of the preclusion of cold calling on the class as a whole. The Court then discusses

15

the evidence regarding Defendants’ compensation structure and the role of internal equity in that

16

compensation structure. The Court finally turns to the documentary evidence that Defendants

17

strove to maintain external equity in compensation among Defendants.

18

i.

Cold Calling and Recruitment

19

Plaintiffs have produced classwide documentary evidence that cold calling was an

20

important recruitment technique that, if used, would have identified the most valuable employees

21

in the animation industry. In particular, Defendants viewed passive candidates—those employees

22

not currently searching for new work but who could be recruited via cold calling—to be key

23

recruiting targets. Thus, an internal Disney recruiting guide indicated that one of Disney’s

24

challenges was “[g]enerating interests from passive candidates.” Talge Decl., Exh. 108. A Disney

25

summary of the “Future of Recruiting” indicated that these passive candidates represented “70%

26

of the population.” Talge Decl., Exh. 107. Similarly, internal emails regarding recruiting at Sony

27

indicate that Sony viewed candidates currently employed by other companies as more valuable

28

37
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

because “if we limit our offers only to people who are unemployed then we’re basically limiting

2

our talent pool to artists who are not in demand.” Talge Decl., Exh. 100. Likewise, an executive

3

summary on recruiting in a 2007 Lucasfilm Board of Directors meeting presentation identified

4

“Passive Talent” as “difficult to find” and stated that Lucasfilm should “[c]hange recruiting

5

strategy from gatherer to hunter” in order to “get the best and the brightest.” Talge Decl., Exh. 62.

6

As described above, see supra Section IV.B.2., Plaintiffs have produced substantial

7

classwide documentary evidence that even though passive candidates represented the most

8

desirable recruiting targets, Defendants entered into anti-solicitation agreements not to cold call

9

each other’s employees. The documentary evidence further supports Plaintiffs’ allegation that
these anti-solicitation agreements in fact did stifle recruitment of passive candidates by

11
United States District Court
Northern District of California

10

Defendants. For example, a 2008 summary on recruiting prepared for ImageMovers Digital states

12

that ImageMovers Digital’s recruiters are unable “to go into about 90% of the companies that we

13

would want to as there are ‘no recruit’ agreements in place with our studio and the top studios in

14

Southern and Northern California.” Talge Decl., Exh. 130.

15

Furthermore, Plaintiffs have identified evidence that Defendants took steps to enforce and

16

reconfirm their anti-solicitation agreements with other Defendants. For instance, in 2006,

17

DreamWorks’ Head of Human Resources Kathy Mandato emailed Pixar’s Lori McAdams to ask

18

McAdams to ensure that Pixar recruiters would not solicit DreamWorks’ employees. Talge Decl.,

19

Exh. 39. In response, McAdams assured Mandato that McAdams would “put a stop to it!” Id.

20

Similarly, Dawn Rivera-Ernster, Disney’s Director of Animation Resources, emailed an ILM

21

recruiter in 2006 to ask ILM to abide by “the Gentlewomen’s agreement” between ILM and

22

Disney. Talge Decl., Exh. 64 (emphasis in original). Likewise, internal Blue Sky emails from

23

2006 note that Blue Sky was successfully able to stop Disney from holding a recruiting event in

24

New York near Blue Sky’s headquarters. Talge Decl., Exh 122. In fact, the evidence tends to show

25

that Defendants considered the anti-solicitation agreements so important that in 2007 Pixar’s

26

President Ed Catmull personally sent an email reprimanding a DreamWorks recruiter who had

27

attempted to recruit a Pixar employee for violating Pixar’s “agreement with Dreamworks not to

28

38
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1
2

actively pursue each others employees.” Talge Decl., Exh. 38.
Additionally, the documentary evidence tends to show that after the DOJ began its

3

investigations in 2009, Defendants began to resume directly recruiting from one another. In late

4

2009, a Blue Sky employee was “poached by Sony,” prompting an internal email at Blue Sky

5

remarking “and so it begins……” Talge Decl., Exh. 123 (ellipsis in original). In 2010,

6

DreamWorks created “a list of story or art people that we might want to proactively call” from

7

Disney. Talge Decl., Exh. 82. In late 2010, a Disney recruiter remarked in an email chain

8

discussing possible recruits that “I guess the theme of this year’s expo is ‘Poach What You

9

Want.’” Talge Decl., Exh. 110. By resuming direct solicitation of each other’s employees after the
dissolution of the alleged conspiracy, Defendants implicitly demonstrated that direct recruiting

11
United States District Court
Northern District of California

10

was a valuable tool that Defendants would have engaged in but for Defendants’ anti-solicitation

12

agreements.

13

Thus, Plaintiffs’ evidence suggests that the anti-solicitation agreements eliminated a key

14

tool of recruitment, cold calling. The documentary evidence further suggests that Defendants

15

eliminated cold calling in order to reduce the cost to retain employees. This common evidence

16

provides support for Plaintiffs’ theory that if the anti-solicitation agreements did not exist,

17

Defendants would have had to take other action to retain employees.

18

ii.

Compensation Structure and Internal Equity

19

As set forth below, Plaintiffs’ documentary evidence tends to show that Defendants

20

maintained formal compensation structures and made significant efforts to maintain internal equity

21

within those structures. This additional documentary evidence further supports Plaintiffs’ theory

22

that the downward pressure from anti-solicitation agreements and collusion over compensation

23

policies would have impacted Defendants’ salary structures and thereby suppressed all class

24

members’ salaries.

25

Plaintiffs produced documentary evidence indicating that Defendants assigned employees

26

to specific pay bands and ranges based on their job titles. In the deposition of Lucasfilm’s Director

27

of Human Resources Sharon Coker, Coker explained that “we had identified levels of positions

28

39
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

offer new hires. For example, a 2005 Lucasfilm email discussing what salary offer to make a new

2

hire notes that Lucasfilm would make an offer “after reviewing possible internal equity issues.”

3

Talge Decl., Exh. 72. Similarly, emails within Pixar discussing hiring new recruits in 2005 state

4

that Pixar’s proposed salary offers were “the best we can do and still keep internal equity which is

5

important to us.” Talge Decl., Exh. 42. Likewise, in 2008, Blue Sky discussed making salary

6

adjustments “across the board with our current staff come June to compensate for higher salaries

7

of bringing people in at market rate.” Talge Decl., Exh. 127. Blue Sky noted that these

8

adjustments would be necessary because “salary inequality breeds huge morale problems with the

9

floor.” Id.

10

In sum, Plaintiffs’ evidence supports their theory that Defendants’ formal compensation

United States District Court
Northern District of California

11

structures, combined with the premium Defendants placed on internal equity, created a

12

compensation system in which any individual class member’s compensation was intertwined with

13

that of her peers. The Court finds persuasive Plaintiffs’ contention that common questions about

14

the impact of Defendants’ compensation structures, Defendants’ prioritization of internal equity,

15

and the effects of these factors on the class members as a whole are likely to predominate over any

16

individual questions regarding internal equity.

17
18

iii.

Compensation Collusion and External Equity

Thus far, the Court has discussed Plaintiffs’ documentary evidence regarding the effects of

19

cold calling, Defendants’ compensation structures, and Defendants’ internal equity concerns on

20

wage suppression across the class. Now, the Court turns to documentary evidence that tends to

21

show that Defendants colluded on compensation and benchmarked their compensation against

22

each other and against common external sources.

23

The Court has already described the substantial documentary evidence that Defendants

24

collusively exchanged compensation information through industry surveys, in-person meetings,

25

and email exchanges. See supra Section IV.B.2.b. Defendants’ internal documents likewise reflect

26

the importance Defendants placed on maintaining external equity with other Defendants. For

27

example, an internal DreamWorks email in 2006 included language explaining the “total pay

28

41
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

package” for employees and stated that pay would be driven in part by “The Market—Comparing

2

our Total Pay package to other companies where we compete for talent.” Talge Decl., Exh. 94. In

3

another email exchange, DreamWorks’ human resources department discussed whether they

4

should “look at what we are paying our lighters and make an adjustment based on what we learn

5

that others are getting elsewhere and what Phil can learn by calling his counterparts at other

6

studios.” Talge Decl., Exh. 96. Pixar prepared a “Talking Points” memo on employee

7

compensation that stated that “Pixar participates annually in the Computer Graphics Industry

8

Survey, as well as the Radford High-Tech Survey to benchmark our compensation against other

9

companies in our industry.” Talge Decl., Exh. 45. Lucasfilm also compared its compensation to
that of Defendants, as exhibited by a Compensation Analysis project that compared compensation

11
United States District Court
Northern District of California

10

at Disney, Sony, and DreamWorks to compensation at Lucasfilm. Talge Decl., Exh. 74.

12

Furthermore, the documentary evidence suggests that the purpose of Defendants’ collusive

13

communications regarding compensation was to reduce compensation generally. Lucasfilm’s

14

President Jim Morris explained in an email that the purpose of the industry surveys was to “keep[]

15

a lid on rising labor costs.” Talge Decl., Exh. 6. Similarly, Pixar’s Vice President of Human

16

Resources Lori McAdams explained to Lucasfilm’s Sharon Coker that “[s]ince money can always

17

be a factor, that’s the other thing we should consider (e.g., we wouldn’t want to offer a lateral

18

move more money than you, and vice versa).” Talge Decl., Exh. 41.

19

The common evidence therefore indicates that Defendants sought to enter into anti-

20

solicitation agreements and colluded on compensation policies in an effort to stifle increased

21

competition for labor and rising wages. To the extent that they were successful, Defendants did

22

not need to increase compensation in order to attract and retain employees as much as they

23

otherwise would have had to increase compensation in the absence of their anti-solicitation

24

agreements and collusion. See High-Tech, 985 F. Supp. 2d at 1205 (documentary evidence of anti-

25

solicitation agreements in a market with internal and external equity indicates that the anti-

26

solicitation agreements suppressed compensation). This common evidence further suggests that

27

the anti-solicitation agreements and collusion on compensation policies reached beyond individual

28

42
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

members of the proposed class and affected the compensation of the class as a whole, including

2

across Defendants. Plaintiffs’ extensive documentary evidence therefore supports the plausibility

3

of their theory that the antitrust violations had a classwide impact.

4

c. Expert Reports and Statistical Evidence
As noted above, to show that common issues predominate for the purpose of assessing

5

classwide impact, Plaintiffs retained the services of an expert, Orley C. Ashenfelter, Ph.D.9

7

Defendants presented a report from their own expert, Michael C. Keeley, Ph.D.,10 to attack Dr.

8

Ashenfelter’s analyses and conclusions. The Court begins its discussion of the expert evidence

9

regarding antitrust impact by describing Dr. Ashenfelter’s methodologies and analyses, which

10

support Plaintiffs’ theories of common impact and harm. The Court then turns to Defendants’

11
United States District Court
Northern District of California

6

criticisms of Dr. Ashenfelter’s work. In considering the expert reports, the Court evaluates not

12

only the admissibility but also the persuasiveness of the expert reports. Ellis, 657 F.3d at 982; In

13

re Rail Freight Fuel Surcharge Antitrust Litig., 725 F.3d at 255.

14

i.

15

Dr. Ashenfelter’s Opinions Based on Economic Theory, Documentary
Evidence, Data, and Statistical Analyses

In Dr. Ashenfelter’s first expert report, which was presented in support of Plaintiffs’

16
Motion for Class Certification, Plaintiffs asked Dr. Ashenfelter whether classwide evidence was
17
18
19

capable of showing that “all, or nearly all, members of the proposed Plaintiffs’ class have been
harmed as a result of the conspiracy.” Ashenfelter Report ¶ 4. In addition, Plaintiffs asked Dr.
Ashenfelter to assess whether there was “one or more accepted and feasible methods for

20
21
22
23
24
25
26
27
28

9

Orley C. Ashenfelter, Ph.D., is the Joseph Douglas Green 1895 Professor of Economics at
Princeton University. Dr. Ashenfelter earned a B.A. from Claremont McKenna College in 1964
and a Ph.D. from Princeton University in 1970. He is the former President of the American
Economics Association and the former President of the American Law and Economics
Association. He has published in labor economics, industrial organization, econometrics, and law
and economics. See Ashenfelter Report, Appendix A.
10
Michael C. Keeley, Ph.D., is an economist and Senior Advisor at Cornerstone Research, an
economic and financial consulting firm. Dr. Keeley earned an S.B. in mathematics from the
Massachusetts Institute of Technology in 1969, an M.A. in economics from the University of
Chicago in 1971, and a Ph.D. in economics from the University of Chicago in 1974. Dr. Keeley
specializes in economic, financial, and statistical analysis, including antitrust and labor economics.
See Keeley Report, Appendix A.
43
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

calculating any damages incurred by the Plaintiff class on a class-wide basis” and whether there

2

are sufficient data available to implement those methods. Id. Dr. Ashenfelter answered both of

3

these questions in the affirmative.

4

In response to the first question relating specifically to classwide evidence of antitrust

5

impact, Dr. Ashenfelter’s analysis proceeded in two steps that correspond to the two steps of

6

Plaintiffs’ overall theory of antitrust impact. First, Dr. Ashenfelter explained that economic theory,

7

documentary evidence, and statistical analyses were capable of showing that the alleged

8

conspiracy suppressed compensation generally. Id. ¶¶ 50–51. In other words, Dr. Ashenfelter

9

illustrated how classwide evidence was capable of showing that, at the very least, Defendants were
paying some members of the class less than they would have been paid in the absence of the

11
United States District Court
Northern District of California

10

alleged conspiracy. Second, Dr. Ashenfelter illustrated how economic theory, documentary

12

evidence, and statistical analyses are capable of showing that this suppression of compensation

13

affected all or nearly all class members. Id. ¶¶ 50, 78.

14
15

ii.

Suppressed Compensation Generally

Dr. Ashenfelter first concluded that classwide evidence was capable of showing that the

16

alleged conspiracy suppressed compensation of class members generally. According to Dr.

17

Ashenfelter, this first step was supported by economic models of asymmetric information. Dr.

18

Ashenfelter noted that classical economic models assume that “the ‘market price’ is simply the

19

price at which supply is equal to demand.” Ashenfelter Report ¶ 52. In reality, however, “actual

20

transaction prices can diverge greatly from the ‘market price,’” particularly in markets that

21

“feature imperfect or asymmetric information.” Id.

22

Dr. Ashenfelter opined that because employers have access to information about the wages

23

and salaries paid to their own employees and because employers can engage consultants to

24

determine “market” compensation paid by their competitors, “[e]mployers are likely to have better

25

information about the prevailing wage rates for a job or industry than workers.” Id. ¶ 54. Thus,

26

“[b]ecause workers do not have access to the same amount of information as do employers,

27

receiving cold-calls is a key avenue by which workers can obtain accurate information about the

28

44
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

value of their services.” Id. ¶ 55. Dr. Ashenfelter hypothesized that, by restricting cold calling and

2

other competition over employees, Defendants’ anti-solicitation agreements impaired information

3

flow about compensation and job offers to employees. Id. ¶¶ 54–59. Additionally, because

4

Defendants were simultaneously engaging in collusive communications regarding compensation,

5

Dr. Ashenfelter argued that Defendants’ conspiracy increased the information available to

6

Defendants. Id. ¶ 60–62.

7

In support of these hypotheses, Dr. Ashenfelter relied on the documentary evidence

8

common to the class as a whole. Id. ¶¶ 60–72.11 For example, Dr. Ashenfelter cited to Defendants’

9

internal documents explaining that Defendants valued cold calling as a way of recruiting “passive”
candidates who were “not actively seeking new employment, but who might be interested in an

11
United States District Court
Northern District of California

10

offer if contacted by a recruiter.” Id. ¶ 66. These documents show that, but for Defendants’ anti-

12

solicitation agreements, Defendants would have been competing for passive candidates by cold

13

calling each other’s employees. Dr. Ashenfelter also cited to emails between Defendants

14

indicating that Defendants participated in a salary survey to share general salary ranges for the

15

purpose of “keeping a lid on rising labor costs,” id. ¶ 63, and that competition for employees

16

“messes up the pay structure” by making pay “very high,” id. ¶ 70. These emails show that

17

Defendants viewed their collusive communications regarding compensation and their anti-

18

solicitation agreements as a means to artificially suppress employee compensation.

19

Finally, Dr. Ashenfelter bolstered his findings with standard econometric analysis utilizing

20

solely classwide evidence and methods. Dr. Ashenfelter performed a regression analysis, utilizing

21

Defendants’ internal compensation data, to illustrate class members’ under-compensation by

22

comparing compensation during the conspiracy with compensation in a conspiracy-free, but-for

23

world. Dr. Ashenfelter concluded that the regression analysis showed that the alleged conspiracy

24

artificially suppressed compensation across Defendants. Id. ¶ 77.12

25
26
27
28

11
12

This documentary evidence is described in greater detail in Section IV.B.3.b., supra.
This regression analysis, which Dr. Ashenfelter also used to show that damages may be
evaluated on a classwide basis, is described in greater detail in Section IV.B.4., infra.
45
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1
2
3
4
5
6
7
8
9
10

United States District Court
Northern District of California

11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

iii.

Widespread Effect

Second, Dr. Ashenfelter opined that economic studies and theory, documentary evidence,
and statistical analyses were capable of showing that this compensation suppression had
widespread effects on all class members.
Dr. Ashenfelter argued that Defendants used “formal compensation structures to ensure
that employees are being paid comparable salaries for comparable work, and to establish relatively
fixed pay grades between groups of employees with increasing responsibilities.” Id. ¶ 80. Dr.
Ashenfelter supported this conclusion with documentary evidence, including internal documents
from each Defendant discussing their salary structures and hierarchies and the techniques
Defendants would use to normalize salaries within these hierarchies. Id. ¶¶ 80–84. Dr. Ashenfelter
explained that “[o]ne of the primary purposes of a formal compensation structure such as those
used by the Defendants is to maintain ‘internal equity’ between employees.” Id. ¶ 85.
Dr. Ashenfelter cited to documentary evidence showing that Defendants were concerned
with internal equity. For example, Dr. Ashenfelter cited an internal Pixar email noting that it was
“difficult” to bring in new hires at pay rates “so much higher than the bulk of our in-house
people.” Id. ¶ 85. Based on this documentary evidence, Dr. Ashenfelter hypothesized that
increased compensation to passive candidates hired through cold calling would result in increased
compensation to other class members through internal equity. Id. ¶ 87.
Dr. Ashenfelter also opined that Defendants were concerned with external equity. Id. ¶ 88.
Indeed, Dr. Ashenfelter noted that one of the features of the conspiracy was that Defendants
directly coordinated on compensation for the purpose of benchmarking their compensation
policies against the compensation of other Defendants. Id. ¶ 89. Defendants also shared
compensation information for purposes of maintaining external equity through the Croner Survey,
which compared detailed employee compensation information across Defendants. Id. ¶¶ 90–92.
In addition to relying upon the documentary evidence, Dr. Ashenfelter looked to standard
economic labor theory and statistical analyses as evidence that the conspiracy would broadly
affect members of the proposed class. Id. ¶¶ 95–111. Through these analyses, Dr. Ashenfelter
46
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

provided evidence that Defendants had semi-rigid compensation structures that exhibited both

2

internal and external equity.

3

First, Dr. Ashenfelter noted that “[s]tandard human capital theory links an individual’s

4

earnings to his or her level of education, job experience, industry and industry experience, and an

5

individual’s job title in a particular industry.” Id. ¶ 96. Thus, using Defendants’ employee

6

compensation data, Dr. Ashenfelter was able to estimate the standard human capital earnings

7

model for Defendants. Id. Dr. Ashenfelter’s standard human capital model found that factors such

8

as age, experience with the Defendant, gender, and job title—many of which were explicitly part

9

of Defendants’ formal compensation structures—are able to explain between 62% to 78% of the
variation in class members’ compensation across all Defendants for each year of the alleged

11
United States District Court
Northern District of California

10

conspiracy. Id. ¶ 99; Ashenfelter Reply Report, Appendix B, Table 1R. Dr. Ashenfelter

12

additionally conducted his standard human capital model separately for each Defendant and found

13

that the model explained between 45% to 93% of the variation in class members’ compensation at

14

each Defendant per year. Ashenfelter Report ¶ 100; Ashenfelter Reply Report, Appendix B, Table

15

2R. Thus, across Defendants and across years, the standard human capital model was capable of

16

explaining a large portion of the variation in class members’ compensation. The large amount of

17

variation in compensation explained by the standard human capital model indicates that

18

idiosyncratic individual variation accounted for a relatively small amount of employee

19

compensation. Accordingly, Dr. Ashenfelter opined that the standard human capital model’s

20

ability to predict a large portion of class member compensation supports the conclusion that

21

Defendants had systematic pay structures that persisted over time. Ashenfelter Report ¶ 101.

22

Second, Dr. Ashenfelter performed a series of analyses to identify the relationship of

23

compensation, and changes in compensation, between job titles within each Defendant. Id.

24

¶¶ 102–11. In other words, for each Defendant, Dr. Ashenfelter analyzed how compensation, and

25

changes in compensation, for different job titles at that Defendant related to each other throughout

26

the class period. Id. Dr. Ashenfelter performed this analysis for all job titles that had at least 10

27

annual compensation amounts. Id. ¶ 102. The correlation analyses showed that, for the majority of

28

47
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

job titles analyzed, compensation for each Defendant-specific job title tended to increase or

2

decrease with overall compensation to class members at that Defendant. Ashenfelter Reply Report,

3

Appendix B, Tables 3R and 4R.

4

Third, Dr. Ashenfelter used standard econometric techniques to separate the impact of the
relationship between job titles from variables unrelated to the relationship between job titles that

6

could cause the compensation for those job titles to change. Ashenfelter Report ¶ 108. Dr.

7

Ashenfelter referred to these analyses as “sharing” regressions. Id. Applying these techniques, Dr.

8

Ashenfelter found that compensation for the overwhelming majority of job titles included in the

9

analysis would change both contemporaneously and over time with compensation changes for the

10

other class members employed by each Defendant. Ashenfelter Reply Report, Appendix B, Tables

11
United States District Court
Northern District of California

5

5R–6R. Based on the sharing regressions and the correlation analyses, Dr. Ashenfelter concluded

12

that his statistical analyses supported the conclusion that changes in compensation for individual

13

class members would spread throughout the remaining class members. Ashenfelter Report ¶ 111.

14
15

iv.

Defendants’ Critique of Dr. Ashenfelter’s Opinions

As noted above, Defendants rely upon the report of Dr. Keeley to argue that individual

16

inquiries regarding antitrust impact will predominate. Dr. Keeley concluded that the documentary

17

evidence, economic theory, and the nature of the labor market for animation workers all

18

undermine Dr. Ashenfelter’s conclusions regarding classwide impact. Dr. Keeley additionally

19

raised a number of challenges to Dr. Ashenfelter’s methodology for conducting analyses of

20

classwide impact. The Court begins by addressing Dr. Keeley’s arguments based on the

21

documentary evidence, economic theory, and the nature of the labor market. The Court then

22

addresses Dr. Keeley’s criticisms of Dr. Ashenfelter’s econometric analyses.

23
24

v.

Dr. Keeley’s Arguments Based on Documentary Evidence, Economic
Theory, and the Nature of the Labor Market

First, Defendants and Dr. Keeley argue that Plaintiffs’ documentary evidence does not

25
show that compensation changes for one job title within a Defendant necessarily mandated
26
compensation changes for other job titles within that Defendant. Keeley Report ¶¶ 142–146. For
27
28

48
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

example, Dr. Keeley cites to Defendants’ internal documents showing that some employees

2

received raises while others did not, and that whether to increase an employee’s compensation was

3

based on more factors than solely internal equity. Id. ¶¶ 143–45. In addition, Defendants point to

4

the Croner Survey findings for 2010, which found that compensation for some titles increased

5

while compensation for other titles decreased. ECF No. 240 (“Lannin Decl.”), Exh. 56.

6

Defendants’ argument is inapposite. Dr. Ashenfelter’s internal equity theory of common

7

antitrust impact is not that all employees and all job titles must experience parallel compensation

8

changes contemporaneously, but instead that Defendants’ use of internal equity as a factor in

9

determining class members’ compensation over time would lead to similar classwide
compensation changes. Ashenfelter Reply Report ¶ 44. The documentary evidence shows, as Dr.

11
United States District Court
Northern District of California

10

Keeley acknowledged in his deposition, see id. ¶ 44 (citing Keeley Deposition at 269:15–23), that

12

Defendants used internal equity as a factor in their compensation decisions, such that Defendants

13

would explicitly adjust the salary ranges for job titles to account for changes made in the salary

14

ranges for related job titles. Id. ¶¶ 44–50. Furthermore, Dr. Ashenfelter’s statistical analyses

15

support the conclusion that Defendants had a relatively rigid wage structure that provided similar

16

compensation to similarly situated employees. For example, as Dr. Keeley recognizes, Keeley

17

Report ¶ 167, Dr. Ashenfelter’s analyses show that common factors such as age, experience with

18

the Defendant, gender, and job title explain most of the variability in employee compensation.

19

That these common factors explain the majority of the variation in compensation supports Dr.

20

Ashenfelter’s conclusion that, overall, Defendants maintained internal equity and sought to

21

provide like compensation to like employees. Thus, the documentary evidence and statistical

22

analyses show that internal equity was a factor that Defendants used to determine whether and

23

how to change class members’ compensation. Accordingly, Dr. Ashenfelter’s conclusion that

24

increases in compensation from cold calling would tend to diffuse through Defendants’ salary

25

structure, impacting class members generally rather than just the individuals who received the cold

26

calls is reasonable.

27
28

Second, Dr. Keeley argues that variations within actual compensation for the same job
49
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

titles undermine Dr. Ashenfelter’s theory of internal equity. Dr. Keeley identifies five job titles at

2

each Defendant that appear to exhibit substantial variation between the highest and lowest

3

compensated workers within that job title. Keeley Report ¶ 137. Dr. Keeley further identifies a

4

single job title at Sony for which some employees with the job title apparently received raises the

5

same year other employees with the same job title saw their compensation decrease substantially.

6

Id. ¶ 139. Additionally, Dr. Keeley created a chart showing compensation change over time for

7

three job titles at DreamWorks, which shows that the three job titles did not experience

8

compensation changes in tandem. Id.

9

Dr. Ashenfelter refutes Dr. Keeley’s criticisms based on the compensation variation within
five specific job titles per Defendant by noting that Dr. Keeley’s analysis fails to account for the

11
United States District Court
Northern District of California

10

other factors, such as years with the Defendant and gender, that influence compensation within a

12

system that accounts for internal equity. Ashenfelter Reply Report ¶¶ 66–67. As for Dr. Keeley’s

13

analysis of one job title at Sony, Dr. Ashenfelter notes that Dr. Keeley’s calculation of hourly

14

compensation for employees with this job title failed to account for overtime hours and thus Dr.

15

Keeley identified changes in compensation unrelated to employees’ actual pay. Id. ¶ 68. By

16

accounting for overtime hours, Dr. Ashenfelter’s analysis eliminates the dramatic differences in

17

compensation change within the Sony job title observed by Dr. Keeley. Id.

18

Regarding the three job titles at DreamWorks, Dr. Ashenfelter notes that Dr. Keeley’s

19

analysis does not account for employee turnover within each job title from year to year. Id. ¶ 71.

20

Dr. Ashenfelter also argues that Dr. Keeley’s analysis does not account for employee-specific

21

factors such as age and experience that affect compensation and changes in compensation. Id. Dr.

22

Ashenfelter, controlling for employee turnover, then conducted his own analysis of compensation

23

trends for the three DreamWorks job titles analyzed by Dr. Keeley. Id. ¶ 72. This corrected

24

analysis shows that compensation across the three DreamWorks job titles followed similar trends

25

over time, as predicted by Dr. Ashenfelter’s theory of internal equity. Id. For the above reasons,

26

the Court finds unpersuasive Dr. Keeley’s concerns regarding actual compensation by job title.

27
28

Third, Dr. Keeley argues that class members would not have required cold calls to obtain
50
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

information about market wages because of the structure of the labor market for animation

2

workers. Keeley Report ¶¶ 106–33. Dr. Keeley points in particular to the fact that many class

3

members were represented by unions, which actively collected and disseminated compensation

4

information among class members. Id. ¶¶ 116–19. According to Dr. Keeley, unions therefore acted

5

as an alternative source of information, such that cold calls were not necessary. Id. Additionally,

6

Dr. Keeley argues that there were high levels of cross-hiring, wherein a class member worked for

7

multiple Defendants during the alleged conspiracy, because many class members were employed

8

on a short-term project basis and would look for new employment each time a project completed.

9

Id. ¶¶ 107–15. Because many class members looked for work frequently during the class period,

10

Dr. Keeley argues that class members would have learned about market wages without the need

11
United States District Court
Northern District of California

1

for cold calls. Id. ¶ 113.

12

With respect to the fact that many class members were represented by unions, Dr.

13

Ashenfelter contends persuasively that information collected and disseminated by unions is

14

necessarily more limited than information available to the Defendants. Ashenfelter Reply Report

15

¶ 116. This is because unions lack information regarding non-union positions and because

16

Defendants’ collusive communications regarding compensation created an additional source of

17

information asymmetry in Defendants favor. Id. Thus, even though unions increased the amount

18

of information available to class members, unions could not overcome the asymmetry caused by

19

Defendants’ alleged conspiracy.

20

As to the short tenures of many class members, Dr. Ashenfelter contends that the

21

documentary evidence supports the finding that as many as 70% of class members were passive

22

and not actively looking for new jobs during the class period. Id. ¶ 110. The documentary

23

evidence further indicates that Defendants considered these passive candidates to be the “best

24

talent,” and thus the most valuable employees. Id. Dr. Ashenfelter additionally criticizes Dr.

25

Keeley for over-estimating the effect of short-term employees because the majority of class

26

members were long-term, not short-term workers. Id. ¶ 111. Thus, by agreeing not to cold call

27

each other’s employees, Defendants artificially halted the flow of information to the majority of

28

51
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

class members. Furthermore, Dr. Ashenfelter argues that even for short-term workers, anti-

2

solicitation agreements would suppress compensation because workers are most valuable when

3

they are in the middle of a project and least valuable when they have completed a project and must

4

find a new job. Id. ¶ 112. Therefore, by refraining from recruiting passive candidates who were in

5

the middle of projects, Defendants were able to recruit candidates only when doing so would cost

6

the least.

7

The Court finds persuasive Dr. Ashenfelter’s explanation that information asymmetry

8

would persist notwithstanding the roles played by unions and short-term workers in the animation

9

industry. Therefore, the Court concludes that Dr. Keeley’s criticisms of Dr. Ashenfelter’s analysis
based on the documentary evidence, economic theory, and the nature of the animation labor

11
United States District Court
Northern District of California

10

market are unpersuasive.

12
13

vi.

Dr. Keeley’s Criticisms of Dr. Ashenfelter’s Statistical Analyses

Dr. Keeley additionally raises a number of challenges to Dr. Ashenfelter’s standard human

14

capital model, correlation analyses, and sharing regressions. The Court addresses in turn Dr.

15

Keeley’s criticisms and Dr. Ashenfelter’s responses.

16

First, Dr. Keeley argues that Dr. Ashenfelter’s standard human capital model does not

17

show that Defendants had a systematic compensation structure. Keeley Report ¶¶ 158–72. Dr.

18

Keeley argues that if job title is removed as a variable from Dr. Ashenfelter’s standard human

19

capital model, the regression no longer explains most of the variation in employee compensation.

20

Id. ¶¶ 167–72. Furthermore, Dr. Keeley contends that the variables included in Dr. Ashenfelter’s

21

standard human capital model would show that all industries, and the economy as a whole, exhibit

22

a semi-rigid compensation structure. Id. ¶ 161.

23

In reply, Dr. Ashenfelter argues, and the Court agrees, that Dr. Keeley’s arguments do not

24

identify flaws in Dr. Ashenfelter’s model but instead highlight one of its strongest findings in

25

support of Dr. Ashenfelter’s theory of internal equity. Ashenfelter Reply Report ¶¶ 51–54. Under

26

Dr. Ashenfelter’s theory of internal equity, Defendants maintained similar compensation among

27

employees with the same job title. Accordingly, the fact that job title explains a significant portion

28

52
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

of the variation in compensation supports rather than undermines Dr. Ashenfelter’s conclusions

2

because it shows that Defendants had compensation structures that enforced internal equity. Id.

3

¶ 53. Similarly, Dr. Ashenfelter asserts that it is unremarkable that most industries would exhibit

4

rigid pay structures because “[m]ost companies in most industries pay their employees based on a

5

formal salary structure.” Id. ¶ 54. The standard human capital model merely confirms that

6

Defendants participated in such an industry with a rigid pay structure. Id.

7

Second, Dr. Keeley argues that by limiting the correlation analyses and sharing regressions
to job titles with at least 10 annual compensation amounts, Dr. Ashenfelter failed to study a

9

sufficient number of job titles. Keeley Report ¶¶ 174–75, 182. Additionally, Dr. Keeley criticizes

10

Dr. Ashenfelter’s sharing regressions for studying how changes in compensation in all but one job

11
United States District Court
Northern District of California

8

title affect the remaining job title instead of studying how changes in compensation of one job title

12

affect all other job titles. Id. ¶ 181.

13

Dr. Ashenfelter addressed both of Dr. Keeley’s concerns in the Ashenfelter Reply Report,

14

and the Court concludes those responses are persuasive. Specifically, in order to include more job

15

titles in his analyses, Dr. Ashenfelter analyzed changes in compensation by grouping job titles

16

together by similar average compensation and then comparing changes in compensation between

17

groups. Ashenfelter Reply Report ¶¶ 62–64. Using this grouping method, Dr. Ashenfelter was able

18

to include in his Reply Report job titles for which he had fewer than 10 annual compensation

19

amounts. This allowed Dr. Ashenfelter to account for 88% of all annual compensation amounts in

20

the data. Id. ¶ 62. The results of the correlation analyses and sharing regressions on the increased

21

data set show that in general, compensation for one group of job titles changes similarly to other

22

groups of job titles. Id. ¶¶ 62-64. Thus, these analyses continue to support Dr. Ashenfelter’s

23

conclusion that Defendants prioritized internal equity. Id.

24

Similarly, in response to Dr. Keeley’s criticism that Dr. Ashenfelter’s sharing regressions

25

do not indicate how changes in compensation for one job title relate to compensation for all other

26

job titles, Dr. Ashenfelter modified his sharing regressions as suggested by Dr. Keeley.

27

Specifically, Dr. Ashenfelter added new analyses, suggested by Dr. Keeley, to examine how a

28

53
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

change in compensation for one job title affects all other job titles within a Defendant. Id. ¶ 59.

2

The results of these analyses show that changes in compensation for one job title tend to result in

3

changes in compensation for all other job titles. Id. In other words, Dr. Ashenfelter’s analyses

4

show that Defendants would tend to adjust compensation for job titles throughout the company in

5

response to a change in compensation for a single job title. This supports Dr. Ashenfelter’s theory

6

that Defendants maintained internal equity by adjusting employees’ compensation based on the

7

compensation of other employees of the same Defendant. Id.

8
9

Finally, Dr. Keeley argues that Dr. Ashenfelter did not sufficiently analyze correlations in
compensation across Defendants. Keeley Report ¶¶ 183–94. Dr. Keeley compares compensation
across Defendants and performs several correlation analyses across Defendants, all of which

11
United States District Court
Northern District of California

10

purport to show that compensation trends varied across Defendants. Id. Thus, Dr. Keeley argues

12

that the statistical evidence indicates that Defendants did not maintain external equity. Id.

13

Dr. Ashenfelter replies that Dr. Keeley’s analysis of correlation across Defendants is

14

flawed because it failed to control for factors such as what job function employees held at each of

15

the Defendants. Ashenfelter Reply Report ¶ 96. In response to Dr. Keeley’s criticism that Dr.

16

Ashenfelter did not sufficiently study the relationship in compensation trends between Defendants,

17

Dr. Ashenfelter performed such a regression analysis that compared compensation between

18

Defendants by matching job titles to Croner job functions. Id. ¶¶ 96–99. Dr. Ashenfelter’s analysis

19

shows that, for the overwhelming majority of the job functions studied, compensation for job titles

20

corresponding to similar job functions between Defendants tended to change together. Id. ¶ 99.

21

Additionally, the similarities in changes in compensation for similar job functions across

22

Defendants were statistically significant. Id. Thus, Dr. Ashenfelter’s analysis of correlation across

23

Defendants supports Dr. Ashenfelter’s theory that Defendants sought to maintain external equity

24

between Defendants.

25

In sum, the Court does not find persuasive Defendants’ criticisms of Dr. Ashenfelter’s

26

methodology. Defendants have not presented analysis that undermines the reliability of Dr.

27

Ashenfelter’s analysis, and many of Defendants’ arguments are contradicted by the documentary

28

54
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

evidence. Accordingly, the Court finds that Dr. Ashenfelter’s methodology, in conjunction with

2

and bolstered by the extensive documentary evidence, is sufficient to meet the predominance

3

standard with respect to impact.

4

d. Conclusion on Impact

5

Plaintiffs’ documentary evidence, along with the expert report and statistical analyses that

6

rely on this evidence, establish that common issues between class members will predominate over

7

individual issues in proving antitrust impact. The documentary evidence and expert report support

8

Plaintiffs’ theory that Defendants had company-wide compensation structures that placed a

9

premium on internal equity. Further, the evidence suggests that Defendants engaged in collusive
communications directly and through the Croner Survey to benchmark their compensation

11
United States District Court
Northern District of California

10

structures against each other. The documentary evidence and expert report also support Plaintiffs’

12

argument that these anti-solicitation agreements and collusive communications resulted in

13

suppressed compensation that affected all class members.

14

4. Damages

15

In addition to disputing whether Plaintiffs can show antitrust impact on a classwide basis,

16

Defendants dispute whether Plaintiffs can show damages on a classwide basis. The U.S. Supreme

17

Court has held that damages “[c]alculations need not be exact, but at the class-certification stage

18

(as at trial), any model supporting a ‘plaintiff’s damages case must be consistent with its liability

19

case, particularly with respect to the alleged anticompetitive effect of the violation.’” Comcast,

20

133 S. Ct. at 1433 (citation omitted) (citing ABA Section of Antitrust Law, Proving Antitrust

21

Damages: Legal and Economics Issues 57, 62 (2d ed. 2010)). In other words, “a damages suit

22

cannot be certified to proceed as a class action unless the damages sought are the result of the

23

class-wide injury that the suit alleges.” Butler, 727 F.3d at 799 (emphasis in original).

24

Here, Plaintiffs again rely on their expert, Dr. Ashenfelter, to demonstrate that they can use

25

reliable methods to compute damages by applying classwide methods and analyses. See Mot. at

26

25. Dr. Ashenfelter concluded that common evidence and a regression analysis could be used to

27

create a model for quantifying the estimated cost to class members resulting from Defendants’

28

55
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

challenged conduct. Ashenfelter Report ¶¶ 112–20. This model generated percentages by which

2

Defendants undercompensated class members in each of the conspiracy years. Ashenfelter Reply

3

Report, Appendix B, Table 8R-E. Dr. Ashenfelter estimated the effect of Defendants’ conspiracy

4

by contrasting compensation during the periods when the conspiracy was in effect with

5

compensation after the conspiracy ended. Ashenfelter Report ¶ 112. Dr. Ashenfelter’s model

6

incorporated a range of variables designed to account for factors including: (1) age, years at the

7

company, and previous year’s compensation; (2) the effects on compensation caused by the

8

alleged conspiracy; (3) the effects caused by factors specific to each Defendant (e.g., total number

9

of employees, annual worldwide box office revenue, and the ratio of new hires to total
employees); and (4) macroeconomic effects (e.g., gross domestic product in the U.S. and the

11
United States District Court
Northern District of California

10

consumer price index). Id. ¶¶ 114–18.

12

Dr. Ashenfelter’s model divides the conspiracy into phases, where a new phase begins

13

each time a Defendant joined the growing conspiracy. Id. ¶ 118. The model additionally divides

14

the conspiracy into a new phase for 2009 to mark the start of the DOJ investigations and a new

15

phase for 2010 to mark the dissolution of the conspiracy in 2010. Id. Dr. Ashenfelter used the

16

model to estimate the average or net under-compensation across Defendants for each phase of the

17

conspiracy. Id. ¶ 119. Dr. Ashenfelter’s model accounted for variations in the effect of the

18

conspiracy over time and for variations among class members and Defendants. Id. ¶¶ 112–19. Dr.

19

Ashenfelter contends that the under-compensation percentages from this model can be used in a

20

straightforward formulaic fashion in conjunction with Defendants’ compensation data to calculate

21

damages for each class member. Id. ¶ 120.

22

Dr. Ashenfelter’s model finds that average under-compensation ranged from 5% to 31%

23

per phase of the conspiracy. Ashenfelter Reply Report, Appendix B, Table 8R-E. Applying the

24

yearly under-compensation to Defendants’ compensation data, Dr. Ashenfelter calculates that total

25

lost earnings attributable to the alleged conspiracy from 2001 to 2010 are $646,074,318. Id.

26
27
28

Dr. Keeley raises a number of criticisms of Dr. Ashenfelter’s damages model to which Dr.
Ashenfelter responds in his Reply Report. The Court addresses below each of Dr. Keeley’s
56
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

criticisms and Dr. Ashenfelter’s responses. As with antitrust impact, however, the Court finds that

2

Dr. Ashenfelter has persuasively responded to each of Dr. Keeley’s criticisms.

3

First, Dr. Keeley argues without citation to the economic literature that it was inappropriate

4

for Dr. Ashenfelter to perform a during/after damages regression that used only the years after

5

Plaintiffs allege the conspiracy ended as a benchmark for estimating the effects of the alleged

6

conspiracy. Keeley Report ¶¶ 81–84. Dr. Keeley argues that Dr. Ashenfelter should have included

7

data prior to the start of the alleged conspiracy in the benchmark period. Id. Dr. Keeley contends

8

that including pre-conspiracy data in the benchmark period eliminates evidence of

9

overcompensation for four years—2003, 2005, 2006, and 2010. Id. ¶ 84.

10

In reply, Dr. Ashenfelter notes that, contrary to Dr. Keeley’s unsupported assertion, the

United States District Court
Northern District of California

11

economic literature supports the use of a during/after damages regression model. Ashenfelter

12

Reply Report ¶¶ 4–5. Dr. Ashenfelter’s citation to the economic literature shows that where, as

13

here, the precise start of the conspiracy is in doubt, the post-conspiracy period is “[m]ore

14

commonly used” as the benchmark period because “the ending of the conspiracy is usually a fairly

15

dramatic event.” Id. ¶ 5 (citing Finkelstein, Michael O., and Hans Levenbach, “Regression

16

Estimates of Damages in Price-Fixing Cases,” Law and Contemporary Problems, 46(4) p. 145–69

17

(1983)). Thus, because Plaintiffs in the instant case allege that the conspiracy began sometime in

18

the 1980s whereas Defendants produced employment data beginning only in 2001, the post-

19

conspiracy period is the more reliable benchmark period according to the economic literature. Id.

20

¶ 6. Furthermore, Dr. Ashenfelter in his Reply Report conducted his regression model including

21

Dr. Keeley’s supposed pre-conspiracy data as a separate phase, and the inclusion of this data did

22

not alter Dr. Ashenfelter’s conclusion that the conspiracy resulted in statistically significant under-

23

compensation from 2001–2010. Id. ¶ 7.

24

Second, Dr. Keeley argues that Dr. Ashenfelter’s damages regression model cannot show

25

that compensation was suppressed during the alleged conspiracy because average compensation at

26

each Defendant did not increase after the conspiracy ended. Keeley Report ¶¶ 57–65. Dr. Keeley’s

27

measure of average compensation at each Defendant is the sum of compensation for all class

28

57
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

members at each Defendant per year divided by the number of class members at that Defendant.

2

As Dr. Keeley acknowledges, Dr. Keeley’s analysis of average compensation does not account for

3

employees’ job titles, macroeconomic changes, or changes among the employees at each

4

Defendant from year-to-year. Id. ¶ 66. For example, some of the employees at issue in the instant

5

case had relatively short employment durations of 1 year or less, such that the composition of

6

employees at each Defendant, and the job titles and pay grades of those employees, could change

7

between years. Id. ¶¶ 107–08. By contrast, Dr. Ashenfelter’s regression model controls for

8

macroeconomic variables, such as GDP, and employee-specific variables, such as age and tenure

9

with the Defendant. Accounting for these variables, Dr. Ashenfelter’s regression model finds that
Defendants undercompensated class members during the alleged conspiracy. Ashenfelter Reply

11
United States District Court
Northern District of California

10

Report, Appendix B, Table 7R.

12

Third, Dr. Keeley argues that instead of estimating the effects of the alleged conspiracy by

13

phase for all Defendants together, Dr. Ashenfelter should have either separately estimated the

14

effects of the alleged conspiracy on each Defendant or estimated a single effect for the entire

15

conspiracy. Keeley Report ¶¶ 66–68. Dr. Keeley argues that making either of these modifications

16

to Dr. Ashenfelter’s model mostly or entirely eliminates the under-compensation found by Dr.

17

Ashenfelter. Id.

18

Dr. Ashenfelter responds that Dr. Keeley’s variations on the model are inappropriate

19

because they aggregate across phases of the conspiracy. Ashenfelter Reply Report ¶¶ 17–20.

20

Aggregating across phases of the conspiracy, Dr. Ashenfelter argues, is inappropriate because the

21

effectiveness of the conspiracy varied by phase. Id. ¶ 17. Dr. Ashenfelter additionally conducted

22

several analyses disaggregating by Defendant and by phase of the conspiracy. Id. ¶ 19. These

23

analyses show that Dr. Ashenfelter’s model produces consistent results when disaggregated by

24

Defendant and by phase, but that Dr. Keeley’s proposed variation on the model does not produce

25

consistent results when similarly disaggregated. Id. Thus, the Court finds Dr. Keeley’s analysis

26

purporting to undermine Dr. Ashenfelter’s results unpersuasive.

27
28

Fourth, Dr. Keeley argues that Dr. Ashenfelter’s model is unreliable because it finds less
58
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

under-compensation as more Defendants joined the conspiracy. Keeley Report ¶¶ 69–72. Dr.

2

Keeley argues that increasing the number of alleged conspirators should instead result in increased

3

levels of under-compensation. Id. ¶ 69. However, this does not alter the fact that Dr. Ashenfelter’s

4

model finds statistically significant levels of under-compensation for each year of the alleged

5

conspiracy. Ashenfelter Reply Report, Appendix B, Table 7R; see also Plaintiffs’ Supp. Br. at 3.

6

In other words, Dr. Keeley’s critique based on the amount of under-compensation per year does

7

not undermine the conclusion that there was under-compensation each year of the alleged

8

conspiracy. The fact that Dr. Ashenfelter’s model shows statistically significant under-

9

compensation each year provides persuasive support for the conclusion that Defendants were

10

effective at suppressing compensation throughout the conspiracy.

United States District Court
Northern District of California

11

Furthermore, Dr. Ashenfelter argues that the economic literature does not require that the

12

effects of the conspiracy remain constant or increase over time, and Dr. Keeley has not provided

13

any citation to the contrary. Ashenfelter Reply Report ¶¶ 12–13. Instead, many factors other than

14

the number of conspirators could contribute to the changing effectiveness of the conspiracy over

15

time. Id. ¶¶ 12–16. For example, the economic literature shows that business cycle fluctuations

16

can lead to varying levels of adherence to the conspiracy and “cheating” by Defendants, which

17

would in turn result in varying levels of effectiveness. Id. ¶ 13–14. Dr. Ashenfelter argues that Dr.

18

Keeley in fact relied upon evidence that Defendants at times engaged in “cheating” on the

19

conspiracy by engaging in cold calling despite the anti-solicitation agreements. Id. ¶ 14; Keeley

20

Report ¶ 131 (listing six email chains describing cold calling by recruiters from Sony, Lucasfilm,

21

and DreamWorks). The Court finds persuasive Dr. Ashenfelter’s explanation, based on the

22

economic literature, that the alleged conspiracy would not necessarily lead to increasing levels of

23

under-compensation as more Defendants joined the conspiracy.

24

Fifth, Dr. Keeley argues that Dr. Ashenfelter should have controlled for changes in

25

Disney’s market power caused by Disney’s participation in ImageMovers Digital as a joint

26

venture and Disney’s acquisition of Pixar and Lucasfilm. Keeley Report ¶ 73. Dr. Keeley

27

conducted a variation of Dr. Ashenfelter’s regression analysis using a purported measure of

28

59
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

market concentration among Defendants that assumed that Defendants consisted of the entire

2

market for animation workers and found that including this variable undermines Dr. Ashenfelter’s

3

finding of under-compensation. Id. ¶¶ 75–76.

4

Dr. Ashenfelter responds that Dr. Keeley’s measure of market concentration is

5

inappropriate because it incorrectly assumes that Defendants are the only employers of animation

6

workers. Ashenfelter Reply Report ¶¶ 24–28. In order to more appropriately address Dr. Keeley’s

7

concern regarding the Disney acquisitions, Dr. Ashenfelter conducted a variation on his regression

8

analysis identifying all employees of companies brought into Disney’s corporate structure as

9

Disney employees following acquisition by Disney. Id. ¶ 29. Unlike Dr. Keeley’s proposed
analysis, Dr. Ashenfelter’s analysis does not require defining the global market for animation

11
United States District Court
Northern District of California

10

workers, so Dr. Ashenfelter’s analysis does not improperly assume that Defendants are the sole

12

employers of animation workers. Id. ¶ 26. Dr. Ashenfelter’s analysis shows that Dr. Ashenfelter’s

13

finding of statistically significant under-compensation is unchanged when modified to control for

14

the Disney acquisitions. Id. Table 5. In light of the fact that Dr. Ashenfelter’s conclusions are

15

unchanged by controlling for the Disney acquisitions, the Court concludes that Dr. Keeley’s

16

criticism that Dr. Ashenfelter should account for those acquisitions does not undermine Dr.

17

Ashenfelter’s analysis and conclusions.

18

Sixth, Dr. Keeley argues that 2010 should have been included in the benchmark period

19

because documentary evidence suggests that the conspiracy may have ended in 2009 once the DOJ

20

began its investigations. Keeley Report ¶ 78. Dr. Keeley finds that including 2010 in the

21

benchmark period results in a reduction of the amount of under-compensation. Id. ¶ 80.

22

Dr. Ashenfelter argues, and the Court agrees, that it is appropriate to exclude 2010 from

23

the benchmark period. In fact, Dr. Keeley conceded at his deposition that the effects of the

24

conspiracy could have persisted in 2010. Talge Reply Decl., Exh. 8 at 60:4–9. Moreover,

25

documentary evidence, such as the fact that the Croner Survey was not altered to comply with the

26

DOJ’s requirements until 2011, supports Plaintiffs’ allegation that certain aspects of the

27

conspiracy may have extended into 2010. Ashenfelter Reply Report ¶ 9. Additionally, Dr.

28

60
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

Ashenfelter’s conclusion that Defendants exhibited rigid wage structures means that the effects of

2

the conspiracy would continue for some time after the end of the conspiracy. Id. Furthermore, Dr.

3

Ashenfelter argues that the economic literature, including the American Bar Association’s

4

Econometrics Handbook, recognize that the effects of anticompetitive conduct may last beyond

5

the time the conduct in question ends. Id. ¶ 10. Thus, because Plaintiffs’ evidence tends to show

6

that the conspiracy may have continued into 2010 and because economic theory indicates that the

7

effects of the conspiracy may have extended into 2010 even if the anticompetitive conduct ceased

8

prior to 2010, 2010 is inappropriate for inclusion in the post-conspiracy benchmark period. Id.

9

¶¶ 9–10. However, Dr. Ashenfelter controlled for the fact that the conspiracy may not have
continued into 2010 by considering 2010 as a separate phase of the conspiracy, such that the data

11
United States District Court
Northern District of California

10

for 2010 would not alter Dr. Ashenfelter’s conclusions regarding the effectiveness of the

12

conspiracy prior to 2010. Id. ¶ 11.

13

Seventh, Dr. Keeley criticizes Dr. Ashenfelter for failing to control for the effects of the

14

High-Tech conspiracy. Keeley Report ¶¶ 85–86. Notably, however, Dr. Keeley does not actually

15

propose any method to control for the High-Tech conspiracy. Id. Furthermore, there are significant

16

differences between the proposed class members in the instant case and the High-Tech class

17

members. Ashenfelter Reply Report ¶ 31. For example, the High-Tech class consisted of salaried

18

employees “who work in the technical, creative, and/or research and development fields” for

19

Apple, Adobe, Google, Intel, Intuit, Lucasfilm, and Pixar. High-Tech, 985 F. Supp. 2d at 1177.

20

Representative job titles for the High-Tech class included “Hardware Engineers and Component

21

Designers,” “User Interface or User Experience Designers,” and “Quality Analysts.” Id. By

22

contrast, the proposed class in the instant case consists of “animation and visual effects

23

employees” employed by Pixar, Lucasfilm, DreamWorks, Disney, Blue Sky, Two Pic

24

(ImageMovers Digital), and the Sony Defendants. Mot. at v. Representative job titles include

25

“Layout Artist,” “Costumer,” “Designer,” and “Senior Effects Animator.” Ashenfelter Reply

26

Report, Amended Appendix C. As Dr. Keeley acknowledges, the different class definitions

27

between High-Tech and the instant case result largely in two qualitatively distinct classes. Keeley

28

61
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

Report ¶¶ 124–27. Moreover, to the extent Dr. Keeley’s argument is that any under-compensation

2

to the class members in the instant case was caused by the High-Tech conspiracy and not by any

3

wrongdoing by Defendants, that argument is undermined by the substantial evidence proffered by

4

Plaintiffs that tends to show that Defendants in the instant case colluded to suppress class

5

members’ compensation, and that Defendants’ collusion had a classwide impact. See supra,

6

Sections IV.B.2–3. Dr. Keeley’s criticism of Dr. Ashenfelter’s model on the grounds that it does

7

not adequately control for effects of the High-Tech conspiracy is thus unpersuasive.

8

Finally, the Keeley Report additionally challenges the number of class members who were
under-compensated according to Dr. Ashenfelter’s damages regression. Keeley Report ¶¶ 87–91.

10

In the Court’s order requesting supplemental briefing, the Court questioned Defendants regarding

11
United States District Court
Northern District of California

9

three methodological flaws in Dr. Keeley’s analysis of the number of under-compensated class

12

members. ECF No. 270. Subsequently, Defendants acknowledged that Dr. Keeley’s calculation

13

was flawed and withdrew that portion of the Keeley Report. See Defendants’ Supp. Br. at 2.

14

In sum, considering the fact that Dr. Ashenfelter’s damages model produces consistent

15

results when modified to respond to Dr. Keeley’s concerns combined with the support for Dr.

16

Ashenfelter’s methodology in the economic literature, the Court concludes that Dr. Keeley’s

17

criticisms of Dr. Ashenfelter’s damages model are unpersuasive. Accordingly, the Court finds that

18

Dr. Ashenfelter’s damages model is capable of calculating classwide damages and thus satisfies

19

the predominance standard.

20

5. Fraudulent Concealment and the Statute of Limitations

21

Defendants also argue that class certification is inappropriate because individual statute of

22

limitations issues will predominate over common questions. The Court “previously concluded that

23

Plaintiffs’ claims are time barred under the relevant statutes of limitations unless Plaintiffs

24

adequately allege either a continuing violations theory or a fraudulent concealment theory” to toll

25

the statutes of limitations. In re Animation Workers Antitrust Litig., 123 F. Supp. 3d at 1193 (order

26

denying Defendants’ motion to dismiss on statute of limitations grounds). Although Plaintiffs did

27

not advance a continuing violations theory in the SAC, Plaintiffs adequately alleged fraudulent

28

62
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

concealment. Id. at 1193–94. Accordingly, Plaintiffs must prove fraudulent concealment to

2

recover.

3

To establish fraudulent concealment, Plaintiffs must show that (1) “the defendant took

4

affirmative acts to mislead the plaintiff”; (2) the plaintiff did not have “actual or constructive

5

knowledge of the facts giving rise to its claim”; and (3) “the plaintiff acted diligently in trying to

6

uncover the facts giving rise to its claim.” Id. at 1194 (citing Hexcel Corp. v. Ienos Polymers, Inc.,

7

681 F.3d 1055, 1060 (9th Cir. 2012)).

8
9

Defendants focus on the second element of fraudulent concealment and argue that
“[a]mong the class members, including the named plaintiffs, the challenged agreements were
widely known or believed to exist.” Opp. at 2. Defendants thus argue that individualized inquiries

11
United States District Court
Northern District of California

10

into whether individual plaintiffs had constructive or actual knowledge (thereby defeating

12

fraudulent concealment) will predominate over common issues. Plaintiffs respond that “courts

13

have been nearly unanimous . . . in holding that possible differences in the application of a statute

14

of limitations to individual class members, including the named plaintiffs, does not preclude

15

certification of a class action.” Reply at 1–2 (alteration in original) (quoting Schramm v.

16

JPMorgan Chase Bank, N.A., No. 09-CV-09442-JAK, 2011 WL 5034663, at *10 (C.D. Cal. Oct.

17

19, 2011)). Plaintiffs further contend that fraudulent concealment issues will be amenable to

18

classwide treatment, and that any remaining individual issues will be few in number and may be

19

dealt with in separate proceedings once common issues are resolved.

20
21
22

The Court first addresses whether a class may be certified where fraudulent concealment
must be proven to overcome a statute of limitations, and then turns to the facts of this case.
a. Whether a Class May Be Certified When Fraudulent Concealment Must Be
Established

23
First, it is clear as a general matter that a statute of limitations defense does not
24
automatically preclude certification where common questions otherwise predominate. The Ninth
25
26

Circuit has repeatedly held that “[t]he existence of a statute of limitations issue does not compel a
finding that individual issues predominate over common ones.” Williams v. Sinclair, 529 F.2d

27
28

63
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

1383, 1388 (9th Cir. 1975); see also Cameron v. E.M. Adams & Co., 547 F.2d 473, 478 (9th Cir.

2

1976) (“We hold that the presence of individual issues of compliance with the statute of

3

limitations here does not defeat the predominance of the common questions.”). However, the

4

parties do not cite, and the Court has not located, any case in which the Ninth Circuit has

5

considered the more specific question of whether certification is permissible when the plaintiff

6

class must prove fraudulent concealment to overcome the statute of limitations.

7

Although the Ninth Circuit has not yet answered the precise question posed in the instant
case, the weight of authority is that there is no per se bar to class proceedings where fraudulent

9

concealment must be shown. In Cameron, for example, the Ninth Circuit considered whether a

10

class-action securities fraud claim could proceed in light of an Oregon fraud statute with a two-

11
United States District Court
Northern District of California

8

year statute of limitation. Cameron, 547 F.2d at 477–78. The statute was tolled by the discovery

12

rule, meaning that the two-year period began to run when “a class member discovered, or in the

13

exercise of reasonable diligence should have discovered, the alleged deceit.” Id. at 478. In a

14

manner analogous to Defendants’ argument here, the defendants in Cameron argued “that the

15

individual issues of when each member of the classes discovered, or should have discovered, the

16

alleged omissions prevents the common issues from predominating.” Id. The Ninth Circuit

17

rejected that argument, holding that “even if there [existed] questions of individual compliance

18

with the Oregon statute of limitations, they are not sufficient, on balance, to negate the

19

predominance of the common issues.” Id.

20

Cameron, in turn, relied upon the Ninth Circuit’s decision in Williams, which involved a

21

similar argument to that brought under the same Oregon statute at issue in Cameron. Williams,

22

529 F.2d at 1387–88. In Williams, the district court found that common issues would not

23

predominate in a securities-fraud class action because “the existence of a statute of limitations

24

defense made necessary separate determinations of the date when each plaintiff discovered or in

25

the exercise of reasonable diligence should have discovered the alleged fraud.” Id. at 1388. The

26

Ninth Circuit reversed, holding that although individual issues regarding the statute of limitation

27

might arise, those issues did not prevent certification “[g]iven a sufficient nucleus of common

28

64
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1
2

questions.” Id. at 1388
Although Cameron and Williams concerned a discovery-rule issue, rather than fraudulent

3

concealment, those cases are nevertheless instructive as both held that a class action could proceed

4

despite the potential for individualized inquiries as to what class members knew or should have

5

known. Thus, the holdings of Cameron and Williams counsel in favor of finding that the issue of

6

fraudulent concealment does not prevent certifying the class in the instant case.

7

While acknowledging Williams and Cameron, Defendants rely upon Broussard v. Meineke
Disc. Muffler Shops, Inc., 155 F.3d 331, 342 (4th Cir. 1998), a case involving the reversal of a

9

$390 million class-action judgment on multiple grounds, to argue that the need to engage in a

10

tolling analysis precludes class certification because a tolling analysis requires “individualized

11
United States District Court
Northern District of California

8

inquiry into what each [plaintiff] knew about [defendant’s conduct] and when he knew it.” Opp. at

12

3 (quoting Broussard, 155 F.3d at 342). However, to the extent Broussard suggests that a tolling

13

analysis always defeats class certification, it conflicts with the Ninth Circuit’s holdings in

14

Williams and Cameron, which permitted class action claims to proceed despite the existence of

15

such individualized inquiries. Moreover, several other Courts of Appeal have directly rejected

16

Broussard’s reasoning in cases involving fraudulent concealment. As explained by the First and

17

Third Circuits:

18
19
20
21
22
23
24

Although a necessity for individualized statute-of-limitations
determinations invariably weighs against class certification under
Rule 23(b)(3), we reject any per se rule that treats the presence of
such issues as an automatic disqualifier. In other words, the mere
fact that such concerns may arise and may affect different class
members differently does not compel a finding that individual issues
predominate over common ones. As long as a sufficient
constellation of common issues binds class members together,
variations in the sources and application of statutes of limitations
will not automatically foreclose class certification under Rule
23(b)(3). . . . Predominance under Rule 23(b)(3) cannot be reduced
to a mechanical, single-issue test.
In re Linerboard Antitrust Litig., 305 F.3d 145, 162–63 (3d Cir. 2002) (quoting Waste Mgmt.

25
Holdings, Inc. v. Mowbray, 208 F.3d 288, 296 (1st Cir. 2000)). The Second Circuit likewise
26
declines to apply a per se rule. See, e.g., In re U.S. Foodservice Inc. Pricing Litig., 729 F.3d 108,
27
28

65
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

129 (2d Cir. 2013) (affirming class certification and noting that “fraudulent concealment issues

2

may sometimes preclude certification under Rule 23(b)(3), but they do not do so here”). Thus, to

3

the extent Broussard is read to require the categorical rejection of class actions where fraudulent

4

concealment is at issue, it sets forth a minority position that is incompatible with First, Second,

5

Third, and Ninth Circuit authority concerning the treatment of statute of limitations issues.

6

Because the Court concludes that no per se rule precludes certification, the question is a

7

fact-specific determination of whether common issues concerning fraudulent concealment, as

8

opposed to individual inquiries, will predominate in this case. See In re Linerboard Antitrust

9

Litig., 305 F.3d at 160–61. To answer this question, the Court undertakes a pragmatic inquiry into
the types of proof the parties will use to prove (or disprove) the elements of fraudulent

11
United States District Court
Northern District of California

10

concealment. See Backhaut v. Apple Inc., No. 14-CV-02285-LHK, 2015 WL 4776427, at *13

12

(N.D. Cal. Aug. 13, 2015) (noting that the predominance inquiry is ultimately a pragmatic and

13

holistic one) ; Wright, Miller, & Kane, 7AA Federal Practice and Procedure, § 1778 (3d ed.

14

2005) (noting that “the proper standard under Rule 23(b)(3) is a pragmatic one, which is in

15

keeping with the basic objectives of the Rule 23(b)(3) class action” (footnote omitted)).

16
17

b. Evidence Regarding Fraudulent Concealment
As noted above, to prove fraudulent concealment, Plaintiffs must first establish that

18

Defendants took affirmative acts to mislead the class and thereby concealed the alleged

19

conspiracy. See In re Animation Workers Antitrust Litig., 123 F. Supp. 3d at 1193. As the Third

20

Circuit has observed, “[i]t generally has been recognized that the question of concealment by [an]

21

antitrust defendant is a common question, subject to being uniformly resolved on behalf of all

22

members of the class.” In re Linerboard Antitrust Litig., 305 F.3d at 160 (citing In re Flat Glass

23

Antitrust Litig., 191 F.R.D. 472, 487 (W.D. Pa. 1999)). Similarly, as another court in the Northern

24

District of California has stated, “the critical inquiry will be whether defendants successfully

25

concealed the existence of the alleged conspiracy, which proof will be common among the class

26

members in each class.” In re TFT–LCD I, 267 F.R.D. at 310 (quotation marks omitted).

27
28

The Court agrees that the question of concealment is a common question in the instant
66
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

case. Here, the Plaintiffs have established that concealment will turn primarily on common proof

2

regarding Defendants’ actions, rather than proof regarding individual class members. Beyond

3

Plaintiffs’ argument that Defendants “misled their employees in public filings with the Securities

4

and Exchange Commission (SEC), wherein Defendants claimed to be in compliance with all

5

applicable competition laws,” see SAC ¶¶ 159–164, Plaintiffs have identified a number of

6

additional specific sources of common proof upon which they will rely. See Reply at 2–3. For

7

example, Plaintiffs argue that they will present common proof of Defendants’ allegedly secret

8

meetings to collude on compensation before and after Croner Survey events. See, e.g., Talge

9

Reply Decl., Exh. 18 (email from DreamWorks’ Head of Human Resources Kathy Mandato to

10

other Defendants’ human resources executives arranging a meeting and stating that the Croner

11
United States District Court
Northern District of California

1

Survey “presents an opportunity for an intimate group of us to get together”).

12

Plaintiffs also contend that they will rely on Defendants’ corporate codes of conduct. For

13

example, Plaintiffs cite DreamWorks’ companywide “Code of Business Conduct and Ethics,”

14

which Plaintiffs contend all employees were required to acknowledge, and which states that “Our

15

policy is to compete vigorously . . . and to do so at all times in compliance with all applicable

16

antitrust . . . laws.” Talge Reply Decl., Exh. 13. Plaintiffs also identify the deposition testimony of

17

DreamWorks’ President Ann Daly, including the following exchange: “Q: Would it violate the

18

code of conduct to agree with a competitor not to hire each other’s employees? [A:] I believe that

19

it would, because it would — you know, I don’t know that that gives us the best opportunity to,

20

you know, bring in the best people, you know, and I think that makes us really competitive to do

21

so.” Talge Reply Decl., Exh. 1.

22

Another source of common proof identified by Plaintiffs is a 2007 “talking points” memo

23

prepared by Pixar, apparently to help its managers “prepare . . . for your conversations with your

24

employees about their 2007 increases.” Talge Reply Decl., Exh. 12. Plaintiffs contend that Pixar’s

25

talking points memo falsely stated that Pixar’s “goal” was to keep salaries “competitive.” Id.

26

This evidence focuses on the actions of the Defendants in concealing the alleged

27

conspiracy. Absent class certification, such evidence would be offered repetitively in individual

28

67
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

trials by individual class member plaintiffs, thus underscoring its nature as common proof.

2

Because “[t]hese allegations of proof are all common to the defendants,” In re Linerboard

3

Antitrust Litig., 305 F.3d at 163, the Court agrees that with respect to the first element of

4

fraudulent concealment, common questions will predominate over any individual issues.

5

While Plaintiffs focus on the first element of fraudulent concealment, Defendants instead

6

focus their efforts on arguing that the second element of fraudulent concealment—that Plaintiffs

7

lacked constructive or actual knowledge—will turn primarily on individualized proof. Opp. at 5–6.

8

In this regard, Defendants cite evidence that they argue demonstrates that class members had

9

constructive knowledge of the alleged conspiracy, or at least suggests that individualized inquiries

10

United States District Court
Northern District of California

11

are necessary to determine whether that is the case. Opp. at 5–9.
Defendants argue, for example, that “some Defendants explicitly told many class members

12

the same type of information that Plaintiffs contend is proof of conspiracy.” Opp. at 5. As

13

examples, Defendants cite (1) a Pixar executive’s statement to a roomful of interns that Pixar has

14

“an anti-poach clause between the Lucas companies and this company,” which was recorded and

15

posted as a video on Pixar’s internal network, see ECF No. 239-8 (“Klaidman Decl.”) ¶ 6; (2) a

16

town hall meeting where Pixar’s Ed Catmull “was publicly reported to have stated” to “hundreds

17

of Disney employees” that he had made a non-poaching agreement with other studios, Opp. at 5;

18

Lannin Decl., Exh. 26; and (3) that a lawyer who has “negotiated over 700 employment contracts

19

with all the major animation studios” was told by a Defendant that Disney and Pixar had an anti-

20

solicitation agreement, Opp. 5. Similarly, as noted above, Defendants argue that “thousands of

21

class members are also members of unions . . . which serve as information exchanges.” Opp. at 8.

22

Defendants also cite examples of Internet postings that could demonstrate constructive knowledge,

23

which Defendants characterize as “public statements about no-poach agreements before the DOJ

24

investigation.” Opp. at 8 (emphasis in original). These public statements include a 2008 comment

25

by “CG Guy” on the Animation Guild’s blog that “I know for a fact that several feature studios

26

have made an unofficial agreement not to poach employees (something that would drive wages

27

up). It stands to reason they have gone several steps further.” Lannin Decl., Exh. 53.

28

68
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

Although the Court limits its consideration of the merits of Defendants’ evidence to the

2

extent necessary to determine whether class certification is appropriate, Amgen, 133 S.Ct. at 1195,

3

the Court notes that Plaintiffs argue that Defendants’ evidence is subject to multiple

4

interpretations. For example, the video Pixar references was edited to delete the “anti-poach”

5

statement three months later and reposted to the Pixar network. See Klaidman Decl. ¶ 7. The fact

6

that Pixar edited the video to delete the discussion of anti-solicitation agreements could be

7

interpreted as an affirmative act of concealment. Moreover, the record does not appear to contain

8

evidence that any class member actually viewed the unedited video. Similarly, the citation

9

describing what Pixar’s Ed Catmull was “publicly reported to have stated” appears to be to an
anonymous blog post, dated four years after the town hall meeting occurred. Lannin Decl.,

11
United States District Court
Northern District of California

10

Exh. 26.

12

Moreover, a common thread through much of Defendants’ proffered evidence, however, is

13

that it is generalized evidence applicable to wide swaths of the class, rather than evidence truly

14

specific to individual class members. For example, the cited blog posts would have been available

15

to the entire class after the date of posting; information circulated through unions would have been

16

available to “thousands” of class members; the Disney town hall meeting upon which Defendants

17

rely would have affected “hundreds” of class members; the lawyer referenced by Defendants is

18

alleged to have worked with hundreds of individuals; and the Pixar video was widely available to

19

Pixar class members on Pixar’s network.

20

Given that constructive knowledge requires an objective inquiry as to whether “facts exist

21

that would excite the inquiry of a reasonable person,” Conmar Corp. v. Mitsui & Co. (U.S.A.), 858

22

F.2d 499, 504 (9th Cir. 1988), whether the common evidence that Defendants put forth—a portion

23

of which Defendants describe as demonstrating “widespread dissemination of information through

24

blogs, town hall meetings, and e-mails,” Opp. at 8—should have put the class on notice is

25

amenable to classwide determination. See In re Monumental Life Ins. Co., 365 F.3d 408, 421 (5th

26

Cir. 2004) (holding, in evaluating whether to certify a proposed Rule 23(b)(2) class, that the court

27

had “no difficulty concluding that whether plaintiffs were provided constructive notice is an issue

28

69
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

that can be decided on a classwide basis” where constructive notice was alleged based on

2

widespread publicity of facts constituting injury). Similarly, whether this evidence was sufficient

3

to trigger Plaintiffs’ duty to diligently investigate their claims will raise largely common issues.

4

See Conmar, 858 F.2d at 504 (“The requirement of diligence is only meaningful, however, when

5

facts exist that would excite the inquiry of a reasonable person.”).

6

Although much of Defendants’ evidence is thus common and suggests that the remaining

7

elements of fraudulent concealment will be amenable to classwide resolution, Defendants also cite

8

evidence in the form of emails to and from class members referencing anti-solicitation agreements.

9

For example, Defendants cite an email in which a Disney employee informed a Pixar recruiter that
he had been told of a “‘handshake agreement’ between Disney and Pixar that prohibits poaching

11
United States District Court
Northern District of California

10

talent.” Lannin Decl., Ex. 3. Similarly, Defendants cite a Yahoo Groups email, apparently sent to a

12

list including named Plaintiff Nitsch, that referenced an agreement between Defendants and asked

13

“how illegal is this?” Lannin Decl., Exh. 36. For his part, Nitsch denies recollection of receiving

14

the Yahoo Groups e-mail. Talge Reply Decl., Exh. 7 at 117:9–13. Defendants contend that these

15

types of emails demonstrate constructive, if not actual, knowledge by the individuals involved.

16

It is true that some emails identified by Defendants represent more individualized evidence

17

than the blog posts, town hall meetings, and group emails upon which Defendants also rely. Such

18

limited individualized evidence may warrant inquiries as to whether particular class members had

19

constructive or actual knowledge, or exercised appropriate diligence if required to do so. As

20

Plaintiffs point out, however, the number of class members for whom Defendants identify such

21

evidence is small (by Plaintiffs’ count, approximately 63 members, of a class exceeding 10,000

22

members), despite the fact that evidence such as the proffered emails between class members and

23

Defendants are presumably within Defendants’ control. Opp. at 7; May 6 Trans. at 17:10–18. The

24

precise source of Plaintiffs’ estimate of 63 individuals is unclear from the record, but is roughly

25

comparable to the approximately 40 individual class members identified by name in the

26

declarations submitted with Defendants’ Opposition.

27
28

In light of the substantial common evidence otherwise at issue with regard to the second
70
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

element of fraudulent concealment, the Court concludes that the relatively small number of

2

individual inquiries which might be required do not defeat predominance. See In re Linerboard

3

Antitrust Litig., 305 F.3d at 161 n.13 & 163 (noting that “[m]any courts faced with similar

4

circumstances have certified class status with the expectation that individual questions concerning

5

fraudulent concealment can be resolved at a later damages phase” and approving class certification

6

despite named plaintiff’s testimony that she was “‘sure’ that something illegal was afoot.”); In re

7

Conseco Life Ins. Co. LifeTrend Ins. Sales & Mktg. Litig., 270 F.R.D. 521, 530 n.8 (N.D. Cal.

8

2010) (“Courts have held, however, that individual issues relating to the statute of limitations do

9

not bar certification where there is otherwise a sufficient showing of commonality.”).13
As a final matter, Defendants argue that “a class cannot be certified on the premise that

10

United States District Court
Northern District of California

11

[defendant] will not be entitled to litigate its . . . defenses to individual claims.” Opp. at 2 (quoting

12

Dukes, 564 U.S. at 367) (alterations in original).14 However, Defendants’ concern regarding

13

Defendants’ ability to litigate defenses is misplaced. Certification of the proposed class will not

14

deny Defendants the ability to raise individualized statute of limitations defenses, should the

15

evidence indicate that the question must be resolved separately as to particular class members. In

16

that regard, Plaintiffs propose that such issues, which will likely be small in number, would be

17

manageably addressed in individualized proceedings at a later phase of the case. As the U.S.

18
19
20
21
22
23
24
25
26
27
28

13

Defendants contend that this case is akin to Backhaut, 2015 WL 4776427, at *13–15, where this
Court found that certification of a class was improper, in part, because of the predominance of
individualized inquiries as to whether individual plaintiffs impliedly consented to interception of
their text messages. Defendants’ reliance on Backhaut is inapposite for at least two reasons. First,
in Backhaut the plaintiffs did not offer any response to the defendant’s arguments that the need for
individualized inquiries into the statutory consent defense under the Wiretap Act would render the
case unmanageable. Id. at *15. Second, and more importantly, the statutory defense at issue in
Backhaut required a determination of the presence or absence of actual consent—not constructive
consent. Id. at *14 (“Consent may be explicit or implied, but it must be actual consent rather than
constructive consent.”) (quotation marks omitted). As a result, the inquiry in Backhaut was
necessarily individualized because it focused on the presence or absence of actual consent of each
class member, not on whether the defendant could demonstrate some form of constructive consent.
See id. Here, by contrast, Defendants’ proof is largely directed to constructive knowledge.
14
Defendants’ quotation of Dukes replaces “statutory defenses” with “defenses,” thus divorcing
the quoted language from the statutory context in which it was made. Opp. at 2 (quoting Dukes,
564 U.S. at 367). No such statutory defenses are at issue here.
71
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

Supreme Court has noted, certification may be proper even where “some affirmative defenses

2

peculiar to some individual class members” will “have to be tried separately.” Tyson Foods, Inc. v.

3

Bouaphakeo, 136 S. Ct. 1036, 1045 (2016) (quotation marks omitted); see also In re Linerboard

4

Antitrust Litig., 305 F.3d at 163 (explaining that “individualized facts of fraudulent concealment

5

may be adjudicated in the same fashion and at the same time as individual damages issues”). The

6

Court agrees that any such issues unique to particular class members may be handled in a manner

7

similar to that used when individualized damages inquiries arise in class action proceedings. See

8

Jimenez v. Allstate Ins. Co., 765 F.3d 1161, 1168 (9th Cir. 2014) (affirming certification of a class

9

where “the district court was careful to preserve [the defendant’s] opportunity to raise any

10

United States District Court
Northern District of California

11

individualized defense it might have at the damages phase of the proceeding”).
In summary, as noted above, the overwhelming majority of the evidence upon which

12

Plaintiffs will rely to show concealment is common proof. Similarly, much of the evidence upon

13

which Defendants will rely with regard to constructive knowledge is common proof as well.

14

Taken as a whole, the record indicates that common issues will predominate as to fraudulent

15

concealment. The potential for a small number of individualized inquiries concerning affirmative

16

defenses with regard to some class members does not alter the Court’s conclusion that classwide

17

issues will predominate. See Cameron, 547 F.2d at 478.

18

6. Arbitration and Release of Claims Agreements

19

Defendants also argue that “individualized issues [are] raised by the thousands of releases

20

and arbitration agreements entered by class members.” Opp. at 10. In particular, Defendants

21

estimate that approximately 1,300 Pixar and Lucasfilm employees were members of the High-

22

Tech settlement class, and have released claims against the Disney Defendants. See Opp. at 10;

23

SAC ¶ 196. Defendants also contend that a substantial number of class members have entered into

24

arbitration agreements with DreamWorks or the Sony Defendants, see Defendants’ Supp. Br. at 5,

25

and that some class members have signed releases against particular Defendants in connection

26

with severance agreements. See Opp. at 10 & n.11. In total, Defendants estimate that at least 5,000

27

class members are subject to at least one such agreement or release. Defendants’ Supp. Br. at 5.

28

72
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

However, as noted above, the Sony Defendants have reached a settlement with Plaintiffs.

2

See ECF No. 273. The Sony Defendants estimate that over 1,040 of their employees were subject

3

to an arbitration agreement and over 450 of their employees were subject to a release of claims

4

agreement. See ECF No. 239-16 (Sollow Decl.) ¶¶ 4, 11. As a result of the Sony Defendants’

5

settlement, those arbitration or release of claims agreements are no longer at issue.

6

Although the Court considers the existence of arbitration agreements and releases in

7

determining whether certification is appropriate, “[t]he fact that some members of a putative class

8

may have signed arbitration agreements or released claims against a defendant does not bar class

9

certification.” Herrera, 274 F.R.D. at 681. While many class members may have signed
arbitration and release of claims agreements, the Court concludes that in the instant case these

11
United States District Court
Northern District of California

10

agreements do not raise the type of individualized issues that would preclude class certification.

12

Specifically, joint and several liability defeats Defendants’ argument that release

13

agreements will significantly complicate resolution of the major issues of liability in this case. See

14

In re Animation Workers Antitrust Litig., 123 F. Supp. 3d at 1207 (noting that the antitrust laws

15

impose joint and several liability as among coconspirators). In contrast to a single-defendant case,

16

joint and several liability among Defendants here means that a class member who has signed a

17

release against one Defendant is not precluded from pursuing this action against the other

18

Defendants. As a result, the existence of releases appears to relate primarily to allocation of

19

liability among Defendants, as opposed to liability of the Defendants to the class.
Similarly, arbitration agreements are unlikely to preclude class members from proceeding

20
21

with their claims against other Defendants in the instant case. As noted above, as a result of the

22

Sony Defendants’ settlement, DreamWorks is the only Defendant still asserting arbitration

23

agreements. The Court has already ruled that DreamWorks’ arbitration agreement with named

24

Plaintiff Nitsch does not preclude his claims against other Defendants.15 See In re Animation

25
26
27
28

15

DreamWorks does not appear to contend that the arbitration clauses in its other employment
agreements differ materially from the Nitsch agreement. A declaration from Dan Satterthwaite,
DreamWorks’ Head of Human Resources, states that many other DreamWorks employees
73
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

Workers Antitrust Litig., 100 F. Supp. 3d 851, 867, 870 (N.D. Cal. 2015). Indeed, DreamWorks

2

itself apparently recognizes that its arbitration agreements are unlikely to prevent class members’

3

claims from proceeding against other Defendants, given that DreamWorks did not pursue

4

arbitration with Nitsch even after the Court ruled that Nitsch must arbitrate his claims against

5

DreamWorks based on his employment at DreamWorks.16 See May 6 Trans. at 11:12–12:1.

6

This case is therefore distinguishable from Lozano v. AT&T Wireless Servs., Inc., 504 F.3d

7

718, 728 (9th Cir. 2007), which Defendants cite in support of their argument that arbitration

8

agreements raise individual issues that defeat predominance. In Lozano, the Ninth Circuit affirmed

9

denial of certification based in part on issues relating to the widespread existence of arbitration
agreements between class members and the defendant. Id. Unlike this case, however, Lozano

11
United States District Court
Northern District of California

10

involved a single defendant, meaning that a valid arbitration agreement with that defendant would

12

potentially preclude each class member subject to the arbitration agreement from proceeding in

13

court. See id. at 722. As noted above, that is not the case with the DreamWorks arbitration

14

agreement because there is joint and several liability amongst the multiple Defendants in the

15

instant case. Second, the claims in Lozano would have required the court to determine whether the

16

arbitration clause at issue was unconscionable under the laws of all fifty states. Id. at 728 (“The

17

district court therefore determined that predominance was defeated because [defendant’s] intent to

18

seek arbitration of the class would necessitate a state-by-state review of contract conscionability

19

jurisprudence.”). No such complexity is present here.

20

The Court concludes that the existence of the arbitration and release of claims agreements

21

raised by Defendants do not change the fact that common issues will predominate in this case. To

22

the extent the releases from High-Tech and individual employment agreements preclude recovery

23

by a particular class member against a particular Defendant, those damages allocation issues may

24

be resolved during individualized damages proceedings at a later phase, together with any

25
26
27
28

between 2003 and 2010 signed agreements with “similar clauses.” ECF No. 239-14 ¶ 5.
16
Nitsch also has claims against DreamWorks based on his employment at Sony. In re Animation
Workers Antitrust Litig., 100 F. Supp. 3d at 865.
74
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

individualized issues that remain concerning fraudulent concealment. In re Linerboard Antitrust

2

Litig., 305 F.3d at 163; see also Levya v Medline Indus., Inc., 716 F.3d 510, 513 (9th Cir 2013)

3

(“In this circuit, however, damage calculations alone cannot defeat certification.”) (quotation

4

marks omitted).

5

7. Conclusion Regarding Predominance

6

The Court’s analysis shows that common issues are likely to predominate over individual
issues. Importantly, the Court’s analysis of predominance includes a “qualitative assessment.” See

8

Butler, 727 F.3d at 801. This qualitative assessment includes some analysis into how this case,

9

should it proceed to trial, would actually be litigated. See In re New Motors, 522 F.3d at 20

10

(“Under the predominance inquiry a district court must formulate some prediction as to how

11
United States District Court
Northern District of California

7

specific issues will play out in order to determine whether common or individual issues

12

predominate in a given case.” (quotation marks omitted)).

13

As such, the Court notes that there is no dispute that antitrust violation can be shown using

14

exclusively evidence that is common to the entire class for the reasons discussed above. The Court

15

further finds that antitrust violation is likely to be a central, disputed issue at summary judgment

16

and at trial. See In re Static Random Access Memory (SRAM) Antitrust Litig., 264 F.R.D. 603, 611

17

(N.D. Cal. 2009) (“Plaintiffs need not show that there will be common proof on each element of

18

the claim. ‘In price-fixing cases, courts repeatedly have held that the existence of the conspiracy is

19

the predominant issue and warrants certification even where significant individual issues are

20

present.’” (quoting Thomas & Thomas Rodmakers, Inc. v. Newport Adhesives & Composites, Inc.,

21

209 F.R.D. 159, 167 (C.D. Cal. 2002))); see 6 Newberg on Class Actions § 18.25 (4th ed. 2002)

22

(“[C]ommon liability issues such as conspiracy or monopolization have, almost invariably, been

23

held to predominate over individual issues.”); Wright, Miller, & Kane, 7AA Federal Practice and

24

Procedure, § 1781 (3d ed. 2005) (“[W]hether a conspiracy exists is a common question that is

25

thought to predominate over other issues in the case.”); cf. Cordes & Co. Fin. Servs. v. A.G.

26

Edwards & Sons, Inc., 502 F.3d 91, 108 (2d Cir. 2007) (“Even if the district court concludes that

27

the issue of injury-in-fact presents individual questions, however, it does not necessarily follow

28

75
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

that they predominate over common ones and that class action treatment is therefore

2

unwarranted.”). As a result, the voluminous classwide proof of antitrust violation weighs in favor

3

of a finding that common questions predominate.
In addition to concluding that common questions will predominate with respect to the

4
5

central element of antitrust violation, the Court, having conducted a rigorous analysis, also finds

6

that common questions will predominate over individual questions with respect to antitrust impact.

7

The extensive documentary evidence suggests that Defendants maintained a formal wage structure

8

and used internal equity as a factor in determining compensation. This suggests that the anti-

9

solicitation agreements and collusion over compensation policies had a structural impact on class
members’ compensation. Furthermore, the Court, having carefully reviewed the parties’ expert

11
United States District Court
Northern District of California

10

reports, concludes that Plaintiffs have presented a methodology that supports a finding that

12

Plaintiffs will use evidence common to the class to demonstrate antitrust impact. Additionally, the

13

Court finds that Plaintiffs have set forth a methodology for calculating damages on a classwide

14

basis.

15

Finally, the Court concludes that common evidence will predominate the fraudulent

16

concealment analysis. Any individual issues that arise with respect to fraudulent concealment, or

17

arbitration and release of claims agreements, will be outweighed by the common issues and

18

classwide evidence that will predominate at trial.

19
20

C. Rule 23(b)(3): Superiority
Rule 23(b)(3) also tests whether “a class action is superior to other available methods for

21

fairly and efficiently adjudicating the controversy.” Fed. R. Civ. P. 23(b)(3). Under Rule 23(b)(3),

22

the Court must consider four non-exclusive factors in evaluating whether a class action is a

23

superior method of adjudicating plaintiffs’ claims: (1) the interest of each class member in

24

individually controlling the prosecution or defense of separate actions; (2) the extent and nature of

25

any litigation concerning the controversy already commenced by or against the class; (3) the

26

desirability of concentrating the litigation of the claims in the particular forum; and (4) the

27

difficulties likely to be encountered in the management of a class action. Zinser, 253 F.3d at 1190–

28

76
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1
2

92.
Plaintiffs argue that “[g]iven the abundance of common proof at issue, requiring class

3

members to proceed individually ‘would merely multiply the number of trials with the same issues

4

and evidence.’” Mot. at 25 (citing High-Tech, 985 F. Supp. 2d at 1228). Although Defendants

5

challenge superiority, see May 6 Trans. at 49:15–19, Defendants raise the same arguments against

6

superiority as they do against predominance, namely, that trial will be overwhelmed by individual

7

issues. See generally Opp.; May 6 Trans. at 49:15–19. The Court has already rejected each of

8

Defendants’ arguments. See supra Section IV.B.

9

The nature of Defendants’ alleged overarching conspiracy and the desirability of
concentrating the litigation in one proceeding weigh heavily in favor of finding that class

11
United States District Court
Northern District of California

10

treatment is superior to other methods of adjudication. Thus the Court finds that Plaintiffs have

12

satisfied the superiority requirement.

13
14

D. Relation Back of Claims for 2001–2003
Finally, Defendants challenge the class definition proposed in Plaintiff’s Motion as an

15

improper expansion of the class alleged in the SAC. Opp. at 11. Specifically, the SAC alleges a

16

class consisting of “persons who worked at any time from 2004 to the present” for the Defendants.

17

SAC ¶ 195. In the Motion, however, Plaintiffs seek to include Pixar and Lucasfilm class members

18

beginning in 2001—three years earlier than the claims in the SAC. Mot. at v. Similarly, Plaintiffs

19

seek to include DreamWorks class members beginning in 2003—one year earlier than the claims

20

in the SAC. Id. Defendants contend that the claims for 2001–2003 are time-barred and would not

21

relate back to the original complaint in the instant action if Plaintiffs’ Motion were viewed as a

22

request to amend the SAC under Federal Rule of Civil Procedure 15(c)(1)(C). Opp. at 12.

23

Plaintiffs respond that the class definition was updated due to discovery disclosing when

24

each Defendant joined the alleged conspiracy, and that Defendants suffered no prejudice thereby.

25

Reply at 7–8. Plaintiffs additionally note that while their revised class definition for claims against

26

Pixar, Lucasfilm, and DreamWorks begins earlier than the 2004 date originally alleged, the class

27

definition for claims against Blue Sky and Two Pic (ImageMovers Digital) now begins later, in

28

77
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

2005 and 2007, respectively, as a result of discovery. Although Plaintiffs maintain that an

2

amended complaint is unnecessary to expand a class definition, Plaintiffs’ Supp. Br. at 2, at the

3

hearing on Plaintiffs’ Motion for Class Certification, Plaintiffs indicated that they would seek

4

leave to amend if necessary. May 6 Trans. at 53:3–12.

5

As Plaintiffs note, courts have reached varying conclusions as to whether a substantive

6

modification to a class definition requires formal amendment of the complaint. See, e.g., In re

7

Cathode Ray Tube (CRT) Antitrust Litig., 308 F.R.D. 606, 620 (N.D. Cal. July 8, 2015) (collecting

8

cases). Regardless of whether formal amendment is required in every case, however, it is the

9

appropriate course of action with regard to the addition of claims for 2001–2003 in this case. The
addition of several new years of claims and class members through a revised class definition is

11
United States District Court
Northern District of California

10

particularly concerning here because any such claims from 2001–2003 are undisputedly time

12

barred absent a plausible allegation of fraudulent concealment. Plaintiffs’ Supp. Br. at 2

13

(“Plaintiffs agree that the claims of class members from 2001–2003 . . . are time-barred absent

14

fraudulent concealment.”). Defendants contend that the SAC contains no fraudulent concealment

15

allegations specific to the 2001–2003 time period capable of sustaining those new claims. May 6

16

Trans. at 42:22–43:6. The Court similarly has not located such specific allegations in the SAC,

17

and Plaintiffs did not identify any at the hearing. See generally SAC; May 6 Trans.

18

Because Plaintiffs have attempted to add claims from 2001–2003 without first amending

19

the complaint, Defendants have had no opportunity to test the sufficiency of Plaintiffs’ allegations

20

of fraudulent concealment as to 2001–2003. In these circumstances, fairness requires that

21

Plaintiffs first seek leave to amend their complaint before the Court will permit Plaintiffs’

22

proposed expansion of the claims. Moreover, in the instant case, fairness particularly requires

23

amendment of the complaint because Plaintiffs’ proposed modification results in a substantial

24

expansion of liability. Plaintiffs do not dispute that the additional three years account for roughly

25

one-third of Plaintiffs’ total damages. See May 6 Trans. at 43:13–14; 52:25–53:2.

26
27
28

For these reasons, the Court DENIES without prejudice Plaintiffs’ Motion to the extent it
seeks to certify a class including claims prior to 2004. Plaintiffs may seek leave to amend their
78
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

complaint to include claims for 2001–2003. The Court notes, however, that leave to amend need

2

not be granted where there is “undue delay, bad faith or dilatory motive on the part of the movant,

3

repeated failure to cure deficiencies by amendments previously allowed, undue prejudice to the

4

opposing party by virtue of allowance of the amendment, [or] futility of amendment.” Leadsinger,

5

Inc. v. BMG Music Publ’g, 512 F.3d 522, 532 (9th Cir. 2008) (quoting Foman v. Davis, 371 U.S.

6

178, 182 (1962)). In addition to addressing the statute of limitations and Foman factors, Plaintiffs’

7

motion must also set forth the basis for Plaintiffs’ claim that the proposed amendment relates

8

back. See In re Syntex Corp. Sec. Litig., 95 F.3d 922, 935 (9th Cir. 1996) (explaining factors for

9

relation back).

10

United States District Court
Northern District of California

11

V. CONCLUSION
For the reasons set forth above, the Court finds that Plaintiffs have satisfied all of the

12

requirements of Rule 23(a) of the Federal Rules of Civil Procedure, as well as the requirements of

13

Rule 23(b)(3). However, the Court finds that Plaintiffs have not shown that it would be

14

appropriate to amend the class definition to include claims for 2001–2003. Accordingly, the Court

15

GRANTS Plaintiffs’ Motion for Class Certification as to the following class:

16

21

All animation and visual effects employees employed by defendants
in the United States who held any of the jobs listed in Ashenfelter
Reply Report Amended Appendix C during the following time
periods: Pixar (2004–2010), Lucasfilm Ltd., LLC (2004–2010),
DreamWorks Animation SKG, Inc. (2004–2010), The Walt Disney
Company (2004–2010), Sony Pictures Animation, Inc. and Sony
Pictures Imageworks, Inc. (2004–2010), Blue Sky Studios, Inc.
(2005–2010) and Two Pic MC LLC f/k/a ImageMovers Digital LLC
(2007–2010). Excluded from the Class are senior executives,
members of the board of directors, and persons employed to perform
office operations or administrative tasks.

22

The Court DENIES WITHOUT PREJUDICE Plaintiffs’ Motion for Class Certification as to class

23

members who worked at Pixar and Lucasfilm in 2001–2003 and as to class members who worked

24

at DreamWorks in 2003. The Court appoints named Plaintiffs Robert A. Nitsch, Jr., Georgia Cano,

25

and David Wentworth as Class Representatives. The Court appoints as class counsel Daniel A.

26

Small of Cohen Milstein Sellers & Toll PLLC, Marc M. Seltzer of Susman Godfrey L.L.P., and

27

Steve W. Berman of Hagens Berman Sobol Shapiro LLP, along with their respective firms.

17
18
19
20

28

79
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION

1

IT IS SO ORDERED.

2
3
4
5

Dated: May 25, 2016
______________________________________
LUCY H. KOH
United States District Judge

6
7
8
9
10

United States District Court
Northern District of California

11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

80
Case No. 14-CV-04062-LHK
ORDER GRANTING-IN-PART AND DENYING-IN-PART PLAINTIFFS’ MOT. FOR CLASS CERTIFICATION