Siddharth Hariharan, et al v. Adobe Systems, Inc., et al

Filing 1

FILED ON 11/07/2013 PETITION FOR PERMISSION TO APPEAL PURSUANT TO RULE 23(f). SERVED ON 11/07/2013. [8856405] (HC)

Case No. __________
UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT

IN RE HIGH-TECH EMPLOYEE ANTITRUST LITIGATION

Petition for permission to appeal from the United States District Court
Northern District of California
The Honorable Lucy H. Koh, Presiding
Case No. 5:11-2509-LHK
DEFENDANT-PETITIONERS’ EXCERPTS OF RECORD
VOLUME II OF VIII
ROBERT A. VAN NEST, #84065
DANIEL PURCELL, #191424
EUGENE M. PAIGE, #202849
JUSTINA SESSIONS, #270914
KEKER & VAN NEST LLP
633 Battery Street
San Francisco, CA 94111-1809
Telephone: [phone]
Facsimile: 415 397 7188
Attorneys for Defendant and Petitioner
Google Inc.

789556

EXCERPTS OF RECORD
N.D.CAL.
DOCKET #

DOCUMENT

PAGE

Volume I of VIII
(District Court Orders—Public Versions)
1.

531

Oct. 24, 2013 Order Granting Plaintiffs’
Supplemental Motion for Class Certification
(public redacted version)

0001

2.

382

April 15, 2013 Order Granting in Part and Denying in
Part Plaintiffs’ Motion for Class Certification
(public redacted version)

0087

Volume II of VIII
(Expert Reports – Public Versions)
3.

518-2 518-4

Expert Report of Professor Kevin M. Murphy
(public redacted version)

0140

4.

424-2

Supplemental Expert Report of Edward E. Leamer,
Ph.D.
(public redacted version)

0340

Volume III of VIII
(Expert Reports – Public Versions)
5.

440

Supplemental Expert Report of Professor Kevin M.
Murphy
(public redacted version)

0402

6.

442

Expert Report of Kathryn M. Shaw, Ph.D.
(public redacted version)

0570

Volume IV of VIII
(Depositions and Declarations – Public Versions)
7.

308-1,
445-2

Deposition of Edward Leamer

0676

8.

538-8 538-11

Declaration of Danny McKell in Support of
Defendants’ Opposition to Plaintiffs’ Motion for
Class Certification
(public redacted version)

0691

1
789556

N.D.CAL.
DOCKET #
9.

516-6

10.

DOCUMENT

PAGE

Declaration of Frank Wagner in Support of
Defendants’ Opposition to Plaintiffs’ Motion for
Class Certification
(public redacted version)

0713

District Court Docket Report

0725

Volume V of VIII
(District Court Orders FILED UNDER SEAL)
11.

Oct. 24, 2013 Order Granting Plaintiffs’ Motion for
Class Certification
(under seal version)

0804

12. 383

April 15, 2013 Order Granting in Part and Denying in
Part Plaintiffs’ Motion for Class Certification
(under seal version)

0890

Volume VI of VIII
(Expert Reports FILED UNDER SEAL)
13.

Expert Report of Professor Kevin M. Murphy
(under seal version)

0944

14.

Supplemental Expert Report of Edward E. Leamer,
Ph.D.
(under seal version)

1144

Volume VII of VIII
(Expert Reports FILED UNDER SEAL)
15.

Supplemental Expert Report of Professor Kevin M.
Murphy
(under seal version)

1180

16.

Expert Report of Kathryn M. Shaw, Ph.D.
(under seal version)

1348

17.

Expert Witness Report of Kevin F. Hallock, Figure 7

1454

2

N.D.CAL.
DOCKET #

DOCUMENT

PAGE

Volume VIII of VIII
(Depositions and Declarations
FILED UNDER SEAL)
18.

Deposition of Michael Devine

1455

19.

Declaration of Danny McKell in Support of
Defendants’ Opposition to Plaintiffs’ Motion for
Class Certification
(under seal version)

1458

20.

Declaration of Frank Wagner in Support of
Defendants’ Opposition to Plaintiffs’ Motion for
Class Certification
(under seal version)

1480

21.

Exhibit 24 to Declaration of Lin W. Kahn in Support 1492
of Defendants’ Opposition to Plaintiffs’ Supplemental
Motion for Class Certification

3

3

Case5 11 cv 02509 LHK

Document518

2

Filed10 07 13

Page1

of

81

UNITED STATES DISTRICT COURT

NORTHERN DISTRICT OF CALIFORNIA
SAN JOSE DIVISION

RE HIGH TECH EMPLOYEE
ANTITRUST LITIGATION
IN

Master Docket

No 11 CV2509 LHK

EXPERT REPORT OF
PROFESSOR KEVIN
MURPHY
THIS

M

DOCUMENT RELATES TO
Date

ALL ACTIONS

Time

January

17

130 pm
8 4th Floor

2013

Courtroom
Judge
Honorable Lucy

H Koh

Case5 11 cv 02509 LHK

Document518

Filed10 07 13

2

Page2

of

81

Table of Contents
I

CREDENTIALS

1

II

ASSIGNMENT AND SUMMARY OF CONCLUSIONS

2

III

BACKGROUND ON THE DEFENDANTS

13

IV ECONOMIC THEORY AND EMPIRICAL EVIDENCESHOW THAT
INDIVIDUAL FACTORS PREDOMINATE OVER ANY COMMON FACTORS IN
DETERMINING WHETHER AND BY HOW MUCH ANY MEMBER OF THE
PROPOSED CLASS WAS INJURED BY THE CHALLENGED CONDUCT

A

The Challenged Agreements Would Not Meaningfully Reduce

the Supply of

17

Information

1

Shows

Evidence
Source

Employees

that

Restrictions

C

The Alleged Conspiracy Would

D

on

Harmed

if

Methods Would Not

Recruiting

Benefit

There

any

is

Affect

Market Compensation

Some Members

of

22

Employees

and Therefore Determining

Individualized

were Injured

and By

How Much

Would Require
25

Analysis

Tremendous

The Composition

Variation

in

Compensation

of Total Compensation

Paid to Individual

Differs Across Employers

Employees

26

V DR LEAMER PROVIDES

NO ECONOMIC SUPPORT FOR PLAINTIFFS

CLASS CERTIFICATION REQUEST
Summary

B

Economic

of

Dr Leamers

Analysis

28
28

Opinions

Does not Support

Dr Leamers

Claim that

Agreements would Reduce Information Flows Limit

Compensation

1

Evidence

Dr

Shows

that the

30
Flow

Leamer Wrongly Claims

Discovery

of Information and thus

the Challenged

Compensation Data Shows

3

the Challenged

Price Discovery or Affect

Generally

Not be Reduced by

2

25

and

Employees

A

20

Class Even

the Proposed

Others

Individualized

1
2

an Important
18

Employee Compensation is Highly
Which

of Other Defendants are not

and Hires

of Recruits

B

if it

15

Price

Discovery Would
30

Agreements

that His EmpiricalAnalysis of

that Restricting

Defendants

Cold Calling Impedes the

Price

38

Process

Data do not Support

Dr

Leamers

Agreements Prevented Increased
Class that Otherwise Would

Claim that

the

Timing

of the

Compensation to Members

Have Accompanied Economic

NonCompete

of the Proposed

Expansion

40

Case5 11 cv 02509 LHK

C

Googles

Big

Document518

Bang Does

Filed10 07 13

2

Not Support Plaintiffs and

Page3

Dr Leamers

81

of

Claim of

Class Wide Evidence

D

42

Economic Theory and Empirical Evidence

a Rigid Salary

Compensation

at

Adjustment

Dr

Dr
Dr

that in

Some Employees Results

of

in

47

of Others

Some Common Factors

then Only

Individuals Compensation

if

Common

is

Factors Potentially

Compensation

Leamers

Process

5

Compensation

44

Structure

Defendants Necessitate

Several

Analysis Wrongly Assumes that

Leamers

Affect

4

Practices

Circumstances Increased

Reduced Compensation

Affected by

Claim that
43

Economic Theory Does not Support

Some

3

Dr Leamers

Rigid Compensation Structures

Defendants have

1
2

Refute

49

Model Does not Demonstrate

Because

Leamers

it

his Hypothesized

Price Discovery

Cannot Explain Compensation Changes

Constant

Ranking

Compensation

Attribute

54
Analysis

is

55

Misleading

6

One

Cannot

Conclude

that Because

Some

Defendants had Policies and Even

Formulas for Annual Compensation

Adjustments

Move

E

Dr Leamers
Dr
Dr

Model

Econometric

Impact Because

1
2

Common

Exercise

Impact across Defendants

Framework Underlying

Statistical

5

Dr

Show Common
57
59

is

Impact

Assumed Not
61

his Regression

Leamer Does not Report any

Whether

Fails to

in its Implementation

No Common

Demonstrates

The

of

56

Undercompensation

of

in

Dr

that a Limited

Flawed Both Conceptually and

Warm up

Leamers
Leamers

is

it

Demonstrated

3
4

Number

the Structure

Cold Calls Would

Additional

his Results

are Robust

Leamers Regression

Dr Leamers

Analysis

Sensitivity Tests from which

is

Improper

70

or Fragile

Model Does Not Explain Changes

in

Compensation

Over Time

6

Dr

72

Leamers

Conduct

Variable

Cannot Capture

the Impact of the Challenged

74

Agreements

7

Estimated Persistence
Discovery

66

to Evaluate

Model and

Effects

are Inconsistent

with

his Claim that Defendants

Dr Leamers

Price

had Rigid Compensation

Structures

8

77

Summary

78

Case5 11 cv 02509 LHK

I

My

Filed10 07 13

2

Page4

81

of

CREDENTIALS

1

Document518

Professor

name

I

of

earned a doctorate

my bachelors

Angeles

in

At

I

I

degree

degree

the University

theory empirical

have taught

in

from

also in

of

labor

economics from

the University

of

Chicago

Economics

I

I

teach economics

in

My

Of

teach

economics and

graduate level courses

the economics
the incentives

of public

of

my teaching

focuses

focus

in

both research

particular relevance

across

wage

industries

to

structure

labor economics

by age

of

the

Political

education

wages and

age

have authored or

economics Those
including

price

issues In these courses

policy

to

how

and the

to apply economic

economic

I

the

use the tools of economics

has been on integrating

articles

by

skill

Several

of

and gender

My

to

analysis

principles

and

on

my papers
My

structure

work

in labor economics

labor

has

as well as the determination of relative

level

have focused on the determinants

work on wage determination

labor mobility I teach

with a focus on the wage

groups differentiated by

I

microeconomics

in

the issues in this matter I have published extensively

and occupations

addressed the links between

5

I

Los

of California

the impacts of regulation

on two things how

and teaching

addressed the market determinants of wage

of the

1986

that motivate firms and individuals

markets and the determinants of wages and compensation

wages

in

analysis

empirical

4

Economics at

of

both the Booth School of Business

markets the determinants of market prices and

of

system Most

data

Service

of Chicago

the University

understand the behavior of individuals firms and markets and
to

Distinguished

1983

since

economics

cover a wide range of topics including

legal

J Stigler

George

1981

and the Department

operation

am the

in the Booth School of Business and the Department

Chicago where

received

3

M Murphy

Kevin

Economics

of

the University

2

is

PhD level

courses

on

also has

empirical

and the determinants of relative wages across

education and skill

co authored

more

than sixtyfive articles

have been published

American Economic Review

in leading

the Journal of

Economy

1

scholarly

in

a

variety

of areas in

and professional

Law and Economics

journals

and the Journal

Case5 11 cv 02509 LHK

6

I

am a

Fellow of the Econometric

Arts and Sciences

Economic

In

1997

under the age of forty
five year fellowship

7

In addition

regulatory

matters

In

the John

was named

I

my position

to

matters

joint

2005

who show

to individuals

consulting

member of

Bates Clark

the

exceptional

Medal

81

of

which the American

American economist

outstanding

merit

of

American Academy

MacArthur Fellow an award

a

I

the University

at

firm that specializes

on a

have consulted

economic and

involving

practices

was awarded

a

Page5

and promise

that provides a

for continued

and

work

creative

Economics a

I

and

Society

Filed10 07 13

2

awarded once every two years to an

Association
1

enhanced

Document518

ventures

and

of

in the application

variety

I

am

of economics

mergers class

of anticompetitive

at Navigant

also a Principal

of antitrust intellectual

legal issues such as

allegations

Chicago

law and

to

and other

property

damages

certification

exclusionary access

labor

tying price

fixing and price discrimination

8

I

bodies

and

US

I

have submitted expert

of the reports I have

attached

per hour for

reports in

Trade Commission and

Federal

my CV

Court

have submitted testimony in Federal

as

Appendix
on

A

Navigant

I

US Senate

numerous cases

have consulted

and the testimony

filed

my work

I

the

for the

I

and to

have

state

testified

US Department

on

behalf

of the

of Justice

have given over the past four years

Economics is being compensated

regulatory

at

is

a

A list

provided

rate of

in

1,250

this matter

II

ASSIGNMENT AND SUMMARY OF CONCLUSIONS

9

I

Google

have been asked by Counsel for Adobe Systems

Google Intel

Inc

Lucasfilm and
by

1

individual

Pixar

Intel

Corporation

collectively

and representative

Defendants
2

Inc

to provide

Adobe

Apple Inc

biennially

until

2009

but

it

Apple

Intuit Lucasfilm Ltd
an economic

Plaintiffs that an alleged

plaintiffs

The John Bates Clark Medal was awarded

Intuit

Inc

now

is

analysis of claims

conspiracy

awarded

annually

among

See

http www aeaweborghonors awards clarkmedal php
2
In

Re High Tech

Certification

Employee

Antitrust

and Memorandum of Law

Litigation
in

Support

Plaintiffs

Notice of Motion and Motion for Class

Motion

2

October

1 2012 p 1

Case5 11 cv 02509 LHK

Defendants to

and

fix

the compensation

suppress

impact and would be susceptible

a

class of

ll

natural persons

Document518

to class

employees

of their

wide proof

employed on

Filed10 07 13

2

Class

AllSalaried Employee

or

Court to

certify

a

creative

and or

research

United

States

Class

Technical

by one

excluded categories

or

and development

potential

10

two

the

alternative

when

classes only

calling each others

that cold calling

impact

essential to their claim of antitrust

who

limited to the particular individuals

Consolidated

4

Amended

Complaint

define the All Salaried

Plaintiffs

in

Re

receive

Employee Class

from March 2005 through December

Lucasfilm

2009

and senior

6

7

8

is

cold

the Defendants

from the Class are

executives

50

between

the

two

2009

c Google
e Intuit

Ph

Motion

D

ways and

Plaintiffs

claim that due to the

Litigation

persons employed

Complaint
on a salaried

1

basis in the

2005 through December 2009

b

2009
December 2009 f

from March 2005 through December

g Pixar

corporate

d

from January 2005 through December

officers

p1

October

of

effects of cold calling are not

from June 2007 through

or

method 6
7

a variety

in

8

Antitrust

recruiting

cold

members of

the

boards of directors

1 2012

46

Complaint

distinguish

42

Complaint

E Leamer

calls

All natural

as

employees

of all Defendants

Expert Report of Edward
Complaint

retail

same

The

to refrain from

a particularly effective

a Apple from March

from January 2005 through December 2009

Excluded

5

Report
do not

the

of their claims

In support

also distinguish

the compensation

that

Adobe from May 2005 through December 2009

5

among

High Tech Employee

United States by one or more of the following

Intel

I

cold calling has a significant impact on employee compensation

that

3

my report

in the technical

and with

periods

typically

and in

more

evidence specific to one or the other

present

employees

same time

Dr Leamer

and

or

a

a salaried basis in the

Leamer

claim that there was a conspiracy

Plaintiffs

employed on

E Leamer PhD

class definitions

I

who work

AllSalaried Employee Class

Plaintiffs

certify

have asked the

Plaintiffs

natural persons

the

wide

2005 through December 2009

alternative

fields that are

Edward

by

arguments and evidence provided
between

ll

more of the Defendants during

offer the Expert Report of

Plaintiffs

a

as

a class

by one

States

4

defined

as identified for the

would have

a salaried basis in the United

Class As an

81

of

have asked the Court to

Plaintiffs

of the Defendants during part or all of the period from January

the

3

Page6

including

According

to

Dr Leamer

orally in writing telephonically

otherwise applied

for

a job opening

ColdCalling
or electronically

refers to

Leamer Report Footnote

3

communicating

with another

3 adopting

firms

directly

employee

essentially the

in

who

same

any manner
has not

definition

as in

Case5 11 cv 02509 LHK

conspiracy

alleged

the resulting

and mobility had a cumulative

Document518

on

Filed10 07 13

9

Class members resulting in

all

from Defendants than they otherwise would have received

11

Plaintiffs

acknowledge

of a small number

consisted

challenged agreements or
Defendants to not cold
conspiracy

do

impact on a class that includes

DNCC

employees

identified

virtually all

among the

which

11

claim that the small number of

Plaintiffs

during the periods

agreements

not cold call

each others

call

US

I

by employees who

did not receive

reduced the compensation of
engineers

to

agreements
job

workers

cafeteria

41

Complaint

Plaintiffs

bilateral

in

general

My

application

if

if

those

potential

in cold calls

a

Defendant

those employees

potential

all

do not

cold call restrictions

to

received

but also

from

firms as well

by

not prohibited

the

challenged

at

issue in this
at

case

is

words

that

they generally

a firmwith which

employment with

in other

if

or potential

before filing an

information

emailing employees

in general

have received

about a particular job opening

had expressed no interest in exploring

new employment opportunities

was

that

simply were gathering

or

According

or almost all employees in their proposed

conduct

include

12

pairs of

the compensation

of the Defendant

by potential applicants

from calling

of the alleged

between

value

their

that they might otherwise

applicants

certain pairs of

agreements had a class wide

not only affected

claim that

would

This definition

understanding of the

were intended to prevent

exploring

Plaintiffs

as the

salaried employees at all seven Defendants

allegedly

calls

my report

Despite the limited nature

other salaried employees

such as responding to inquiries

openings

agreement

all

cold

lower compensation

agreements between

Defendants reduced the information available to employees about
Plaintiffs the reduced flow of information

81

seven Defendants

refer to in

claim that the reduction

Plaintiffs

of

10

conspiracy

that the alleged

of bilateral

Page7

of competition and suppression of compensation

elimination

effect

2

the

it

had an

Defendant

or in

they were a totally passive

candidate
9

10

11

Complaint

123

Dr Leamer

says that he

Report
12

110 emphasis added

Complaint

According

who

are

s that

Defendants

entered

into several

Plaintiffs

leverage

claim that
that

the

e outreach

cold calling

i

seeking employment and other active

compensation

negotiating

by restricting

to Plaintiffs

not actively

depressed

showing

understand

additional

agreements Leamer

22

by impairing

information

to solicit applications

competition

for

flow about compensation

and job offers reducing

of employees and minimizing movement of employees

Dr Leamer

Defendants

describes

agreements

preventing class members from discovering

abundant

evidence

would tend
the true

to

of

their

between

to all Class

suppress employee

value

4

common

work

from candidates

employees the agreements

firms Motion

members capable

compensation

Motion

p 16

p3
of

generally by

Case5 11 cv 02509 LHK

classes

were

by an amount

affected

conventional methods and

12

In his report

Dr

common

Leamer

Document518

that can

Filed10 07 13

2

be measured on a class wide

Page8

81

of

basis using

evidence

1 whether

two questions

addresses

there is

common

proof

to

Classwide
Non Compete

each proposed class capable of showing that the
the competition

sic compensation

13

of its

members

or formulaic method capable of quantifying

suffered

by each class

that

or nearly

all

14

Dr Leamer

all members

Dr Leamers

2 whether

and

amount

of both classes

had

their

reliably using

materially reduce the information available

to

there is

reliable

yes and

is

suppressed

economic methods

First the challenged

to Defendants

a

both questions

compensation

standard

analysis has three essential steps

reduced

artificially

of suppressed compensation

concludes that the answer

aggregate amount that can be quantified

13

the

Agreements

by an
15

agreements must

employees Second

that reduction

informationmust cause the salaries of individual

employees to be reduced Third the

rigid compensation

must cause the reductions

some employees

of

to

reduce compensation

First the labor markets from

recruiting practices of

number

would

of employees

Defendant

the information

15

wide basis Economic theory and

provided

common

to others

compensation

of the Defendants

Leamer Report

at

10 see

would

not

at

be

to one

11

5

effects

would be

common

at

the second

Finally

if

highly

In particular

employee likely would

an impact on one

21 57

the

directly affect only a small

of the proposed class

This would stop the chain

would

enormous and diverse and

Second any

members

across

are not rigid then

Leamer Depo

hire are

is critically

the information levels of employees at any

the first step

Ibid

Leamer Report

that

that reduced the informationprovided

impact on individual

14

not meaningfully affect

and would not be

same conduct

structures

of employers

This would break the chain

individualized

13

a class

which Defendants

a small number

the

on

somewhat

in the compensation

evidence demonstrate that his analysis of each of these essential steps

empirical

flawed

structures of the Defendants

in

increase

step since the

the compensation

individuals

compensation

Case5 11 cv 02509 LHK

would

14

Thus

None

based on

my analysis

I

with economic

Opinion

1 The

implausibility

suppressed

as

logic

by

which

I

explain

was dwarfed by

to

the

11

percent

number

Opinion

by

estimated

2 Empirical

challenged

during the

Dr Leamer

recruiting

employee turnover
movements between
turnover

agreements

Given

to recruit

that

from

method

their

16

DNCC

recruiting activity

See Table 1 below

to

of Defendants

subject

and from other Defendants

to
will

were

tend to overstate

6

new

actual

their

that

implausible

employees by the large

that

the

employees

whether resulting from cold calling

one

was

this

percent

2001

to

of Defendants

2011 Employee

means that during
by

unaffected

1A

still

and

1B

available

DNCC agreement the

agreements

was

agreements was even lower

and see Exhibits

a

It is

Defendants

to

hires and separations over the period

of potential

And

the

class wide impact on the information

for only about

Defendants that had

opinions

VB1 below

other Defendants

accounted

that

of over 8,000

hiring activities demonstrates

forms of recruiting of other than cold calling

employees

movements

or

empirical

at these firms

they received

as a proxy for underlying recruiting activity

about 99 percent

16

to

opinions

my core

are

combined workforces

economically significant

Employee movements

Dr Leamers

an average

hired

undercompensated

see Part

at

class period demonstrates

compensation

of their

evidence of Defendants

had no

conduct

or another

81

three

not surprising

is

it

about labor market opportunities and compensation available

a

of

detail in this report

of applications

for five years these firms consistently

amount

in

2005 and 2009 Defendants

equal

all

The following

evidence

empirical

of Dr Leamers claim that average
Dr Leamer and Plaintiffs claim

workers per year

alone

to their allegations

high level of hiring by Defendants

Collectively between

hiring

Page9

by documentary and

theory and contradicted

Plaintiffs claims are not supported

Plaintiffs allegations

let

conclude that Plaintiffs claims and

Given the lack of economic

regarding

hold

links in the chain

of the required

are both inconsistent

evidence

Filed10 07 13

2

compensation for other members of the class This would stop the chain

not increase

the third step

Document518

fraction

the actual

Using
this

period

the challenged

for details

17,18

Given the

and used during the class period
of

turnover

accounted

for

importance of the challenged

by the

Case5 11cv 02509LHK

relative unimportance
outside

effect

of

the class period

cross

hires

separations
before and

and

any

in that

restriction

hires

cross

separations

movement would not have a

represented

after

The

the class period

Defendants

1.1 percent

just

1.2 percent

the Defendants

of

on

counterfactually

1

As such

the information available

material

collective

hires

total

hires and
shares from

clearly demonstrate that

employee

of the labor market

changes in those flows would have no

to Defendants

those flows and the associated

81

Defendants

of their total

account for a minute fraction

activity for employees of these firms

effect

data in Table

of

Defendants both within and

the class period

represented

Page10

a share that is not materially different than the corresponding

movements between

substantial

Filed1007 13

2

employee movement between

on compensation Moreover during

from other Defendants
hires

Document518

employees even

recruiting activity represented

if

the only

source of information available to employees

17

For purpose

According

of this discussion

to Plaintiffs

Google Intel began

in

I

use the period from 2005 to 2009 to approximate the

agreements

2005

Together

annual employment at the Defendants
18

This analysis

the

uses

involving
these

hires

and separations

slightly

one year windows
between

out

four

companies

from 2001 to
to identify

previous employer and the hire date

at

Defendants

due to the two different windows

the

of the seven Defendant

four

accounted

class

period

Adobe Apple

about 92 percent

of total average

2011

hires

looking back

new employer and

in a

for

companies

given year the numbers

used

7

one year between

separations
of

hires

the separation

and separations

date at

year For
may differ

looking ahead one

Case5 11cv 02509LHK

Document518

Filed1007 13

2

Page11

81

of

Table 1

Summary of

Hires

and Separations

Defendant Companies

at

Annual Average
2001

From Other Defendant Companies

2010

2001

2009

2011

2011

5,795

Overall Hires

2005

2004

8,814

11,435

8,193

35

From Other DNCC Defendant Companies

95

159

85

28

69

123

64

From Other Defendant Companies

0.6

1.1

1.4

1.0

From Other DNCC Defendant Companies

0.5

0.8

1.1

0.8

12,182

15,985

16,525

14,700

Overall Hires and Separations

FromTo

Other Defendant Companies

71

FromTo

From Other DNCC Defendant Companies

Source

b

Based on

analysis

There were

many

in Exhibit

sources

Defendants averaged
hires

total

1 averaged
suggests

20

of

percent

market and the associated

During

looking

for

Defendants

labor market

to

work

of

and employees
total

agreement was in place

1A

leaving
at

Therefore

in a typical year

co workers

some

This

with the labor

Other sources
of

at other firms dedicated

whom may
internet

of

have

sites

based advertising as well as cold calling from

employers and from Defendants where
Plaintiffs theory

8

new

of Defendants

during the year

employees had direct contact

friends working

Under

workforce

about 11.3 percent

workforce

total

informationfrom

non Defendant

Defendants

roughly 7,200 a year based on Table

information

such as job boards and media and internet

number

hires

the class period

1 averaged

of Defendants

labor market information include

DNCC

1.1
0.9

during the year while separations

that about

the very large

1.8
1.4

about 78,000 employees a year see Exhibit

about 9.2 percent

been actively

1.2
0.9

Defendants

including

roughly 8,800 a year based on Table
workforce

127

of labor market information available

non Defendants

to

168

239

1A and 1B

employees other than cold calling
Defendants to go

305

139

0.6
0.5

Other Defendant Companies

191

57

From Other DNCC Defendant Companies

no

of the spread of information

Case5 11cv 02509LHK

Document518

informationfrom these other sources

Filed1007 13

2

which

any

vastly exceeds

Page12

reduction

of

81

in information

resulting from the challenged

Defendants employees even

c

agreements would have been widely disseminated among
if

Yearto year

fluctuations

that might have

Defendants

widely from

varied

roughly one percent
the hypothesized

fluctuations

in

overall

in cold

There was no reduction

DNCC

2005

for

by Defendants

as a

whole

Thus

of 4,100 in

Opinion

3 A reduction

because

there are

a

Market

agreements

alleged

a all other

in inter Defendant cold calling

including

by which Defendants

the price employers

agreements affected

words they

affected

available

fill

to

compensation

b

As a

matter

of

neither the

those jobs

VB1 below

see Part

Defendants or

2005 2009

with

b Defendants

as during the

Dr Leamers

economic theory

would

would not
recruit

2001

central

employee

number

there is

wide harm

for labor

number

no reason why they would
see Part

conspiracy

when

not lower compensation

demand

jobs nor the

of the class generally

in

other

of employees
affect market

IVB below

to restrict

a small number of

approaching employees at one or a

on a class wide basis

agreements were not commitments to reduce salaries or

9

in class

determined by supply and demand for labor

is

of available

the alleged

result

employees

neither the supply of nor the

Therefore

or compensation

other firms

large

pay for labor and thus the compensation

employers from using a single recruiting tool

few

Such

Defendants during the class period The

the data are inconsistent

earned by members of the proposed class

The

a

Defendants

many channels

price

2009

would be caused by

premise that the agreements reduced informationflows and consequently

movements between

hiring

with economically significant

agreements was essentially the same during

2011 period

low

that

calling due to the challenged

hires between

cross

the class period

to a

conduct on class wide compensation

in

any hiring changes

aggregate changes dwarf any changes in the

percentage of Defendants hiring from either
with

Over

agreements

hiring activity are inconsistent

effects of the challenged

pairs of Defendants

hiring activity vastly exceed

of total hires accounted

reduction

calling between

a high of 12,700 in

of over 8,500 hires These

difference

no cold

resulted from the challenged

by Defendants

d

in

there were

restrict

The challenged

employment and would

Case5 11cv 02509LHK

not have

changed

Document518

Defendants had to

positions

demand

the supply of or

a

certain employees through

The

fill

for labor overall

or the

Page13

number

agreements only affected

alleged

method

particular

Filed1007 13

2

cold

Even

calling

if

of

81

of job

recruiting of
the agreements

reduced recruiting of certain employees from particular employers and potentially
affected

certain individuals

a

new employee for
Adobe

rather than from

class benefitted

c

As a

matter

rather than
potential

of

the person hired

see Part

Defendants had

This reduction

Plaintiffs theory of information

employees as well which
fact that a reduction

could be positive

would

Opinion

a

is

If

such as Microsoft

member of

to the extent

less cold

the proposed

has an effect could raise

it

number

calling reduced the

would reduce

it

could increase

the

or

flow

this

the opposite

would

the pool of potential

amount

increase

of the effect

negative Moreover
for

of

of compensation

hires

that the

employees from the smaller resulting labor pool

some

compensation

by

hypothesized
available

to

members

The

both sides of the

in

is

ambiguous

cold calling

and reduce demand for others
across

Under

of other

Plaintiffs

the fact that the reduction

individuals

on wages would not be common

implies

of the proposed class

that
see

IVB below

4 Defendants

Defendants had

highly

between

below

compensation structures are not rigid

and exercised substantial

employees

individual

was

calling

in cold calling affects the options

demand

increase

the impact

Part

a

Apple

firms and workers means that any overall impact on compensation

market
it

cold

by Defendants

contacted

offer to attract

to

non Defendant

from Microsoft

reduce average compensation

for those Defendants

Adobe

those

IVC below

economics reduced

candidates

by Defendants through

agreement with

from a

an open position

recruiting through

to increase

benefit those hired

For example if as a result of an alleged

channels
recruited

which would

channels

other unrestricted

would be

the impact

as a result

Dr Leamers own

similar individuals

in every year and for each

compensation

model

with large variations

individualized

observationally

flexibility

unexplained

by

Defendant

in setting compensation

implies

that

employee compensation

even within particular job categories

see Part

IVD below

there is substantial

Dr Leamers model Dr Leamer
10

of

As

I

and

demonstrate

dispersion

in

employee

has shown that different

Case5 11cv 02509LHK

jobs have

Document518

different average compensation

compensation

Filed1007 13

2

but not that increases

Page14

of

81

in an individuals

resulting from a cold call results in higher compensation

for other

employees

b Dr Leamers

premise

a change

not imply that

also

in

A rigid

flawed

compensation

wage

increase

for all

structure

even

if

one existed would

more employees would

for one or

because the cost of increasing

entire structure

enormous an

is

compensation for one employee

employees and would be

Thus Dr

resisted

the

shift

would be

Leamers

theory makes no economic sense

c

Finally

Dr Leamers

alternative

analysis cannot distinguish
that compensation

hypothesis

the impact

of Defendants

he hypothesizes from an

employees

is

broadly

determined by competition in a vast labor market and that adjustments for individual
employees unique circumstances
individualized

Opinion

methodological

outside

offer are highly

VD3 below

see Part

5 Dr Leamers

such as an attractive

conduct

and

regressions suffer from severe conceptual

flaws and are completely unreliable and thus uninformative His regression

methodology provides evidence that

is

with his conclusion

inconsistent

of class

wide impact

and damages

a

Given the nature

of Plaintiffs allegations

common

challenged conduct was
there is class

Dr Leamer

during the conduct period were
methodology
values of his

some

suggests

so called

Dr Leamers own

common

whether the impact of the

is critical

effect

is

not

actually

and

substantial

variation

overcompensated

across

Indeed

common

Specifically

positive

statistical

across

a class

see Part

whether

wide

changes

application

of his

the estimated

Defendants

and for

VE 2 below

methods which

I

Thus

critique

Defendants in the estimated

as the result of the challenged

11

on

that compensation

conduct effects vary substantially
the

understanding

Defendants

across

were

to

can be measured

to demonstrate

regression specification

below show

some employees

the impact

fail

that the changes

of the Defendants

further

Defendants

across

wide impact and whether

Data analyzed by

basis

the question

conduct

impact with

Case5 11cv 02509LHK

b Dr Leamers

data

theory

how

members

When

firm
level compensation The
shows

the conduct and

theory

effect

estimated

Defendants

economic

SP

500

existence

My

economic

their

analysis

evidence to support

on members

fails

of these factors invalidates

should not change

periods

see Part

as follows

recruiting

IV I

hiring

show

Moreover

benefitted
there is

these

same

facts

that he fails to support

his findings

of class

if

Plaintiffs

conduct

effect the

members for

all

as measured

by changes

in

the

or even

In Part III I provide background

and compensation

practices that is relevant

that there is neither

is

information on the

economic

logic

many

my

nor empirical

conduct would have a

only one of

to

common

impact

recruiting tools and other

These facts together

refute

imply that some members of the proposed class will have

basis to certify

the

VE 5

agreements would reduce compensation on a class wide

from the same conduct that

no economic

see Part

VE 5 below

Cold calling

Plaintiffs claim that the challenged

his statistical

undercompensation

smaller

Plaintiffs claims that the challenged

of the class overall

determinants

Similarly simply controlling for changes in overall

yields substantially

estimates

In Part

regression

For example limiting his regression analysis

Defendants are not an important source of hires for any Defendant

basis

conduct

to account for important

and financial market performance

report is organized

Defendants and

Leamers

to

but also a major error

conduct effects are unreliable

VE 4 below

index

highly

not only contrary

is

determined

implying overcompensation

positive

conditions

overcompensation

15

is

is

effects also are highly unstable reflecting the

estimated

post conduct

see Part

stock

This

Yet doing so completely changes his estimated

correct

is

81

VE3 below

Dr Leamer

imprecision with which they are estimated
to

is

Dr

estimated

see Part

that his claimed

below Dr Leamers

of

evidence that the challenged agreements reduced compensation

of the proposed class

analysis and

Page15

and obvious feature of his

a critical

compensation

properly

In his conduct regression analysis
of

Filed1007 13

not independent

are correlated

an individuals

no meaningful

provides

c

of

inference

in statistical

of

only because he ignores

significant

that his observations

own

his

2

impact of the challenged agreements on compensation

estimated

statistically

Document518

Plaintiffs

a class

allege

Part

harmed

other

employees which means

V critiques Dr Leamers

economic requirements for

12

class certification

analysis and explains

Case5 11cv 02509LHK

Document518

2

Filed1007 13

Page16

of

81

Case5 11cv 02509LHK

Document518

2

Filed1007 13

Page17

of

81

Case5 11cv 02509LHK

21

Third

bonus

to

on employee referrals

current employees

important channels are

new

who

website

LinkedIn

and dice

com

changed

over time

throughout

22

graduates

is

LinkedIn

for

28

example

for

by

implications First

It

implies

that

the use

a

utilize

81

use of

in

companys

to the

networking

firm Other

importance

may have

mid2000s

since the

job

such as

sites

of these different channels

by Defendants

of

many

but the

different recruiting

and

in the use of one channel can

or at least reliance

that both employers

implies

applications

the

the recruiting practices of these firms

reduction

information on hiring and compensation

members that

of

more

and

increased

it

unsolicited

increased

by

ultimately are hired

The importance

different channels has characterized

important

compensated

who
27

com professional

and job fairs

the past decade

Page18

have formal referral programs that provide a

refer individuals

From an economic standpoint

channels

Many

such as monster

job boards

many

26

university

applicant

use of

Filed1007 13

2

use a large variety of channels for recruiting employees They tend

the companies

rely heavily

to

Document518

on

other channels

it

these other channels will have

implies

that individuals

expanded

opportunities

be

This has two

and employees have alternative

Second

will

sources

including

critical

of

class

as a result of the

reduced cold calling

IV

ECONOMIC THEORY AND EMPIRICAL EVIDENCE SHOW THAT
INDIVIDUAL FACTORS PREDOMINATE OVER ANY COMMON FACTORS
IN DETERMINING WHETHER AND BY HOW MUCH ANY MEMBER OF THE
PROPOSED CLASS WAS INJURED BY THE CHALLENGED CONDUCT

23

The

allegations

pairs of Defendants

Declaration

of

Steven

to

in this matter concern the impact of the challenged

cold calling on compensation

eliminate

Burmeister in Support of Defendants

received

Opposition

For example employee
of

Fichtner
28

referral

new hires See

Opposition to Plaintiffs
27

the

Defendants

Motion

to Plaintiffs

26

percent

by

for Class

p3

Certification

40

agreements between

Motion

for

Depo at 177 210 See

See Declaration

and Declaration

of Jeff

is

the

most important recruiting

Declaration

Vijungco

of Tina Evangelista

of

Jeff

Vijungco

Class Certification

also

pp

Declaration

of

of

method

Declaration

of Jeff

Tina Evangelista

23 Declaration

of

p3

15

for

Adobe

Adobe Systems Inc

Frank

Wagner

in

accounting for about

35

Support of Defendants

Vijungco

p2

p 12
p 10 Declaration

of Chris Galy

p2

Case5 11cv 02509LHK

Document518

employees The challenged agreements

salaried

24

The

who

five individual

certified

at

issues

in

and that there

each class

member without

Plaintiffsclaims

compensation

many

members

other recruiting channels

claim to represent

salaried

in

things both that

the

of quantifying

common

how

that analysis explains

if

2 do not

2 that

had a

issues

predominate

in

to

analysis can

1 do not

agreements that

another

3 do not

compensation Thus

labor markets operate

at

damages owed

of

An economic

evaluations and

reference

or

at

restrict

the relevant

Defendants

economic

an agreement that potentially limited one of

one Defendant might have been made aware

Defendant would reduce compensation

theory has three essential elements

reduction

relative to what

in

gentlemens

a Pixar employee

in

of

by

received

under his theory

In particular

by members

informationmust lead to a reduction

they would have received

of the compensation

class wide reduction

Pixar

of the seven

affect direct determinants of an employees

or performance

opportunities

amount

the

analyses

agreements to affect compensation received

rigid nature

29

persons

all

order to have such a class

compensation

August

would

2 2012

notify
at

at the defendant

that

Lucasfilm

it

if it

in compensation

for internal equity

would not

counter

123 16 126 15 134 23 135 6 148 3

16

if

and

5

3

a

This sequence

Lucasfilm

offer to

1

for those

firms must then generate

Pixarmade an

order

by Defendants

absent the challenged agreements

through the pressure

agreement with Lucasfilm

and that Pixar

see McAdams Depo

structures

in

of the proposed class

those agreements must materially reduce the level of informationpossessed

individuals

or fix

of the proposed class

for the alleged

employees

virtually

employees

technical

understand that

I

individualized

methods by which employees

Dr Leamers

25

on

way

cause class wide changes

employment

81

many employees could be hired

in this lawsuit

among other

reasonable

relying

only

whether given how

recruiting

specific

a

such as promotions

hiring nevertheless
issue is

is

compensation

relate directly to

of

determining whether class members have been injured by the alleged

conspiracy

support

Page19

29

2005 and 2009

must demonstrate

Plaintiffs

did not restrict

or in the alternative

any time between

over individual

named

plaintiffs

were salaried employees

Defendants

all

compensation

or any other element of

Filed1007 13

limit how

prohibit hiring employees of other Defendants

wages

2

made an

a Lucasfilm

offer to

employee

Case5 11cv 02509LHK

which underlies

Dr Leamers

and inconsistent

with economics

A

The

Document518

price discovery

and

Filed1007 13

2

and

internal

evidence

empirical

equity frameworks

of

81

speculative

is

show below

as I

Agreements Would Not Meaningfully

Challenged

Page20

Reduce the Supply of

Information

26

As a

matter

of

economic theory

the impact

source will depend on the size of the supply
the extent

source

is

which supply

to

reduced Here

challenged agreements
provides

leads

to

is

the supply of

Dr

In

Plaintiffs

Dr

and

at

Classwide impact

A would

Defendant

period

A has

a

depend on

DNCC

25

Dr Leamer

31

Cold calling

available

is

any specific

fraction

recruited
32

at

my discussions

A has DNCC

is possible

or almost all class

to

employees

importance

of cold calling

32

recruiting

If

outside

As hires while employees

agreements account for one percent

in the

with

either

Defendants

them

that

than reduces

compensation

above

or

and the evidence

does not disagree
that there

that

is

not

tracked

data

with this but he simply claims that his conduct

However

as

by

directly

the

to

managers

explain
I

do

not

in their

have

channels

does not account

that

other
for

many employees

a very

were

referrals

was an impact which means that

40 425 183 223 413 21 414 7

of

an assumption he

Therefore

cold calling generally

from the Defendants recruiting

such as employee

of other

below market

data and their recruiting

this generally

and interviews

the class

gained by cold calling could reveal

information

is

with which

of other Defendants

measures of the importance of cold calling relative to other recruiting

of recruiting

pudding

that

compensation

provided in declarations

through other channels

Dr Leamer

in the

It

not clearly identified

and in

than information
large

nor explains

or employees that current

declarations

1 the

importance

assumes that all price discovery raises rather

acknowledges

recruiters

of

of Defendant

percent

for all

the

that cold calling

the information

hires then the share of Defendant As hiring potentially affected

Defendant As

neither

was reduced by

agreements on informationpossessed by employees

2 the

and

when supply from one

less information available

combined impact

the

Defendants with which Defendant

30

in

or

recruiting activity with information flow

agreement as a source of potential

calling accounts for

cold

in

of the challenged

relative to other recruiting channels31

Defendant

the reduction

Leamer equate

and claim that reduced cold calling results

27

increases

less price discovery and lower compensation

In effect

of market supply

the elasticity

information that allegedly

model

Leamers

30

members

and

restriction

to the market from other sources

it

supply to the market from one

of eliminating

I

explain

does not demonstrate impact

17

this

combined

below

his

effect

conduct

regression

provides the

must be large
regression

is

Leamer

proof

Depo

so flawed that

it

Case5 11cv 02509LHK

agreements during the class period
important

from firms in general

A did

As

heavily

Even

expand

not

companys employee
a

no cold

A had

Defendant

1

DNCC

has

was

the impact

hired

of Recruits

Defendants
Defendants

new

33

hires

no

was

of

agreement s

because

it

as

assumes that

more

likely that another

A did

with which Defendant

of the Defendants

recruited

Other Defendants

a

without

with which

call

cold

a

B would

cold

loss of

the top

not

single firm

just

for

of other

I

if

20 previous employers
34

by Defendant

of

in a

from an agreement between

impact on the information

available

cross hires between

a very small fraction

the Defendant

Defendants employees

summarized the former employer of

opportunities

call

not have a meaningful

have accounted

3 shows

are not an Important

employee compensation by

agreements affected

the relative importance

based on recruiting data provided
that

more

one

or even

earned by employees of either company

Exhibit

as

cold calling employees of

a Defendant

Using Defendants data

The

A and B could

compensation

it

81

and Hires

on

recruiting efforts

Defendant

make

is to

non Defendant

that the challenged

likelihood

reducing informationdepends

Defendants

A avoids

someone from

Evidence Shows that Employees

The

29

either

agreement a

call

agreements

is

other recruiting channels

utilizing

Defendant

if

the effect

of

lost information is replaced

Source

28

it

by

A has DNCC

Defendant

overestimate

recruiting efforts

Page21

assuming that cold calling

a cold call agreement if that employee

by which

process

its

would

Filed1007 13

2

only 0.25 percent

likely

this

below

explained

Defendants with which

not have

is

hiring from the Defendants with which

in

Defendant

Document518

of their total

new

A striking

and

to

A and

Defendant

hiring

anyway

hires at each of the Defendants
observation

from this exhibit

firms accounts for more than six percent

is

of hires at

any Defendant and that the top 20 firms combined typically account for less than 20 percent

of

33

Despite the availability of this information
to Plaintiffs
deposition
hires
34

First Set of

that

he had no

by a Defendant

My

Defendants

way to

for

which

understand was provided by the Defendants

I

Document Production

quantify

See Leamer

staff standardized

field in the

Requests

Depo

the
at

October

importance of other

Defendants

in

response

claimed during his

as the former employer of new

45 18 48 25

employer names in the recruiting databases
data typically

3 2011 Dr Leamer

was self reported

to the extent

by the applicant

18

possible

and was entered

the prior

employer

as free form text

Case5 11cv 02509LHK

Document518

Filed1007 13

2

Defendant and
hires

total

collectively other Defendants

Thus

even a policy

stronger than the limitation on

periods
class

that Plaintiffs

members

30

typically accounted

that eliminated

a

would

In fact measured

over

all

for less than three percent

hiring from other Defendants

all

single recruiting channel

here

challenge

only about one percent

from

certain Defendants

not meaningfully affect

Defendants

numbers could mask

a narrower

time period

by

low

of employees

of other Defendants

by one Defendant

that any reduction

were an effect for individual
of the importance

to

31

The number

initiated

to

identify

by cold

number

as a fraction

be lost by
of hires

4B shows

from these exhibits

these

separations

same

the total

by one Defendant

even outside

the class

agreements would

Thus

even

if

calling

among Defendants

Defendants versus between

year At

the

from another

and

Exhibit

at

4A

shows

the

the seven Defendants

RD

class

out of the Defendant

same time movements between

As

can be seen

firms is large

the defendants are

the class period

If

hiring

Defendant were economically important in the

price discovery process then employee movement between

19

a

Defendants and other firms

Creative and

of employees in

provides

price discovery that could

regardless if one looks before during or after

of employees

analysis

transitions that might have been

transitions to and from Defendants

figures for the Technical

to

there

any damages they suffered

as a percentage of total employees

movement

and highly variable from year
miniscule by comparison

and thus

But the extremely

the proposed class

labor market

of all employee

on cold

restrictions

and

on

conduct

of the challenged

information and potential

of

broken out by movements between
Exhibit

quantify

Defendant to Defendant

of

summarize the amount

conceivably

effect

or narrower

and of cold calling as a recruiting channel would be

and

those individuals

calls

accounts for

employees or small groups of employees an individualized

of other Defendants

needed

the alleged

in cold calling because

any significant class wide economic

not have

during certain

share of hires

for a substantial

been affected

implies

much

the flow of information to

particular employees that might have

period

is

of

of total hires

Theoretically these aggregate

level of hiring

which

hiring from other Defendants

group of employees where other Defendants accounted

total

81

of

hires Other Defendants account for at most three of the top 20 former employers at any

total

way

Page22

Defendants should account for a

Case5 11cv 02509LHK

substantial

part of the overall

opposite

true Even

calls

is

amount

the

members

if

all

of the proposed class

B
32

Market

price

separations

including

agreements affected

affected

jobs

neither the

Therefore

compensation

33

Even

there is

no reason why they would

if contrary

a meaningful

class wide basis

While

some employees

that

that

upward

Plaintiffs

Affect

Market Compensation

demand

number

for labor

affect market

compensation

pressure

The

decline

in

overall

a reduction

same

in

reduction

above

the decline

recruiting efforts

The

reduction

on compensation

fact that the reduction

in this

that effect

has been

reduced

in potential

at

hires

required

Defendants

to

would
fill

raise

Vijungco

would

not

on

a

hires for those

the level of recruiting of

the open positions which

would

of the effect hypothesized

the opposite

in cold calling affects the options

available

to

by

both

ambiguous even

if it

were

discovery process given that information on compensation

to

a small

only at the later stages
group

that

then

is

pp

56 and

Declaration

of the recruiting

interviewed

and Intuit clearly state that they do not discuss compensation
of Jeff

was

scenario the fact that there would be more demand for some

most commonly provided to candidates

Declaration

those

fill

or

in cold calling

the pool of potential

in recruiting reduces

Hiring should be a reasonable proxy for the price

candidates

to

they

cold calling would reduce the number of firms contacting

and the level of compensation

material Moreover

is

The

words

in other

of employees available

employers and employees makes the overall impact on compensation

35

for

more intensively

compensation and certainly would not reduce compensation

same amount

other individuals

put

the evidence presented

to

necessarily reduce overall

by

into account the incentive

and

of the class generally

sufficient to cause

firms

by

received

determined by supply and demand for labor

jobs nor the

of available

by cold

initiated

pay for labor and thus the compensation

neither the supply of nor the

number

the

limited both in terms of its magnitude

the price employers
is

81

of

that exactly

impact on compensation

Would Not

Methods

Page23

shows

exhibit

other recruiting channels

earned by members of the proposed class
alleged

The

even before taking

fluctuations

by using

Filed1007 13

and from Defendants were

to

would be extremely

on Recruiting

Restrictions

workers

of

lost and the potential

of information

recruiters to compensate

and

2

35

movement

hires

relative to other market level

Document518

pp

process

once the number of

a job or job opening

until the later stages

of Chris Galy

20

for

34

Both Adobe

of the recruiting

process See

Case5 11cv 02509LHK

and

individuals

common
hires

less

members

across

would stand

34

One

demand

types

employees

of

employment

are widespread

in

of software

engineers

in the United

employment

of software

engineers

in the industries

35

The economics

Employees

when they

are

Exhibits

leaving

1A

and

1B show

about 11.3 percent

of the

performed this comparison

conservative

37

Robert

Details

the Defendants

reason

why

of several

in

jobs

of

or less of
of

operate

36

compensation

of

Defendants
and

to

relocate

Defendants are very

of employees hired at

for

of employees at Defendants

engage demonstrates
and Michael

by Defendant

2

1992
are

a

Google

Intel

in

hires at Defendants

during a year

of workers

Defendants averaged

participants

of the average

9.2 percent

in the job market

software engineers and limited the analysis

that

e employees

i

year This means that on average about 20

were active

Defendants

P Ward
p 440

shown

new

the class period annual

focusing on software engineers

calculation

Journal of Economics
38

percent

willing to change

based on CapIQ information for the Defendants

compete

H Topel

which

percent

leaving

and restricting total employment to firms engaged
Defendants

two

Defendants average number

e workers

Defendants combined workforces

I

in

more

other

with any single

States and only about 10 percent

6

During

employed during

of workers

Defendants

industries

for

and the labor forces

that turnover
38

i

while annual separations

number

and

and

or joining is substantial

represented

36

37

For example as shown in Exhibit

and Pixar were

36

young

are

of both proposed

engineers

meaningfully by changes

will not be influenced

recruiting and hiring practices

young

across

of labor mobility provides an additional

Defendants employees

geographically

for software

accounted

employment

would be harmed

accounting for only a small fraction

collectively

5 Defendants

Exhibit

81

of

in the pool of potential

and a large portion

and

geographically

the seven Defendants

As shown

are left out potentially

opportunities

Page24

on compensation would not be

Workers who remain

to all Defendants

Employment

software engineers

employer or even

who

Filed1007 13

2

that the impact

implies

while those

common

job category

is

for others

of the proposed class

to benefit

classes

Document518

in

account

Job Mobility

Appendices

because that

1A

is

the profession

businesses of the type
for

21

that
of

in a general

in

2D

Men

in

which

even this

the

named

sense

in

Plaintiffs

which

only a small share of job opportunities

and the Careers of Young

through

of

in a typical year

to industries

simply to show

percent

107 The Quarterly

Case5 11cv 02509LHK

the sense that they

had sought

out a

Document518

new

This extensive

source of information on market compensation
also

shows

that

employee movements

extremely small fraction

37

Thus

data

of total

show

Filed1007 13

by

job or been recruited

informationabout compensation

obtained

2

to

or cross

across

is

for compensation

employers because

of employees

means

that Defendants

The
if

38

it

same

the labor markets in which

skills

Defendants

a large

a

to

number

of other

firms than others but the

and experience

these firms recruit

and

to

equalize

hire is broad

of employees into and out of Defendants

employees have access

for an

and

and other firms

vast flow of informationabout market

and compensation

opportunities

C

at

for employees against

of employees with the

employees are mobile The movement

thereby

provides a natural

Defendants accounted

companies The competition may be more immediate with some
tendency

81

employer and

another

employee turnover

hires between

compete

of

both employees and employers The exhibit

hires and separations

that Defendants

Page25

Alleged Conspiracy

Would

Benefit

Some Members

of the

Proposed Class Even

Harmed Others

Even

if

the alleged

the proposed class because

cold call the necessary

conspiracy

reduced the compensation

they did not receive

corollary is that

it

information or a job opportunity

of

because of a

lost

compensation of other members of the

increased

proposed class by opening up opportunities

by some members

received

that they otherwise

would

not have

received

or

under Plaintiffs theory providing them with information that they otherwise would not have
obtained

Thus

Plaintiffs

uniform across class

and

Dr

membersnor

at least

in principle

Defendants generally follow the same process

open position

39

arguments imply that the impact

Adobe

Intuit

reaches

out to the

and Intel

all

processes See Declaration

of

describe
Jeff

companys

the roles

Vijungco

p 2 Nearly
defendants use
Jeff Vijungco pp 23 Declaration of Frank
Tina Evangelista
p3
Evangelista

all

is

neither

even harmful to all but rather a mix of benefits to some

caused by the same conduct that could

39

own

Leamers

for filling

open jobs

recruiting department

of managers sourcers

p3

have injured others

Declaration

multiple

Wagner

to a

39

talent

22

Declaration

with an

manager or

and recruiters as key to their recruiting

of Chris Galy

p 2 and

Declaration

sources to find potential candidates

p 10

A manager

of Chris Galy

of Tina

See Declaration

p 2 and

Declaration

of
of

Case5 11cv 02509LHK

Document518

Filed1007 13

2

Page26

81

of

thirdparty
who

recruiter

either before or at the

of

pursue a variety

who

is

of avenues

responsible

candidates

inappropriate

interviews

methods

The manager

Plaintiffs claims

and

the recruiters role

the decision

is

how

affected

Plaintiffs

to

fill

fill

that job is led

only the methods

alleged

if

agreements

including

DNCC

would be
job

by

DNCC

or hired through

See Bates Document

initiate

is

institutional

cold

open

positions

from

fill

fill

some open
a

also use

contact

on

networking

sourcer

with potential
and

or
for

important in evaluating

Dr Leamer

the impact that

the recruiter

cold

framework

The

in person

recruiters

would

23

the

decision

manager and

Since the role of a
find candidates

calling employees at certain companies

companies such

The

at

would be

recruited

through

by

the

as Defendants

firms with which they

net effect of the challenged

channels other than cold calling employees of a

76550DOC000014

claims to

agreements challenged by

or recruiting employees

that candidates

whether

The

made by

is

alleged

calling employees at other

agreement

the likelihood

and

to hire generally

agreements through other channels

to increase

In order to

used to find qualified job candidates

they were constrained

with which there was no
have

to

by

fairs

may

then generally use phone interviews

incentives

who

and selection of

To add

to hire

given the relevant

recruiter is to identify candidates
other channels

identify

to

make economic sense and whether

a job to

marketing

including

that are unqualified uninterested

who

and

candidates

the recruiter

a small group of promising candidates

ultimately decides

has a logical magnitude

whether there

to obtain

of finding

employees

current

candidates

Recruiters

applicants

and on

an employee makes a

efforts

recruiter

by

referred

efforts
40

way

to fill some Defendants

difficult

active

to eliminate

for the position

Understanding

estimate

making more

any

posted internally

jobs website

important

and sponsoring job

through cold calling

including

other screening

40

for

more

An

is

the friend then submits an application

to identify additional

typically those that are

job opening

companys

the

not require

and candidates

such as LinkedIn and attending

positions

40

may

an employment opportunity and

the pool of internal applicants

websites

on

that it is posted

through employee referrals which

friend aware
to

same time

such as monstercom and LinkedIn com

websites

is

The

will assist with the search for candidates

agreements

interviewed

DNCC

offered

Defendant

a

Case5 11cv 02509LHK

cold

including

would

who

denied to the employees

100,000 inquiries a month on

receives

begin

applications

41

while

The consequence
would

the class

common

a

receives

that

already

salary

working

by doing

hardware engineers

so

for the

reduction

probability that

in potential

person hired
off

was

If

able to obtain

Declaration

See

harm

The

class

one

company who

of

Jeff

the

class

call

but where

cold

worked

Vijungco

to

a

accepted
a

to Plaintiffs

only because

of

is

Fichtner

not be

In Plaintiffs

compensation

but for world
for the position

hired might be

an

different external candidate

employs

etc

fairs

etc

employees The

non Defendant

all

that recruiters

called or hired increases

means

software

including

that he

the

use

43

with any

fact that the

was made

that person

better

becomes a

claims has been injured even though he

the challenged

conduct

44

Thus

under the

p3

him it

this

person would not have been hired for

that

he would not have received would have been higher

24

done

claim harmed the

html

Depo at 144 521 147 12 19

44

of

gets hired

employees have access to

the job

all

some members

member who

http online wsj com article SB10001424052970203750404577173031991814896

43

which

two could

professionals

a Defendants

employed by

thus according

a better position

calls

is

example

jobs and potentially

switches

hire might be

of those other candidates

hires through

and

Instead

member who

human resource

accountants

he was previously

of the class

41

42

The

that

million resumes in 2011.42

of the hundreds or thousands of firms that

wherever he previously

by doing so

member

two

received

for

the portal through

site

employee referrals LinkedIn monstercom job

other methods

Adobe

and distinguishing

Or the new

Defendant firms where recruiters do not cold

The

career

Dr Leamer

the cold call

81

value allegedly

the Plaintiffs logic from the conduct that Plaintiffs

to

did not receive

someone

Through

their

hires is vast

not be class wide

some were harmed

if

someone who was employed at one

42

Google

evidence such as that offered by

higher

engineers

adobe com

that there could

benefit even

member who

internal hire

its

of

and possibly the benefit of a better job

the cold call

of such potential

reported

is

it

is

has benefited according
class

The pool

Page27

Defendants

information about

the

did not receive

with one of the Defendants

with

obtain

Filed1007 13

2

non DNCC

or other channels directed at

calls

process these other individuals

41

Document518

is

irrelevant

whether

Case5 11cv 02509LHK

Document518

Filed1007 13

2

Page28

very theory put forward by Plaintiffs the challenged conduct would benefit
even

harmed others There

if it

D

is

Employee Compensation

Which

if

any

no class wide harm even

if

Highly Individualized

is

Employees were Injured and By

some

and

some class members
are injured

individuals

Therefore

81

of

Determining

How Much Would

Require

Individualized Analysis

1
43

There

is

Tremendous

The tremendous

Variation in Compensation Paid to Individual Employees

variation

in annual compensation

7A

each Defendant shown in Exhibits

at

theory

that a rigid compensation

and

structure

employees resulting from cold

individual

each year the range of

at odds

is

would be

changes

of the proposed classes

with a central tenet of Plaintiffs

that changes

necessitates

calls

compensation

total

7B

members

for

transmitted

in compensation

substantially

differs

the class

across

for

45,46

In

Defendants

across

47

44

Thus

compensation

during the alleged
percent

45

The

the
is

conspiracy

Dr Leamer

by

presented

analysis

below

his data

no

promoted

cannot

I

he analyzes

be moved

discuss

in
48

Changes

in

Appendix

example

tenure

sex

base salaries also show

4A

through

Company specific

for

job

Pixars

4D

14A

Dr

very large

the distribution

performance
bonuses

that

also

of

affects

are tied to the

48

In

a particular Defendant

same conclusion
differences

Even

and

this understates

3

14B

Leamer

takes

variations

in

point

out in

any

year

a 10

received

in

response

is

employee

of

not

Dr
in

pay variation

to external

since

e

pressure

one of
there

g

explained by individual

and Appendix

also

compensation

films

critique

in his regression

3A

Appendix

levels which

my

even for individuals

the level of

account

into

success of individual

25

I

this variation

see

compensation
changes

As

compensation

in

across job classifications

below and show in Exhibits
age

at

the Defendants

This implies that under Plaintiffs theory the

from Software Engineer 2 to Software Engineer

characteristics
47

this

substantial

46

As

raise

supports

show

over 4,000 narrow job categories

no reason individuals

in lock step across

period some employees

raise while others received

data

Leamers

move

does not

show
at

analyses

3B

substantial

I

also

show

variations

a particular Defendant

see McAdams Depo

at

42 243

3

Case5 11cv 02509LHK

propensity

for salary changes

coworkers

would vary

terms

employees received
across

for an individual

substantially

show

these data

Document518

that the requirement of

internal

would have
understand

2

any employees compensation

affected

first

equity and

increases

why

one employee received

The Composition of

a much

In

of

his or her

Dr Leamers

annually

differ substantially

for each Defendant

In order to understand whether

and

if

81

the degree to which

His theory would have to be tested and evaluated

separately to understand the source of the variation

across

of the proposed class

similar percentage compensation

Defendants

Page29

employee to be propagated

members

across

Filed1007 13

2

so by how much it

is

a cold call

necessary to

larger raise than the other

Total Compensation Differs Across Employers and

Employees

45
at

8A

Exhibits

the Defendants

components

of

and

8B summarize the

These exhibits

show

that Defendants

differ

in their relative reliance

employee compensation base salary bonus and

example during

conspiracy

the alleged

equity

likely varies across

internal equity

Plaintiffs

Defendants as well
evaluated

46

49

For

compensation

structure

which compensation
the degree to

theories apply

is

which

would vary

across

This implies that the validity of the Plaintiffs theory would have to be

The composition
Exhibits

9A

Apple and Google that

50

to

the three types of compensation

rigid

options

three

separately for each of the Defendants

Defendant

49

and

or

on

period a

Since the degree
individualized

by employees

composition of compensation received

Leamer Depo

at

of compensation

and

9B show

Dr Leamer

also varies substantially

the composition of total

analyzes in his Figures

across

are in

Appendices

15 through

26

5A

titles

within each

compensation for the jobs at
17.50

278 25 282 22

Corresponding exhibits for other Defendants

job

through

5E

Case5 11cv 02509LHK

Document518

2

Filed1007 13

Page30

of

This variation
the validity of Plaintiffs

employees

47

individual

level as

Dr

and

Plaintiffs

The impact

theories

would need

Leamer have

of differences

well The

in

be

in

job offer will depend

on

his preference

riskaverse employee

or

one who

the share of compensation

on

may

options and

stock

Thus

salary

an offer of substantially

company

another

will affect

how

The

and

all its

stock

match

if

the outside

on

stock

will vest

an employee

offer is heavily

compensation
vest

offered to an

when

the

that are unvested

employee without

51

same compensation
Consequently

employer wants to match

of

may place

may

receives

is

in

another

how

All else equal

company where

273 1

4

the equity

because

I

value

of base
to

from

offered

to

by

offer that he

at

same

outside

his current employer

an employee

that

would

a potential

interest

that holds substantial

the employees

inquiry

individualized

soon an employees

the

current employer

options

if

the

offer also will likely differ

For example Plaintiff Siddharth Hariharan testified that while employed by

employment elsewhere

little

base compensation

creates

stock options

the response

the outside

A highly

be worth less

an outside

allowing them to be exercised in the near future will be more likely to

hire than

51

he holds

options

forms

weighted toward stock options then

by some Defendants

options

and how many

different

in

resulting

a corresponding amount

because the impact of the challenged agreements will depend on
options

and any potential

than lower compensation

options

that offer might require only a small increase

reliance

the

compensation in base salary These same factors

an employer might respond when

asks his employer to

matching

almost

that provides

less than

to

provided in base salary

employers frequently

much

extends

greater expected total compensation

consists of a large expected bonus

if it

48

to change

value expected bonus

call

compensation

for receiving

expects

that

do so

failed to

the composition of total compensation

difference

implies

separately for each group of

established

and equity means that the value to an employee of a cold

bonus

him

to

81

was

still

vesting

was growing again and

I

at

Zynga

I

would have to

27

Zynga

was not looking
start all over

to

again

he turned

work

for

down

an offer

another

Hariharan

Depo

at

Case5 11cv 02509LHK

V

Document518

Filed1007 13

2

Page31

81

of

DR LEAMER PROVIDES NO ECONOMIC SUPPORT FOR PLAINTIFFS
CLASS CERTIFICATION REQUEST

49

support

Plaintiffs

class certification motion

their

with the Leamer Report which they

claim demonstrates that the challenged agreements suppressed the compensation of
all

Class members52

Leamers

and so provides

reduction

in compensation

of the proposed class were

A

for proving impact

Summaryof Dr

Dr Leamer

50

evidence

types

divides

capable

is

generally

54

of

or the

link between

showing

amount

2 Defendants

of

that the

to

of cold

52

price discovery

to

Motion

53

In his

asked

at

or nearly all

has not provided

or

members

a class

wide

First he argues that

agreements suppressed compensation

mean that on average members

which he says

negotiating

of the proposed class

In support

leverage and movement

3 empirical

that raises compensation

of employees

their

evidence that job

supports

He

offers three

about compensation

documents which he claims demonstrate

stayers which he claims

he

supported by economic literature of a

is

b informationflows

calling and

class wide

his conclusions

and

between

concern about the

movers

receive

that cold calling

explained at his deposition

that he also

p 23

report

Dr

Leamer does not provide a

in

his deposition

and when

Greater than 50 percent

Leamer Report

clear

definition

of

what

he means by

all

or nearly all

When

Dr Leamer what percentage are you confident class
that
my opinion is that most members of each class were

your expert opinion

members were undercompensated
undercompensated

54

all

was an average

damages

non compete

impact of cold calling on compensation and

leads

Dr Leamer

that

because of the challenged conduct

internal

higher compensation than

members nor

Dr

Opinions

1 economic theory

a the

53

neither that there

or nearly

However

for class certification

his analysis into three parts

job opportunities employees

firms

amount

Leamers

by which he appears

of evidence

of class

undercompensated

lower compensation

received

support

analysis and the evidence he offers demonstrate

generalized

method

the required

all

asked

he replied
what

Leamer Depo

65 heading IV A

percentage

is

322033 10

emphasis

added

28

most

and asked to provide a

range

he responded

Case5 11cv 02509LHK

regards

his conduct regression

agreements was to suppress

Document518

Filed1007 13

2

his claim that the impact

as supporting

Page32

of

81

of the challenged

55

compensation

generally

noncompete
51

Dr Leamer

Second

claims that

class wide evidence

agreements suppressed the compensation of

employee class and

class

technical

that he claims to establish through

members

nearly all individual

support

part of his

this

56

which

I interpret

argument he again

more

that Defendants

Agreements

52

55

don’t

was

would

Dr Leamer

the aggregate

not

at

require

have the

made

information

I

80725
a

that tells you

Leamer Depo

56

57

58

opines that

amount

Q How
I

members

consequence

of the agreement

done

to
it

answer

that

It’s

determined

is

through that damage

the

impact of these

at

agreements

100heading IV B

And

And

Leamer Report

or

some members To

1

results in

economic theory
somewhat

rigid

I

by common

that

factors and

NonCompete

analysis are capable of

I

don’t

Apple and Adobe

between

A Well

have the cold calling agreements

secondly

I

to translate that into some measure of
you can

haven’t

Q Your

largely

claims shows

thing

that

101

added

29

do So

had a database
I

tell

regression

101

Leamer Report

to

all

suppression to members of the proposed

model because

emphasis

harm

of

cold calling that was made and all the cold calling that

the

all

97 19 100 12

Leamer Report

average impact

58

was suppressed

no simple

question

of the allsalaried

broad effects on compensation of the

indicated

I already

have the information on

that the

undercompensation

such that one would expect

information

have

that the

forms of econometric

standard

much

don’t

members

regression analysis that he

of compensation

going to be very difficult

econometric exercise
have indirectly

e the

rigid salary structures

data set that

don’t

as a

is

3 multiple
class

showing

documents which he claims confirm concern with

have widespread effects on compensation

to

Leamer Depo

that

maintained

Finally

computing

and

earned by individual

compensation

internal

of

of evidence

equity

internal

all

opinion

reflects

offers three types

specifically demonstrat

Non Compete Agreements 57

an

as

of the proposed class and not just

2 Defendants

and

internal equity

capable

or nearly

his first set of analyses

that he claims demonstrates that concerns with
salary structures

all

is

that

you need to carry

would allow me

you that before or after
analysis

A The

to

out an
do it But

I

during comparisons

regression

yeah

See also

Case5 11cv 02509LHK

class

caused by the

model

statistical

53

As

I

Non Compete Agreements

2

Filed1007 13

59

words he claims

In other

explain none of

Dr

Leamers

opinions

are supported

analysis Empiricalevidence

including

evidence he ignores

interpretation of evidence that

he offers

contradicts

members have
members

B

Analysis Does not Support

Compensation

Dr Leamers
is

available

Dr Leamers

all

or nearly

all

economic theory does
by many ways

characterized

to employees

Price Discovery or Affect

Generally

on

available

of recruiting

jobs

of the labor market at issue here

not fit the facts

employees

a vast

amount

useful an economic

fit

the key characteristics

Dr

modeled and

model must

Leamer has

not attempted

theory of compensation impact
the competitive

nature

to

of the environment

to match

in

economic frameworks

and that these frameworks
can suppress

worker

Leamer Report
Leamer Depo

explain

compensation

135 emphasis
at

price discovery

various

In order to be

framework

and

their

and thus

employees operate

Price Discovery

to

link the challenged

useful

He

in evaluating

agreements

claims that

to

there

the

are three

the impact of the agreements

mechanisms by which anti Cold Calling

generally

Would

Agreements

claimed by Plaintiffs

that are particularly

Defendants

or market that is being

of the industry

of Information

on economic theory

widespread effect on compensation

movement

which Defendants and

Not be Reduced by the Challenged
relies

his

where

evidence on the amount of available informationand

available

Evidence Shows that the Flow

Dr Leamer

of information

and market compensation mobile employees and

employment and employee

60

to

Claim that the Challenged

Flows Limit

of

59

60

his opinions and demonstrates that class

account for only a small fraction

55

to provide a

as well as proper analysis and

tremendous density of employers and employees in small geographic areas

1

81

of the proposed class

Economic

one that

of

by proper economic

generally and that there has been no harm

not been injured

Agreements would Reduce Information

54

Page33

aggregate damages and demonstrating causality

for calculating

now

Document518

He

added

24825 16

30

focuses

primarily

on

the

agreements

market price

Case5 11cv 02509LHK

framework

discovery

workers

down

slows

occur

transactions

Calling

an

is

many

Calling

uninformed

logic

channel of information

labor contracts

employees

price discovery

63

are negotiated

by members

is

inconsistent

a

and valuing

62

According

he concludes

claims

is

many

Dr Leamer Cold

to

opportunities and

unequal bargains

and

the unique features of

Dr Leamer

Consequently

levels

81

of

absent

Cold

informed and

between

that the challenged

agreements

of the proposed class and cause employees to be

Dr Leamers

that he claims supports

compensation

61

Page34

very sluggish price discovery

about outside

in

The consequence

However

undercompensated

process

prices far from equilibrium

important

Filed1007 13

of uncovering

task

the price discovery
at

2

that labor markets can have

arguing

expensive and time consuming

that the

restrict

Document518

argument about

price discovery

is

invalid

and

the

reduced cold calling and class wide reduced

a link between

with assertions that he makes to support

Dr Leamer Exaggerates

that link

Information from the Challenged

the Loss of

Conduct

56

Dr Leamer

how much

provides no evidence of and concedes

informationmight have been lost

from the challenged agreements
Defendants plays

in

available

57
between

and

62

63

64

65

calling

his logic

and

4B

at

among

Defendants

for

I

discussed

above

Ibid

75

Leamer Report
Leamer Depo

at

80423

Leamer Depo

at

members

in Part

47248 14

31

the

employees

calling

by

other

amount

A1 showed

IV

after

the impact

evaluate

to

of information

of the proposed Class

before and

know

he has done no

that

roughly one percent

during the periods

73

Leamer Report

of

be used

would have on

a very small fraction

including

he claims cold

data that can

the Defendants

and does not
potential

he provides and concedes

on compensation

which

Defendants accounted

employee flows

61

by

4A

Exhibits

the central role that

recruiting and hiring flows

on cold
thus

Despite

Defendants employees and

the price discovery process

analysis of Defendants
that restrictions

64

to

he has not studied

65

movements

of the overall

the challenged

Case5 11cv 02509LHK

Document518

Filed1007 13

2

Page35

81

of

toyear
Thus

conduct

any reduction

informationflow from the challenged agreements would be

in

extremely small relative to the level of the overall flow of information or to the natural year
fluctuation

58

and

Plaintiffs

employees

the flow of

in

Dr Leamer

allege that the informationflow that

data

show

that these were

through

become aware
obtain

indicator

of the importance

extremely small

mean that

This does not

a cold call

Some

of this only because

employees

compensation However even

greater

compensation

for

Plaintiffs theory

some employees
that

would

it

same conduct would

if

may

not create

and

information

opportunities

may

compensation

then use the informationgained to

the loss of cold calls

benefit

flows and

under certain

may have below market

of a cold call and

of the

of cross

of such reduced information

an employee could not benefit from additional

circumstances
obtained

a good

is

members

lost to

The number

proposed class occurs as a part of the hiring and recruiting process
Defendant hires

was

resulted in lower

harm Indeed

class wide

some members

under

of the proposed class as I

explained above

Dr Leamer

59

also ignores

employees to compensate

the informationflow through

restriction

on

calling

among the

number

the vast

would

Google

at

including

60
software

engineers
candidates

engineers

in

cold

5 shows

Exhibit

these exhibits

calls

to

the

ability

including

also

to

accounted

targeted

for

two

Defendants

operate

Consistent

demonstrate that the reduced information

32

so

total

the

use of cold

did not cold

call

Defendants
on

the restriction

good candidates

would be expected

would have been

the Defendants

for the

economics

other

if

hire

and

cold calling to employees

to

at

They
cold

calling

Other firms

change

their

behavior

employees

percent

States and only about 10 percent

in which

suitable

recruits

their

non Defendants

in the United

skills

recruiters at Intel

job boards more intensively

that the Defendants

the industries

with

eg

Intel

meaningfully restricted

both Defendants and

and make additional

if

recruiters

one channel by

Defendants need not even reduce the

of other firms from which

Google

at

their

a matter of

during certain periods they likely increased

also use other channels

employees

As

other channels

For example

calling in their recruiting processes

employees

and

for restrictions in informationflowing through

increasing

cold

of both employers

the incentive

of

or less of

employment

many other
with

employment

results

of

of software

employers had
established

above

flow through a limited channel would

Case5 11cv 02509LHK

not have

impact on the

a meaningful

through

efforts

by

provided

1A

Exhibit

Page36

still

would have access

firms other than the Defendants

as well as to

of

available

to

81

to

the recruiting

Defendants

recruiting

other channels

A simple exercise

61

Filed1007 13

2

flow of labor market information

total

Indeed employees

employees of the Defendants
opportunities

Document518

the realities

illustrates

hiring from and

movements

of total hires over the class period

As shown

of this marketplace

to other Defendants accounted

A conservative

one

for roughly

to help understand

calculation

Table 1 and

in

percent

how much

information potentially could have been reduced by the challenged agreements could use the

2010 2011

higher post period

lost hires as the difference

rate of 1.4 percent

between

as a base of

hire level in the post and class periods

the cross

Defendant cross hires were lower by only about 0.3 percent
an annual difference
with

Defendants

each of the 30 additional

if

provided

information

would be 60
provided

of cross hires of roughly

of one percent

employees

movements

seeking work elsewhere

material

economic

all

sources

effect

on

left

available

to

total

labor turnover

and the firm to which

In other

it

moved

there

to add to the total bits of information

8,800 7,200

Defendants employees

sources friends

etc

would be even smaller Such

overall

compared

year

from one Defendant to another

through other channels

internet

Inter

of about 7,200 per year at

of 16,000 bits of information

in the bits of information

informationfrom

who moved

bits of information annually

by employee

30 employees per

of Defendants

both the firm that employee

information on market conditions

actively

in

to

additional

of 0.38 percent
obtain

twotenths

or less than

66

the

of total hires during the class period

hires of about 8,800 per year and total departures

total

words

number

in the

comparison and measure

such as

new

the actual

67

or an increase

Since employees

hires

co workers

percentage reduction

a small difference

would have no

compensation

66

This calculation
as

a benchmark

slightly

Even

conservative
is

because

as

I

discuss

share

this

employee

of

above using both

no change in aggregate crosshiring among

higher cross hiring in the post class period

increased
67

is

there

may reflect

the

the

pre class and post class

Defendants

periods

during the class period

growth of Defendants

and thus their

employment overall

percentage

is

movement and

much too large because
ignores

information

it

assumes

all

information

obtained by employees

33

in other

acquisition

ways

comes through

The

Case5 11cv 02509LHK

62

This change

also is de minimis

Over

recruiting activity

12,700

in

2005

That range

is

Document518

low

to a

compared with yearto year
by Defendants

the class period hiring

2009

of roughly 4,100 in

more than 200 times

Filed1007 13

2

the variation

the challenged

Leamers

agreements

marketplace

implied by the calculations

the evidence presented

small

of the relevant

Even

Plaintiffs

if

there were

employee compensation
Technical

are poster

children

by

characterized

facilitate

opportunities

Dr

to other forces

simply untenable

24

22 and

given the

do not
and at

wide impact

is

discovery that apply

in

Amounts of

to

lost

with

not grounded

in

the proposed class

Available

some markets with

such a group

a large portion

The

labor force

Information

limited access

to

of the

In particular

members

of the

AllSalaried Employee

labor market in which

they participate

class

is

use of internet and other channels by both employees and employers

and information flows

contacts

allegedly

so that incremental cold calling might affect

represent

least

such as monstercom dice

and

of the information

analysis ignore and are inconsistent

Vast

of employees

compensation

an informed

and

Have

and Defendants

Plaintiffs

extensive

current employees

friendships

is

his claim of class

dynamics of information

Employee

for

mobility

market and

some groups

informationabout appropriate

Fichtner

due

above demonstrates that any such effect

His theory and empirical

labor

of the specific

b

proposed

flows that

provides no evidence of the importance

vanishingly

consideration

68

in cold calling due

reduction

in hiring

roughly 30

realities

Dr Leamer

the nature

to

above

based on his Figures

of total compensation

incremental information

to

because of the agreements

64

from a high of almost

claims that the impact of the challenged conduct was economically significant

imply a sensitivity

would be

81

a range of roughly 8,500 employees per year

Given the degree of fluctuation

implying effects as large as 20 percent

63

of

in hiring and

fluctuations

varied

employees per year of what would be caused by the hypothesized
to

Page37

university

many publicly

com

graduates

with fellow students

salarycom

who

available

etc

are recruited

and colleagues

Depo at 45 946 4

34

data sources

and networking

by technology
68

on

salaries and

among both

firms

and maintain

Case5 11cv 02509LHK

Dr Leamer

65

implies

Defendants employees

Document518

that the labor market

2

Filed1007 13

from which Defendants

when

Page38

81

of

and where

hire

watercooler
compensation

obtain

information that they can use

consists of hapless

by

weakly with employers who
market

of technical

centers

contradicts

Dr Leamers

hiring

and separations

annual hires and separations
percent
that

70

compensation

at

of average annual

lacks

changing number of

Valley and other geographic

implication

71

firms to

But this characterization

in Silicon

The high

Defendants shown in Exhibits

as a fraction

consulting

members

that

high costs for transactions

involve

on

of information which

scraps

large and constantly

unsupported

their

rely mostly

often hire private

time money and personal dislocation

services including

obtain

to

and other employees located

proposed class are employed in jobs that

turnover

sources

The agglomeration of Defendants and a

other employers

technology

69

Internet

provide aggregated informationabout
credibility

who

and poorly informed employees

talk perhaps supplemented

they then use to bargaining

they negotiate

of the

involving

labor

rates of employee

4A and

4B

with the

employment between

sum of

10 and 25

during the conduct period demonstrates the substantial flow of information of the type

Dr Leamer

Dr Leamers

claims was restricted into and out of these firms and contradicts

claims that these employees were immobile

c

Lost Information

will not

have Class wide Impact

if it

is

Unique to

Individual Employees

66
valuing

According

to

Dr Leamer

expensive and timeconsuming task of uncovering

the

the unique features of workers

that cold calling helps uncover

slows

unique

down

the price discovery

features of potential

process

employees

is

72

and

But his claim

inconsistent

with his

claim that there would be class wide impact from reduced cold calling through the price
discovery process Indeed he stated

69

70

71

72

Leamer Report

75

Leamer Report

74

Leamer Report

73 emphasis

his deposition

75

Leamer Report

at

added

35

that

employers including

the Defendants

Case5 11cv 02509LHK

often differentiate compensation

information would

lost

would

Filed1007 13

2

the price discovery

process

employee specific

d

is

employee specific

unique

well accepted
economic

Dr Leamer

Claims about

Lost Information

provides

economic models

the labor

which

Leamer
to

lower

market but

cites

all

prices might rise faster

cites

but the

rockets

for understanding

limitations

that paper does not

why

prices rise quickly

in

by

itself

information

by Joseph

show

socalled

than they

fall in

Stiglitz

many

demonstrate that

imperfections

The

limitations

other three

and feathers phenomenon

rockets

is

or claim that

75

as a result of information

the

framework

cites

markets with imperfect

information

according
76

Dr

in the labor market leads

of information

and feathers model does not imply that a restriction of

information in the labor market would cause a reduction
only

Price Discovery

nor the broader

the cited literature

of his claim that restriction

these in support

wages

Neither

and incorporate

involve

and

claims One paper he

neoclassical

that acknowledge

economic papers he

that

Literature

model has

for his

undercompensated

employees are

to

74

literature

support

argues that the fullinformation

markets including

The

cold called and

the extent

flow and price discovery

claims that his information

in the economics

literature

73

There would be no class wide impact

Dr Leamers

Finally

81

features

not

To

features

of

then the effect of reducing cold calling also

are not Supported by the Economic

67

who was

to the unique features of the worker

relate

Page39

unique

employees to account for such

across

any impact on employees without those

not have

will be

Document518

in

wages

Rather

response to positive information but

fall

these models explain

slowly in response to

unfavorable information

73

I

Leamer Depo
know

dont

occur

are

are

Leamer Report
Joseph Stiglitz

Review
76

The

460 June
cited

people

for

who

promotions

thinking

result in the price

these

internal

turned out to be extraordinarily

A

discovery process

promotions

the big

bumps up

that

good

Information and the Change in the Paradigm

in

Economics 92

American

Economic

2002

paper by Green et al

Ye 2006

internal

66

major over the counter
and

Q Dont

would be relevant Because Im

it

compensation

74

75

316 16 317 17

at

that

develop

financial
theoretical

2010

is an empirical

market not
models

paper documenting

asymmetric

price

a labor

market The

cited

papers by Tappata

explaining

asymmetric

price

responses

36

adjustment in a

2006

and Yang

Case5 11cv 02509LHK

68

The economic

closely

to

on

and in turn

Samuelson made an

results

Dr Leamer

to

parties have

would

knows

prevail

with

more

is

77

information

78

some mutually

Similarly more recent economic

Dr

beneficial

Leamer

Thus

are forgone

asserts in his report that lack

employers and employees

the available

supported by the economic

by

William Samuelson
In

February

Ausubel Lawrence

80

81

Science

full

when

similar results

J and

BV

beneficial

model

Samuelsons

party

who

when

does not generally
to

information but instead

the

demonstrates
79

there is asymmetric

information

of information would disadvantage both

Leamers

empirical

his claim that the challenged

price discovery

framework

1986

Ibid

Motty Perry

study

unique

features of individual

agreements had a class wide impact

52

July

81

1984

bargaining game with asymmetric

to calculate

from the behavior

Sequential

not

evidence of Defendants recruiting and

a sequential

In order

is

correctly

of his

his

informed

Bargaining under

payoff

the

opponent

uninformed player

p 1004

Cramton and Raymond J Deneckere
Bargaining with Incomplete
Game Theory Aumann Robert J and Sergiu Hart eds Vol 3 Amsterdam

Peter

of

chapter

Leamer Report

6870

Leamer Report

50 2002

73

37

And

Asymmetric Information Academic

2 1986

M

Information Handbook
Elsevier

or

and draw the proper inferences

Grossman Sanford

Press revised
79

literature

a similar vein Grossman and Perry

must anticipate

literature

Bargaining Under Asymmetric Information Econometrica

information and obtain

see

However

Further his argument that cold calling uncovers

employees contradicts

78

some mutually

evidence shows that the challenged agreements would not

Dr

77

that

80

meaningfully affect information flows

hiring practices

81

of the proposed class

information is more favorable

with asymmetric

trades

of

to the informed party than the price

favorable

informed party than the price that would prevail with

69

members

that the informed party has superior information will take this into account

establish that the price that prevails

Indeed

of

information Rather he explains that an uninformed

full

formulating his strategy

that

Page40

informationcorresponds more

literature showing

asymmetric

does not establish that the resulting price
that

this

Filed1007 13

hypothesizes that reduced cold calling affects

under compensation

early contribution

when

are foregone

in

2

with asymmetric

bargaining

mechanism by which

the

negotiations

trades

literature

Document518

Case5 11cv 02509LHK

2 Dr Leamer Wrongly

Document518

Filed1007 13

2

Claims that His Empirical Analysis

Page41

81

of

Defendants

of

Compensation Data Shows that Restricting Cold Calling Impedes the Price
Discovery

70

In addition

Process

wrongly claiming

to

economic theory and economic

that

show

literature

that reduced cold calling limited informationflows and price discovery and thereby

ed

suppress

82

employee compensation on a widespread basis

analysis that he characterizes

Cold Calling would
discovery process

83

Dr Leamer

compensation

He

Defendant85
this

However

claims that

packages

a

show

6 and

symptom

empirical

evidence capable of showing that

instead

restricting

the price

no common evidence

that there is

of price discovery at

who moved

for those

7 which

between

compare

Defendants than for those

of

it

does not support

this

process

his conclusion

better

who stayed

84

compensation

total

Defendants and employees that

2001 and 2011 demonstrate

were meaningful

work would be

median base and

the

move between

each year between

comparison

provides

employee compensation by impeding

his data

of employees that

in

common

Leamer

compensation

says that his Figures

respectively

additional

suppress

artificially

suppressed employee

71

as

Dr

86

stay

However

at

a

even

if

that the challenged

agreements impaired informationflows and price discovery

72

First the economics

generally

receive

atypically large

occur generally and

82

83

84

85

wage increases so

the pattern

89 emphasis

Leamer Report

shows

shown by

that job changers

Dr Leamer

would

Economic theory and evidence imply

not evidence of disequilibrium

Leamer Report

mobility

80

91

Dr Leamer

There

are

thousands
and the

of

between

those

who stayed

at

a Defendant

91

but his

stayers category also

stayers category only excludes from the Defendants total employees

the individuals

Defendants

very few
stayers

movers in

Dr

his data

Leamer has

only between

not controlled

26

to

178 movers

in a

for potential differences

year compared
in

mix between

with tens of
the

few movers

many stayers

Topel Robert

Journal

added

stayers as

describes

new hires His

who moved

87

on between employer

87

is

Leamer Report

includes

86

literature

of

H

and Michael

Economics

May 1992

P Ward
Bartel

Job Mobility

Ann

P

and the Careers of Young

and George

38

J

Borjas

Men

The Quarterly

Middle Age Job Mobility

Its

Case5 11cv 02509LHK

that an

employee who moves

costs

However he

portion

of their higher

Second

the process

movers are

desirable to another

their

compensation for

same compensation

movers do
employee

increase

not increase

firm matching

if

stayers and
they had

the compensation

moving costs and what

of such

attractive

to

means that ex ante

unique

features

because

it

In other

words

The

because they are

Dr Leamer says but

as

nothing about

reveals

not evidence that stayers

of stayers

uniquely

will be

and were chosen

exactly this conclusion

supports

moving

and

generally

moved

that the relevant

movers

are selected

their

is

employers must

in

promoted

not affect compensation

than observably

must net out movers

Employees who move on average

in effect getting

81

of

for this rather than reflects a disequilibrium

movers

by which

recognizes

acknowledges

the magnitude

compensates

firm perhaps because of

movement does

appropriate

88

Dr Leamer

suppressed

is

no evidence about

compensation

stayers are not equivalent
the hiring firm

First as

Page42

in compensation

a larger increase

whether compensation

provides

Filed1007 13

2

moving Dr Leamer

employees for the cost of

disparity for evaluating

earnings

likely will obtain

employees for two reasons

similar incumbent

compensate

Document518

would have

compensation increases

economic

vast

received

but does not rely

literature

on

the

of

on

disequilibrium

or

undercompensation

73
their

Second and
movement

compares

critically evidence that

affects the compensation

the compensation

that the compensation

of

of

affects

argument requires that the change
Without

stayers

this

concerns that converts

unproven

one

Determinants and Consequences
Borjas

George

J

Topel Robert

Journal

of

compensation
for

which he wrongly

Working Paper

stayers is not evidence that

comparison

year

Yet

but provides

No 161 NBER

is

closed by

employees

no evidence

the compensation

there is

paper

May 1992

Determinants and Consequences

Bartel

Job Mobility and the Careers of Young

Ann

P and

Working Paper

George J Borjas

No

161

39

NBER

of

internal equity

Working Paper Series

Cycle Working

it

his price discovery

movers also changes
claims

is static

No

no

support

for

January 1977

233 NBER Working

1978

H and Michael P Ward

Economics

of stayers

raise into raises for all

Job Mobility and Earnings Over the Life

Paper Series February
88

stayers in a particular

in compensation

link

persons

Dr Leamers

of stayers

movers and

movers

movers earn more than

Men

The

Quarterly

Middle Age Job Mobility

Its

Working Paper Series January 1977

Case5 11cv 02509LHK

on

his claim that a restriction

Document518

information

employees of other Defendants affected

74

Thus

not support

Dr Leamers

his claim that

compensation

3

According

2005

because

its

expansion period began

we

2004

in

in

his Figure

9 he

uses

agreements was motivated
expansion with employees

average
provides

data

He

was

peremployee
support

expansion

after

from Apple

by Defendants

incentive

89

90

91

revenue

at

Proposed

of the

Expansion

these agreements were

90

members

He

put in place

of expansion

period

of informed

in

would

that

claims that this

of the proposed class

referencing

his Figure

if

8

91

his claim that the timing of the

sharing

improved

Apple

during the

profits

NonCompete

the risk of sharing

the gains
of

Apple with average per employee compensation in his Figure 9

Dr

Leamer appears

employees

to

increased

DNCC

94

Leamer Report

Members

Non Compete

a threat to the firms high levels of profits and that the comparison

434 17 18

Leamer Report

of the

Economic

Apple revenues surged and when

for this conclusion

them

at

to

rise

to

desire to avoid

to enter into information restricting

Leamer Depo

conduct or that the

with periods

to support

agreements as evidence that the Defendants incentive
that the cost to

the alleged

do

of labor markets and

claims that Apple data show that the

went into effect when

with the workforce

to

compensation than would be expected absent the challenged agreements

less equity

Agreements

why

and that subsequently

use 2010 and 2011 as the relevant

Then

members

of class

calls

members generally

Claim that the Timing

surprise

81

equilibrium

the timing of the agreements coincided

otherwise have caused compensation

received

a

of

through cold

obtained

Have Accompanied

not

Page43

operation

Compensation

Increased

Dr Leamer

to

the normal

from reaching

Class that Otherwise Would

89

of class

was impaired by

Dr Leamers

Data do not Support

Agreements Prevented

75

reflect

price discovery

conduct prevented compensation

alleged

would have been

that

6 and 7

Figures

Filed1007 13

2

98

40

to view the timing of the challenged

reduce informationflows

when

their

profits

equivalently

increased

agreements also increased

at

so their

that

time

Case5 11cv 02509LHK

Document518

2

Filed1007 13

Page44

of

81

Case5 11cv 02509LHK

Document518

2

Filed1007 13

Page45

of

81

Case5 11cv 02509LHK

79

The

change

in

heavily

influenced

Google responded

fact that

compensation

by

Dr Leamers

Thus

Economic

of Big

the

Bang

uncommon nature

is

offers three additional

class

101

a

were

points in

time

common evidence

Dr Leamers

Facebook

Claim that

Rigid Compensation Structures

Dr Leamers

analysis in support

evidence that he claims demonstrates

the link between

specific

at

81

instituting

of evidence related to factors such as

Refute

of

at Defendants

not only fails to demonstrate

AllEmployee Class102 To supplement evidence

showed

outcomes

to different Defendants

employee compensation would have been widespread
the

Page46

from Facebook by

that compensation

specific

essential element of

motion

Filed1007 13

to a unique challenge

view

factors that were

have

The second

certification

the

2

Theory and Empirical Evidence

Defendants

80

supports

discussion

but in fact highlights

D

Document518

that the artificial suppression of

extending

which

I

to

all

or nearly

above

discussed

of analysis that he says support

his claim of

members

all

that

agreements and suppressed compensation

the challenged

types

of Plaintiffs class

of

he claimed

generally he

wide spread

impact on

members

See

http

for

example

techcrunch

a

November 2007

online

com 2007 11 21 facebook

Facebook

article

Googlers At

Stealing

stealing googlers at an alarming

rate

a

An Alarming Rate
May 2008 online article

Google Finds That Perks Cant Keep Some Employees From Leaving

http

www dailytech com Google Finds That Perks Cant KeepSomeEmployeesFromLeaving article

11794 htm

at

President

of global

communications

and

public

affairs Elliot

Schrage

jumped ship to work

week Just two months prior Sheryl Sandberg had left to become
Facebook and a May 2009 Wall Street Journal online article Google Searches

Facebook

this last

the

number two

for Staffing

at

executive

Answers

startups
http

recent

online wsj com article SB124269038041932531

weeks amid

display advertising

the departures
chief

of top

102

Facebook

Leamer Report

html

including

David Rosenblatt Meanwhile

engineering director Steve Horowitz
like

executives

Inc and

Twitter

101 emphasis

Concerns

advertising

about a talent exodus
sales

midlevel employees

and search quality chief

Inc

added

43

Santosh

like lead

Jayaram

have revived

boss Tim Armstrong
designer

in

and

Doug Bowman

continue to decamp

to

hot

Case5 11cv 02509LHK

Economic

theory

principles

Defendants

Multiple

As

I

now

through a

the impact

regression

common

of the proposed class though

maintaining a
discuss

somewhat rigid

compensation
considerable

structure

difference

objective

individual

do

105

the sparse

106

evidence

is

employ

to

evidence

faculty

Journal

by

of the

rapidly

105

a

or circumstances

more

in

flexible

There

hand and

performance

and

does not

is

seniority and other

the one

in

broad labor

the

and other

my review

106

Based on

of declarations

support of this claim in detail
its

employees

market for members of the proposed

times and the unimportance
price

2003

However

by Facebook

discovery

find that

I

and

class with

of cold calling among

process

procedural

equity

is

more important than

sample

Do New York Basic Books 1984
World Can Unions and Industrial Competition
vol 22
Winter 2008 pp 153 176
What Do Unions

a Dynamic

Economic Perspectives

in

about cold calling of

at various

and Honoree
in their

B and James L Medoff

Sluggish Institutions
of

Dr Leamer

concern

information flows and the
Terpstra

T

at each Defendant

by

However he

compensation and compensation changes

offered

hiring

work

Freeman Richard

Coexist

that

would

loyalty

contributions

compensation on

in setting

goals

loyalty than

unionized workforces

to the university

For example in their

Hirsch Barry

to maintain worker

relative to other compensation

he cites relate to Googles

in contributing

equity

Dr

with

structure

to assure internal equity

salary structure

to determine

employers such as Facebook

internal

104

101

not address

Defendants

compensation

firms have incentives

of workers

by another Defendant This

new

rigid

with compensation managers

note that documents
not

a

promotes greater worker

between

characteristics

Leamer Report
I

to

internal equity

evidence are inconsistent

systems of the Defendants that rely on individual

my interviews

to

agreements on compensation

that emphasizes and rewards individual

characteristics

compensation

104

Dr Leamer

salary structure

rigid

by adhering

103

Theory Does not Support a Rigid Salary Structure

the strength of this incentive

which a

103

to

loyalty

81

of

evidence demonstrates a widespread impact on compensation of

members

According

salary structure

theory and empirical

claim that

81

maintain worker

Page47

analysis

Leamers

Economic

rigid

of the challenged

explain both economic

1

to

that he says reflect adherence

internal documents

and

principles

Filed1007 13

firm incentives

implicating

of internal equity

2

Document518

1

44

p 135

Case5 11cv 02509LHK

Document518

2

Filed1007 13

Page48

of

81

Case5 11cv 02509LHK

importance
Alexandre

Document518

of this effect or even that

Mas111

to support

his

is

it

performance

argument but Mas does not

than the one
role of

at

issue

demand and

this article

new

acrossthe board wage cuts

of

employee

translates

Dr Leamer

no

adjust compensation

hiring

call

their

was reduced by

of incumbent

the challenged

theory in

less

Microsoft

those challenged

Yahoo

or

at

any

Adobe

internal

no

decline

in

not explicitly considered

information

113

or analyzed

made

Leamer Report Footnote 126
Quarterly

Leamer Report Footnote 126
Labor Economics 243
at

referring

See

faculty

to

of

time

Defendants with

Albert

Rees

by a

pressures

employment

that

in

resulting

DNCC

to

or

place

Since

Dr

the

according

However

from which

it

if

new

The Role of

1993

87 101 18 4132 20

46

Apple did
hires from

employees

obtained

same adjustments to maintain

Leamer has

Mas Pay
2006

to

agreements than

called or otherwise recruited

Apple and

783

but

but simply that Defendants agreed not to cold

cold

Alexandre

112

received

any hypothesized

not

shown

and indeed has

whether the challenged agreements resulted

Journal of Economics

the

the information that results

and the available evidence demonstrates that there was not

Performance 121

Leamer Depo

to be

to

describes

claims

do not claim

agreements were not

but instead

who

maintain internal equity there was

to

of the hundreds of other companies

equity would have had

members job

adjustments for other

hires transmit

Plaintiffs

agreements

then the same information was transmitted

113

new

employees

employee movement between
if

by

higher compensation

Dr Leamers

employees of certain other Defendants for certain periods

not cold call employees

of

how

at all into compensation

to maintain internal equity

would have occurred

112

inherent in

in this source of informationand thus

reduction

81

the dispersion

in

salary structures of university

the fact that if

ignores

adjustments to compensation

in

on

forces

if

police union

of

article

changes

by Albert Rees

article

fairness which does not support

employees to preserve

Finally

an

also cites

the impact of market

on

an

cites

consider

the effect

Page49

a very different labor market and different event

simply emphasizes the uncertainty

or incumbent

84

Dr Leamer

here

He

context

in this

wages within an employer Rather he examines

of

111

present

Filed1007 13

2

Reference

Fairness

in

it

is

in

any

loss of

irrelevant

Points and Police

Wage Determination 11

Journal

Case5 11cv 02509LHK

Document518

2

Filed1007 13

Page50

of

81

Case5 11cv 02509LHK

informationconsidered

in

employees For example
Each manager

overall

setting an overall
in

2004

Intel

the poorest

the top performance

87

Assume

rating averaged

Assume

that the

implication
receiving

Leamers

the managers

to

much

smaller

based heavily

Intel

on

compensation

Based on my

with similar

budget that year

increases

discussion

filed in this

matter

employees

willingness

I

the fact that the total

a cold call

percent

is

116

from

salary increase

who
This

an

is

not only that the individual

five percent

generally

is

too

he manages 25 employees

market adjustment then his other 20

Thus

the exact

the budgeting

a system

where

amount depends

process
granting

that other employees receive

implied by

Dr

Leamers

amount budgeted

that drives

above average

below average

theory

for salary increases

generally

by thirdparty companies

surveys performed

on

like

76583DOC001487pptx
with compensation

understand that Defendants
to

the

and responsibilities

skills

than five percent

average increases

data from the compensation

data

a 20

market compensation

that

who

received

a cold call

received

provide information

calls

among those 20 employees

ignores

to increase

that salary increase both for the individual

some employees may require

Dr Leamer

who

for

115

ability

a job offer with

received

claim that cold

increases rather than the above

116

match

salaries

guidelines

Intels

of those employees received

give five of his employees a 20 percent

to

employee

employees were that employees

salary increase

changes at Defendants essentially creates

salary increases

115

to

salary increase

the salary distribution

compensation

2005

In

learns that one of his engineers

undercompensated but

a job offer is

employees will receive

88

when one

that the engineer

manager wants

Dr

of

and he wants

is

policies

across

over 10 percent

the job offer and for other engineers

received

If

at one Defendant

81

companies

performance ratings should get no salary increase while those

Defendant and

another

low

salary increase

for a group of employees

manager

a

in

of

to allocate her pro rata share of that budget in salary

Given this system a manager would have only limited

compensation

Page51

for the

change

a four percent

some cases by company

distributing the aggregate budgeted
received

Filed1007 13

2

budget for salary increases

budgeted

how

then decided

to her staff guided in

increases

Document518

pursue outside

reasons and would remain so even

if

managers

at the

infrequently

opportunities

as

evidence

he received an increase

48

Defendants

and

counter outside

in

that

pay

the

my

review of Declarations

offers

employee

because they consider an
is

disaffected

for other

Case5 11cv 02509LHK

Radford and
employees

on

not

117

The informationeach Defendant

that information is specific

positions

at

but his ability to

least

in

process

the short

obtain

to

much

may even

81

of

hires and separated

and similar firms is

broader than the Defendant firms and

Defendant which then

by

a raise

will not provide a basis

threatening

on which

are matched

move

to

the

company Due

compensation increases

lead to smaller

to another

would decide

the Defendant

salaries throughout

in favor of increasing

it

new

from Radford

receives

the

at

from

Page52

with the

An

which the Defendant benchmark employee compensation

might be able

do so

market intelligence
budgeting

types of jobs

the firms against

employee

individual

to

Filed1007 13

2

information obtained

idiosyncratic

from data on salaries paid in a marketplace

derived

same

Document518

for other

to

Defendant
to ignore

the fixed

employees

at

run

3 Dr Leamers
Affected by

Analysis Wrongly

Some Common

Assumes that

Factors then Only

if

Individuals

Common

Compensation

Factors Potentially

is

Affect

Compensation

Dr Leamer

89

who would

claims that changes in compensation of a small number of employees

have received

a cold call

would have class wide impact
increase

compensation

most in demand

common

at

across

any

to

of

let

However

alternative

117

Indeed

current

Plaintiff

in

part

To

support

by common

individuals

factors

provides

such as experience

because they received

that the level of compensation

Hariharan testified that he did not share
a

new

job

Hariharan

120

49

on

paid by Defendants

His analysis cannot distinguish

employers when considering

Leamer Report

claim he

this

that compensation

185 16
118

and related practices would be expected

his analysis does not demonstrate that changes

class wide

hypothesis

time

show

alone a small number of

compensation

Defendant but for the challenged conduct

the board rather than be narrowly focused
118

factors analysis

another

Cold Calling

that

point in

employees can be explained
education

from

those

the impact

compensation
at

an
to

that are

empirical

individual

job

title

compensation

a cold call

of Defendants

Depo

in

the skills

to

would

and
of a subset

affect

he hypothesizes from an

employees

information

is

broadly

with prospective

104 105 18 136 24 137 12 184 5

or

Case5 11cv 02509LHK

Document518

Filed1007 13

2

Page53

of

81

determined by competition in a vast labor market for similar employees and that adjustments for
unique circumstances

Dr Leamers

90

members tended

to

of particular employees are highly

individualized

claim to be able to demonstrate that generally

move

claim of class wide impact

119

time

over

together

Such movement

is

is

neither

the compensation

nor informative

surprising

of class

about his

As

the hallmark of a competitive marketplace

I

explained above the labor market in which Defendants compete for employees and in which

members

of the proposed class seek

extensive

flow of informationthrough a variety of channels and high employee

it

is

not surprising

common factors

that

employers and jobs

across

the market

would have

compensation

more or

pay variation

An

attracting

less attractive

11A

a

claims that

119

120

variation

Leamer Report
The degree of

across

to

121
Exhibits

Google

jobs

123

that

If

several

with

makes

Leamer never even examines

he had he would have discovered

grades and

He

many

a regression

variation
it

with job categories

will understate

ignores

and

11B show

Dr Leamer

the ability of

that both the level

within his job categories

122

titles
titles

and that

which

analysis that

see

left

built

relatively little scope

about

that

the distribution

across

of annual changes in total compensation

analyzes in his Figures

job

categories

companies

for the

pay
in

for the

using the same algorithm that

121
122

50

Apple

17 Appendix 6A through 6E show
major jobs at the other five Defendants

15 through

jobs at Apple and Google

Leamer Report

employers to differentiate

events

major at these

Leamer Report

the ability of

Dr Leamer

on

high level

he claims shows

employers to move individuals

of annual changes in total compensation

123

Dr

that

Defendants had highly structured compensation systems

identify

122

unique characteristics

vary greatly across individuals

distribution
the

not competitive

and keeping good employees At the same time

provides

cold calls or other

11A

is

130

individuals since

response

Thus

mobility

in average compensation

compensation that

ranges of salaries for grades and

established

for individual

offered

of the variation

rapid and

11B120,121

and

two dimensional matrix with

management

much

any given employer

within these job categories

Dr Leamer

91

to

by

broad and characterized

employees because each possesses

and rate of growth of compensation
Exhibits

is

explain

who

employer

difficulty

varies across

that individual

employment

uses to identify

and

the
I

the top ten

Case5 11cv 02509LHK

90

percent

age

of the variability

in a class

tenure gender location job

Leamer concludes

Document518

124

and employer

title

time can be explained by a

common

he bases this conclusion

the

a

that is explained

by

81

of

in each year

Based on his regression results

125

characteristics

that measures

statistic

The

statistic

the fraction

by

Dr

salaries within each firm at each point

in

set of employee

Rsquared

is

Page54

members compensation can be explained

almost the entire variation

that

Filed1007 13

2

in

on which

of the variance

Rsquared
in

the dependent

variable

power and

zero no explanatory

means

of 0.9 or so

variables

do

an

individuals

good job

of explaining

12 compares

92

perfect explanatory

that the regression

which

job title

Exhibit

a

one

the independent

is

power As used

has a good

equation

employer

variables and that lies between

an

fit and the independent

age

specific

Dr Leamer

by

tenure gender and location

compensation

the persons

Rsquareds

reported

in

Dr

Leamers

Figures

11 and 13 with

the

employeespecific
when

values

only the employerspecific job

age

factors of

same with

not demonstrated

claims that he has

as he

he would acknowledge

compensation

among employees

is

The

exhibit

employee

or without the

affect

are included

variables

tenure gender and location

regression is almost the

that even

title

but not the

shows

specific

an employees

largely explained

compensation

but only that variation

by employerspecific job

compensation

average pay across job categories

fact that job titles explain

mean

124

Regression analysis

many
the

126

identifying

a statistical tool to

different

individuals

the

measure
in

with different

The wide

of the Plaintiffs

because

variation

titles in

dependent
or

time periods

variables and crop yields

the

here age

by using data for

with different

dependent

to the

here a class

variables

variables

the

does not

addition

variable

control

on the dependent

and or different

However

in compensation

within job

independent

126

values for
the

amount

variables

129

variation

classes

in

different levels of average

impact on a

variables

titles

can be used to understand the relationship between

sunshine and fertilizer the independent

Leamer Report

the

one or more

characteristics

analysis

firm
wide

in compensation

impact of the independent

variables For example regression

proposed

of the

variation

annual compensation of changes

etc

of rainfall
125

is

in

a large fraction

that there is not substantial

members
tenure

wide variation

Leamer has

controlled for important employee specific factors

job titles have

is

Dr

variables

employees with different employerspecific
and there

fit of the

that the

in

pay across

Hence Dr
class rather

job

Leamers

categories
finding

is

a

consequence

of a high

than any homogeneity

51

R squared

of the broad definitions
to

a large extent

of the Plaintiffs

reflects the heterogeneity

Case5 11cv 02509LHK

variation

across

Document518

In fact the data used in

titles

Filed1007 13

2

Dr Leamers

Page55

regression analysis

of

show

81

exactly

the

opposite

A simple test

93

employees

individual

12 and 14

Figures

of the ability

shown

is

drawn from

schedule

compensation

The

2008 and 2009

in

which

is

that remains unexplained

Leamers

Figure

compensation

Defendant
example
actual

explain

in Exhibits

12

that

Leamers

there is substantial

in

2007

and predicted

between

regression

dispersion

127
Exhibit

14B shows

between

plus or

minus

of

compensation

rigid

15.4 and 1.7 percent

more

8.4 percent

predicted

their

shows that while an
dispersion

accounting

and actual

an individuals

in the unexplained

15 13

In the

same year

differences

based on

52

title

of his

based on the results in
compensation

Dr

and the

In each year and for each

portion

9

job

individuals

of compensation

the percentage difference

5 11 5 and

resulting

employee specific factors

actual

127

predicts

Plaintiffs

remains in the portion

for

the distribution

model

named

of

percent

or

for

Dr Leamers

or

Figure

For

between

more for half

Apple Google

the percentage difference

compensation was about plus or minus 25 percent

the distribution

for

there is considerable

compensation was about plus or minus 10 percent

Intuit Lucasfilm and Pixar respectively

and predicted

after

in the middle of his conduct period

Adobes employees and

actual

even

which shows

of the difference

Dr

the

Second

compensation

wide

his compensation

14A

Brandon Marshall

models

2006 than

respectively

factors regression analysis

compensation

is illustrated

in

were overcompensated

Plaintiffs

Dr Leamers

and actual

predicted

helps

to

from his

with a rigid compensation system

and employer

This

named

in the difference

inconsistent

between

difference

of

conclusions can be

2006 and 2007 respecively and 37.9 and

years and individuals

Dr Leamers common

from

in

Two

while Daniel Stover received

predicts

compensation

use regression estimates

of the Defendants

greater compensation

than predicted

than predicted

across

based on

compensation

I

to explain

would have been earned by each named

that

he was employed by one

Dr Leamer

by

asserted

compensation

94

regressions

13A and 13B

in Exhibits

17.4 percent

received

less compensation

variation

Leamers

this table First for the most part the

relative to their predicted

example

Dr

to predict the compensation

in each year that

Plaintiff

of

more

14

for

10

of

Intel

between
percent

of

Case5 11cv 02509LHK

Adobes employees and

plus or

Document518

minus

44 40 14 29 22
128

Intel Intuit Lucasfilm and Pixar respectively

imply that 10 percent

his regression estimates

Filed1007 13

2

and 30 percent

Based on

of Googles

scope to

pay

differentiate

129

Therefore

studies

structure

employees

of over 4,000 specific

to

Dr Leamers

model demonstrates

who

earned by employees

his regression which

employees even

across

contrary

his regression

by

predicted

in

2007

salaries

accounting
percent

for

of the

demonstrates that there was wide
job classifications

he

claims of a relatively rigid compensation

that there is substantial

year included

in

in compensation

variation

have the identical values of the characteristics

job titles in a typical

Apple Google

more than 40

or

the narrow

within

for

81

of

the 5th and 95th percentile

approximately 900 classmembers earned less than 40 percent

average compensation

Page56

including

Dr Leamers

being

regression

in

model

one
130

This evidence shows that the Defendants did not have the type of formulaic compensation
structure

would

that

challenged conduct

95

Even

unexplained
based on

a given job

Dr

Leamers

Dr Leamers

compensation

128

there is large variation

regression

and the compensation that

which was

The

title

132

In

2007

Exhibits

model

amount

the

in

15A

the

of Software

Developer

the actual

predicts for the top ten

Engineer 3 employees at

Dr Leamer

job based on the algorithm

simply because observations

the distribution

between

model

regression

of compensation

15B show

and

of the differences

Dr Leamers

ten percent

the top ranking Apple

fits Intel data better

regression

model

Figure 12 regression

Apple and Google jobs
Apple

would be class wide impact from

131

within

by

Plaintiffs claim that there

support

from Intels employees

used in his

such a

constitute

highperforming
large
129

about 60 percent

portion

This

of

course ignores

of the regression

the firms ability

and otherwise potentially

data

to differentiate

undercompensated

compensation

employees

across

into

new

employees

jobs

by moving

with higher average

compensation
130

According

over 100,000
60,000
131

As

employees
I

noted

compensation
across
132

The

shown

Dr

job

Leamers

employees
in

2005 and 2009

about 7,000 different

the

proposed

All Salaried

job titles and the proposed

Employee

Class

Technical Class includes

includes

over

about 2,400 job titles

above even
in

response

these

figures understate

to external

the flexibility that

the

Defendants

pressure because they ignore Defendants

had to differentiate

ability to

move

individuals

titles

distributions
in

data between

in

Appendices

of the of differences

7A

through

in actual

and

predicted

7E

53

compensation

for the other

five

defendants are

Case5 11cv 02509LHK

Dr

by

predicted

again

necessary

to

model

job
by

predicted

Together

assumed

and

because of other factors
were not harmed by

his

which any

Model

VD

minus

percent

that individualized

was under

individual

analysis

would be

or overcompensated

from the

of undercompensation

the conclusions

compensation

that

Does not Demonstrate

discuss

Dr Leamer

in

his Hypothesized

Price Discovery

detail the regression analysis

which

offers as evidence that the challenged

and damages allegedly

Dr Leamer

suffered

draws from it are inconsistent

that allows

structure

him

to

infer that the

by

refer to as

I

conduct affected

price discovery to affect

all

members

his conduct regression estimates I simulate

with his claim that there

assumed loss

identical individuals

based on the empirical

year for each Defendant

133
I

person so

in

that

subsequent

into

account

2004

the

I

result

only difference

years

compensation

and thereby

the persistence

for individuals

otherwise similar people

compensation can

diverge

shows how

Assume

that

in their
predict

effects
I

identical

in all characteristics

54

Using

compensation in each

two

compensation can diverge

who

individuals

are

comparable

in their residuals

compensation

each subsequent

based on

2004 The

over time

common way

is the difference

in their

from the prior two years

were identical in

equity causes

unexplained

performed the same experiment 50,000

who

internal

over

in all

unexplained portion of their compensation for each

2005 compensation

the difference

rigid

over time of otherwise

dramatically

there are

or the

a

in compensation

distribution of

randomly draw residuals

randomly drawn residuals

of

The

perform the following experiment

characteristics

in

133

is

of informationand

of the proposed class in a

the change

the

the Class That analysis and

reduced price discovery combined with Defendants commitment to
localized

who

challenged agreements

aggregate compensation of members of the proposed class and that he uses to estimate

amount

of the

Cannot Explain Compensation Changes

it

conduct regression that

employees at Google

III

paying damages to members of the proposed class

my report I

of

Engineer

because of the challenged conduct rather than

structure

to avoid

Process Because

In Section

to

81

of

2007

and indeed could have benefited

4 Dr Leamers
97

in

on compensation means

wage

rigid

model

Page57

or less than the compensation

of the employees earned plus or

Dr Leamers

the evidence

more

Similarly for Software

percent

determine the extent

relative to the

Filed1007 13

2

earned approximately 42 percent

Leamers

the top ranking

compensation

96

17

15 through

Figures

Document518

that

Dr

in

Leamers

times to obtain

resulting

distributions

Dr Leamer

model and

I

do

the

same

year taking
the

a distribution

new

of

resulting

show how compensation

claims explain

an individuals

Case5 11cv 02509LHK

Document518

due

time for otherwise comparable individuals

Dr

by

that remains unexplained

Leamers

Filed1007 13

2

to the portion

conduct

of

an

Page58

individuals

81

of

compensation

which

regression each year the effect of

cumulates over time

98

Compare for example two

characteristic

controlled

tenure and location

Dr Leamers

for

Dr

Leamer

model

implies

e 2004 and

that these

By 2009

by 24 percent

demonstrate that otherwise

Thus Dr

compensation
compensation

schedule

134

Such

conduct

own model

His

somewhat rigid wage

structure

results

regression

estimates

results provide

the proposed class could

no

the compensation

shown

model

different

a

his claim of

contradicts

imply that he has no basis

to

salaries

17

end up with tremendously

rigid

conclude that

class wide effects through his claimed

to

in both the levels

As such Dr

even within job categories

for Plaintiffs claims that the

support

By 2006

two

of the

16 and

in Exhibits

Empirical evidence shows wide variation

and rates of growth of employee compensation
Leamers

e 2003 compensation

between

changes in compensation would translate

individual

age gender company

i

prior

identical employees can rapidly

Leamers

are identical in every

two employees would on average have

the difference

would be around 37 percent

individuals

2004

of

in his conduct regression

i

as well as current

conduct

that differed

by

who as

employees

amount

of

harm

members

to

of

be determined on a class wide basis

5 Dr Leamers

Attribute Compensation

Constant

Ranking

Analysis

is

Misleading

99

compensation

I

excluded

and by

Lucasfilm

of the pairs
at

least

86

18B show

and Pixar

If

differences

would

the

likely

evidence

masks

years

in compensation

Exhibit

exhibits

situated

the

17 shows

The

55

variation

two firms

90th percentile

the

I

have converted

the analysis to include

Exhibit

16 shows

figures indicate

by at least 56 percent

the entire distribution

were promoted

be even larger

relatively stable

and expanded

that differed

numbers assume

employees

of

substantial

because of missing data for those

after five

identically

titles

would have compensation

of employees

percent

two

Dr Leamers

both by firm and overall

differences

from 2005 through 2010 In both
each year

that

into annual changes

and the 90th percentile
percent

and

within and across job

trends

15 17

his Figures

134

18A

Exhibits

two employees
at different

of differences

the

that
after
in

mean

ten

two

years

each year

have the same job

rates

the

title

then the compensation

in

Case5 11cv 02509LHK

25 job

top

titles

Dr Leamers
fashion

groups

linked

across

Dr Leamer
structure

15 17

The

at

his deposition
136

not rigid

18A

Exhibits

that such evidence

Moreover

and

link of the

form

required

relevant

than compensation

changes

in

compensation

by

Dr

in

18A

within Defendants

constant

and

the overall

upward

theory Changes

Indeed

was

substantial

compensation

compensation

in

Dr Leamers

relatively stable then one

variation

across

titles

in a year and

mask

that Because

does not

are

would be

and

from year

claims that
If

Dr

this

Some

across

101

Dr Leamers

logic

level that substantially

to

his Figure

135
I
136

show

exceeds

7 ripples through

charts

Q Could

implies

for the

Would Move
that

year The

and
scale of his

variation

had

Defendants

of

that of current employees
the rigid compensation

a nonrigid wage structure

8A

forces

Policies

and Even
of

the Structure

a small number

top 10 jobs in Appendices

that

job titles

Formulas for Annual Compensation Adjustments that a Limited Number
Additional Cold Calls

Leamer

a given year

large positive

to

more

would expect

similar in

Defendants

are

which would be driven by market

trend in compensation

One Cannot Conclude

in

compensation changes with

attribute

a wage

stable ordering

changes in compensation for others

titles

independent of any internal equity concerns

6

are in any

true would not be sufficient to establish a

if

for different job titles also

changes in compensation across

negative
figures

in

even

Leamers

for one group drive

18B show

and

across

alone causally related across groups

let

levels because they more closely proxy

average compensation

But Exhibits

a roughly

135

parallel

consistent with

limited claim that changes

which

were correct that compensation across job
changes

would be

the maintenance of

18B examine the

similar for the different job categories
causal

in a

of compensation

fact that the ordering

even mean that changes in compensation are correlated

100

and Google

job titles for Apple

imply that compensation increased

81

of

through concerns about internal equity or other forces

job categories

was

10

Page59

stable over time does not imply that changes in compensation

admitted

that

Filed1007 13

2

rather than just the top

highly misleading

is

job titles is relatively

way

Defendant

for each

claim that his Figures

across

Document518

and

as you’ve defined

Depo 283 23 25

56

movers

eg 20

hired at a compensation

or

30

percent

structures of Defendants

higher according
to

cause an

8B
it

lead to parallel lines

A Yes

it

could

Leamer

Case5 11cv 02509LHK

or similarincrease

equivalent

unreasonable to expect

earned by a

amount

cold

compensation

not

who

the person

compensation

new

102

hire and not

One way

in

if

compensation

would

it

is

Because

is

by

75,000 and a senior software
engineer receives

110,000

The

with

an outside

firm can respond

perfectly

rigid

compensation

by promoting

impact

E

let

causing a

of all

employees

employee receives

to a position

if

an

outside

higher

that provides

on compensation

within

wage

the firms

engineer

125,000 and

a job title
structure

position

a junior

Dr

to

of

senior software

ripple effect on compensation
Leamers

Rsquared would be 1

Dr Leamers
The

third

Model

Econometric
it

is

between

of

no

compensation
the challenged

of

would show a

regression

yet there is

Undercompensation

Flawed Both Conceptually and

issue addressed

by

Dr

Leamer

is

whether

be used to demonstrate that the challenged agreements
of

any

if

ripple effect

of junior

and

conduct had any

alone a class wide impact

Impact Because

103

a valued

no informationabout whether

provides

higher

directly gain information

the junior software engineer

and the rigid structure reflected in the fixed relationship
senior software engineers

job but

job titles a junior software

Using data such as this
his

the

25

as the result of a cold call with compensation

offer

structure

who

engineer position that pays

125,000 in compensation without

other junior software engineers

substantially

employees only

which occurs even

two

by a

in compensation

increase

regression focuses

firm has only

required

it

cost the firmnot only

current

with a promotion

countering

not be able to identify this type of effect

that pays

engineer

substantial

Dr Leamers

rigid To explain assume a

software

a

required

or even

than

employer would be willing to offer a

only employees

to

which an employer can respond when

offer of higher compensation

total

it

An

hire if

performed

compared with

increase

make the

percent

previously

the proposed class

in

impact were limited to similar employees or
from the

by 25

greater compensation

employee would

called

economically

is

it

81

of

whether to offer employment to a

employee would

cold

Page60

employees But

percent

of all salaried employees

for all employees

a substantial

of current

who demands 25

call

In effect hiring the

Filed1007 13

2

A firm considering

current

more than was earned by

percent

mover

a

attribute

compensation

increasing
lesser

constant

in compensation

to occur

this

identified through

candidate

Document518

members

of the proposed class

He

first

presents

57

in its

standard

Fails to

Show Common

Implementation
econometric

analysis can

generally suppressed the compensation

a simple analysis of the change

in total

Case5 11cv 02509LHK

compensation

an

for Defendants

claims that this

he

analysis

later

Filed1007 13

2

described

Page61

warm up

as a

evidence that there was undercompensation

suggestive

is

Document518

81

of

He

exercise

during the class

period

Dr Leamer

then performs a

undercompensation

from the alleged

104

conduct regression

He

agreements

to

attempt

indicated

at

the aggregate

to quantify

his deposition

that his

conduct regression also had a central role in making up for the lack of direct supporting
evidence for other parts of his theory
to

and

evaluate

agreements

quantify

amount

105

Thus

Dr Leamers

forms of econometric

conduct

process

if

the

amount

of computing

his

conceptual framework

other independent empirical

logic

links the

it

of information reduction

evidence

informationand thereby

hampered

structure

106

to

each

However

or almost

and

Dr Leamers

and

conduct

137

Leamer Depo

at

413 21 414 7

cold calling declined
identified
138

139

in his

price

period

framework

at

Non
support

for

which he has no

the lynchpin in his chain of

on compensation

somewhat rigid compensation

the proposed class

common

not

that

in multiple

common impact
to all Defendants

he does not have evidence

but stating that the regression

is actually

there

135

Leamer Report
Leamer Depo

discovery

is

acknowledging

during the conduct

is

for

standard

of compensation

but also as empirical

price discovery

regression is also flawed

effect he estimates

amount

to a measurable impact

implementation masks the fact that he has no evidence of

undercompensation

139

through his claimed

each member of

demonstrate that

agreements reduced cold calling and

price discovery

both on average for the proposed class

substantial

Employee Class caused by the

His conduct regression

that the challenged

possibility

to

the aggregate

aggregate damages

and to measure

was

137

suppression to the AllEmployee Class and Technical

Compete Agreements

though he lacked evidence

of information lost

regression is not only offered

analysis are capable

138

said that even

because his conduct regression would show

during the conduct period

to affect the price discovery

he

information that was lost because of the challenged

of

he did not need this evidence

undercompensation
enough

the

In particular

24 825 36

58

will

ways

First his

but rather that the

When

on whether

disaggregated

the

decide whether

amount

of

the effect

Case5 11cv 02509LHK

by Defendant

overcompensated

independence

fails

his

model

by each

implies

that his

are demonstrably

of

with his claim of a

simple sensitivity tests

107

Before presenting

his Figure

compensation

calculated

illustrating

how

by contrasting

the before

for the

US economy

he can

illustrate

employs

in

Defendants

the

relevant

140

141

142

143

144

Motion

and

after

Demonstrates
econometric

NonCompete

the

after

warm
up

works

143

2001 and 2011

if

comparing

in relevant periods

before

No Common

analysis

his review

144

at

379 910

Leamer Depo

at

380 21
140

59

that

Impact
in

on employee
when

analysis
of

provides

the Agreements
141

142

a

At

growth cycle

18 he

his

way

of

periods

concludes that

damages which he

for measuring

and after

showing

of

Dr Leamer

Agreements

the regression

company

no

Members

the average change

136

fragile and

is

140

in compensation

at

also

the

during the conduct period with

He

selects

periods because he concludes that these years

Leamer Depo

Leamer Report

to

of

regression provides

summarized in his Figure

they were a single

lack

NonCompete Agreements

Based on

p3

Leamer Report

conduct

model

compensation during the periods

before and

model by
as

The

structure

regression

or virtually all Class

before and after methodology

pooled together

non conduct

Leamers

standard

analysis as a

between

his regression

what was happening

all

of the effect of the

this

determined independently which

class wide methods and evidence are capable

his proposed

were in effect with compensation
deposition he referred to

Leamers

Dr

For these reasons

statistical

imprecise and are not reliable

Dr

Warm up Exercise

an estimate

19

are highly

impact Third

suppression of compensation affected

1 Dr Leamers

is

somewhat rigid compensation
estimates

Dr Leamers

Second

false In particular his analysis assumes that

a Defendants employees

model

for Plaintiffs claims that

support

81

of

employees during the conduct period

or proof of class wide

estimates

Page62

conduct regression model would show that some Defendants

received

inconsistent

Filed1007 13

own

assumptions regarding

is

2

their

his

compensation

Document518

2004 and 2011

reveal

the kind of

as the

Case5 11cv 02509LHK

compensation

Document518

that occur in expansion periods

increases

Based on this comparison he calculates
for all Defendants

108

as a

whole

Dr Leamer

While

19

19

I

model

warm
up

by an expanded

version

Dr

Leamers

defendant by defendant

implies

that the

and

theory

and

as

Dr

implies

Defendant that allegedly

109

However

substantial

as

at

same

he

the judgment

type of effect

in Exhibit

underlying differences

according to his methodology

undercompensated
overcompensated
by more
Leamers
individual

145

146
the
147

148

than

27

their

percent

that
147

19 Dr Leamers

Defendants

was

better to

because the

firms

namely undercompensation

at

Plaintiffs claim of

each individual

Figure 19 methodology

undercompensation

in estimated

Intel

it

his

the conspiracy

at each

and Google are the only two companies

masks

Defendant

Indeed

that

employees during the conduct period while the other five Defendants

employees Apple employees
and Pixars

he

estimates

for

example were

employees by more than 70 percent

Figure 19 shows an overall average

result

an average

that blends together

negative

overcompensated

This suggests
opposite

effect only because

two

effects

that

Dr

at

out of seven

140

Leamer Report

Dr

their

in

rather than

examined

model

more efficient

In

did in his Figure

individually

a minimum suppression of compensation

participated

shown

145

2005 2009

19 analysis

during the conduct period should be found for each Defendant individually

common impact

for

Leamer

his claim that

made

then

pool across firms in order to create a more coherent
148

81

problem with his

of his Figure

that I perform the analysis for each Defendant

are sufficiently similar

of

2005 2007

percentages

a fundamental

and after comparisons

before

them together

simply pooling
regression

is

similar to

underpayment

estimated

Figure 19 as a

uses

perform the same

The only change

that were

Page63

146

regression analysis can be illustrated
Exhibit

Filed1007 13

2

Leamer assumes

that there

would have been no change

weak economy
Leamer Depo

at

364 24 365 16

Leamer Depo

at

364 20 365 7

60

in

compensation

in

2008 and 2009 because

of

Case5 11cv 02509LHK

Defendants show a negative

110
for

impact and one

of the employees in the

percent

warm

This

Dr Leamer

whether there

up

him pause about

of the

when

Class

of

81

Intel accounts for about 60

by company should have

large magnitudes

of his methodology

Page64

149

whether his regression

to consider

rather than reflecting the impact

Filed1007 13

two Defendants

disaggregated

common impact The
the ability

2

AllSalaried Employee

exercise

and caused him

is

Document518

is

raised a red flag

capable of determining

of the effects also should have

conduct

to identify the effects of the challenged

of other factors that differ between

the conduct

and

given

non conduct

periods

2 Dr Leamers Common
Demonstrated

111

After

using Figure 19 as an illustration

analysis which he says

is

defendants as well as for
explain

model
Figures

112

total

Impact across Defendants

a

better approach

employees

150

Dr Leamer

because

He

he calculates annual

his Figure

conduct had

150

151
at

See

Dr

model

among

that attempts

20 and

his Figures

percentages

because

it

23

151

to

Using this

by Defendant and year

his

is

Leamers

not

common

impact

Defendants

Figure

Leamer Report

at

Dr Leamers

regression

lower compensation

Once

disaggregated

common

at all

Defendants

by Defendant

and for

all

Dr Leamers

impact and implies instead

that the

Defendants

3

141

Dr

across

underlying

assumes rather than establishes or demonstrates that

regression analysis completely fails to demonstrate

149

a regression

19 analysis the approach

common

or virtually all employees of the

impact

conduct regression

his

allows for differences

employees

undercompensation

analysis is fundamentally flawed

alleged

presents

24

As suggested by

the challenged

it

constructs

annual compensation of individual

22 and

Assumed Not

is

in his Regression

Leamer

said

his deposition

that

total

compensation

121 821

61

is

the relevant

variable

for

analysis Leamer

Depo

Case5 11cv 02509LHK

a

Summaryof Dr

Dr Leamers

113

indicator

variable

and

variable

variable

Leamers

model uses

regression

year as the dependent

An

Document518

includes

when

for

This variable

is

the interaction153 between

linger over

152

He

includes

employees within

Dr Leamers

Because

variable

to

in those

In
effect

purport

two years

154

of the

two

The hiring

for

variables

rate variable

Leamer Report

allegedly

variables

on

participated

in

that represent

and age squared and

age

reflect

how

a time trend and employer

normal

variation

in

over time and across Defendants

on

these variables

the effects

to measure

the extent

Dr Leamer

compensation

not full

across

155

with age and hiring

along with average employee

to which

then combines

indicator

caused by the challenged agreements

initial annual undercompensation

age and hiring

by company

undercompensation

in

one

his initial undercompensation

conduct years he assigns a

on
is

interaction
the

of

independent

measured

two explanatory

value

Since his
year

estimates

of 0.5 and 0.25 to the conduct

variables

measures

the multiplicative

or

joint

variable

as the log

of the ratio of

firm in the previous year
155

conduct

the

respectively

model an

a regression

also includes

which he claims

effects

undercompensation

calculate

2005 and 2009 were

153

effect

variable that is turned

that Defendant

and employee

characteristics

these to control

estimates

company

variables

industry

the industry

remains in subsequent years

152

zero one

in

Defendant 154

the immediate

uses the coefficient

persistence

variables

claims that his conduct variable alone and interacted

identify

rate at a given

or

in each

time

variables

He

Leamer

the conduct variable

characteristics

rate together

Dr

81

of

of each employee

agreements were

dummy

essentially a

Page65

Compensation

the following independent

or lagged compensation

Employee

114

of

real annual compensation

the challenged

any of the challenged agreements

Persistence

Model

during the period when

for a particular Defendant

the hiring rate at a given

Filed1007 13

2

142

62

new

hires to the

number of employees

at the

Case5 11cv 02509LHK

and the persistence

22 and

24

115

At

measures

Document518

Filed1007 13

2

his deposition

Dr Leamer

the average impact

for

acknowledged

of the conduct across

Defendant

that the conduct variable

Thus

Defendant

Defendants

all

measurement

specific

he claims that

but claimed

of the impact

However

conduct will differ at the two Defendants

his

model

same

even

if

assumed

same

for

those individuals

for a Defendants

provision

still

the impact to be the

forces

the hire rate is the

makes no

firms is

of the effect across

unique

Defendant

another

at

at

individuals

employed by

characteristics

each

of

any

to note that

by

his

same age

of the

his

are younger or

rather than demonstrated

two

are

that the

the aggregate impact
important

is

it

model

because he interacted

that employees at one Defendant

the extent

to

employer has a slower hiring rate than

commonality
since

his Figures

in his

conduct variable with variables measuring the age of employees and hiring rate

alleged

81

156

regression allowed

their

of

by company

annual undercompensation

effects to calculate total

Page66

model

as long as

He

different Defendants

to affect the potential

impact of the

challenged conduct even though his theory says that he should

b

Once Disaggregated Dr Leamers

Undercompensation

116

Given the nature

challenged conduct was

of Plaintiffs allegations

common

across

Defendants

used the regression framework offered by

have

to estimate

done

of individual

criteria

156

157

is critical

observations

for belonging

Figure 22 presents

technical

and

22

employee

The

Dr

as

for

to the

Leamer acknowledged

a

particular Defendant

class some

his results for the

class I will focus

central

conclusions

Dr Leamer

Leamer Depo

at

on

a class

to address

this

allsalaried

my

are the

analysis

same

employee

on

365 816

63

wide basis Thus

question

which

annual data for

class

all

are estimated

Dr

One way

employee

putative
class

I

the large

employees

who

is

have

to

do so

Leamer claims

to

number
fit

the

only because he can

while Figure 24 presents

his results for that

for the technical

whether there

in his deposition despite

of his coefficients

results differ
157

understanding

to

separately for each Defendant

the regression

however

whether the impact of the

the question

class wide impact and whether the impact can be measured

would be

Does Not Show

Regression

for All Defendants

class

though

his results for the

his exhibits

some

Figures 20

of the individual

Case5 11cv 02509LHK

combine

158

Defendants

data across

Document518

Therefore

in order to test whether

challenged conduct was similar across Defendants
conduct variables for each Defendant

separate

estimate

117

conduct effect

of the aggregate

Defendant

20

estimates

I

show

22 and 24

Figures

I

a combination

reflects

In stark contrast to

for all Defendants

common

undercompensation but

throughout

The

during the period

agreements according

magnitude

also

to

and

Plaintiffs

Dr

overcompensation

show

Leamers

Dr

and the sign of the estimated

2005 and 2009
conduct

for

Apple

generally

are

positive

two

for one or

hub

much

years

of the network

of

Dr

smaller than

the conduct regression

undercompensation

reported

In fact for

for most years while

the

from

results

the

Lucasfilm and

estimates

overcompensation

Leamer

impact

undercompensation

overcompensation

estimates

from Dr Leamers

Leamers

Class two Defendants

once disaggregated by Defendant

substantially

Dr

impact from the alleged

instead

undercompensation

Thus

results

not only differ

Dr

specific

of these disaggregated

in every year between

Google shows

the period

Adobe and Intel

other companies

Leamers

that includes

Defendant

of the regression analysis

nature

both the AllSalaried Employee Class and Technical

two

separate

81

of the

of his regression

and also includes

159

results

disaggregated analysis does not suggest

percentages

the effect

of

from the disaggregated model and compare them with

results

percentages

negative

Pixar show no

use a version

Page67

estimates

specific

In Exhibit

Leamers

By

with age and hiring rate

interactions

Filed1007 13

2

but also vary greatly across

in both

results

companies

and over

160

time

158

Dr

sector
the

Leamers

model becomes

period
159

2003 2011

regression

includes

overspecified

2006

160
Results

unavailable

As a

multiplying

outputs

a Pixar specific

which were

result

when

many

its

to

variables

Dr

is

for

that

only vary by year

example

the

new

for

hire ratio

using annual data from a single

effect

Leamer

I

See

was

example change
variable

company

As

for the

in

IT

a result

nine year

estimated

from the disaggregated model are provided in Appendices

conduct

2006 revenue

the reported

have included in
Pixar

reported

revenues
for

my

regression

2005

Pixars

9A and 9B

revenue

data

In order

after

2005

2011 xlsx Pixar was acquired by Disney

only nine months

I

annualized the 2006 number by

number by 12 9

of disaggregating

by Defendant

regression In Appendices

10A to 10C

an

specification

aggregated

estimated

over which the regression

Detailed regression

to estimate

in

conduct

employment in San Jose or by company by year

regression

I

can also be illustrated using a simplified
provide regression
that includes

details

a single

64

and

conduct

version

undercompensation

of

Dr Leamers
estimates

variable and excludes interactions

from

Case5 11cv 02509LHK

Dr Leamer

118

Document518

Filed1007 13

2

should not be surprised by the finding

Page68

uncommon impact

of

Defendants if as he believes the challenged agreements had an actual impact
spurious

or unrelated

differences

theory

These include

firms that

differed

119

At

so

He

efficiency

Dr Leamer

e

i

to

offered

no

conduct

Leamers

the potential

shown
the

in

hiring

with the single

11C

Appendix

four Defendants

undercompensation

162

163

rate

the level

of the

varies across

a firmand that pressure

information was reduced

the

and

quantify

would have been reduced

Comparison

11A

model except

conduct

Pixar Lucasfilm

to

still

and
to

11C

I

Adobe

at

at

the impact

However

gain from
of the

of the ability

to analyze

Figures 22 and 24 shows that the

the results differ

somewhat from Dr

undercompensation
specification

will

details

and

for all Defendants

be informative

as to the

undercompensation

and even the sign of the estimated

increase compensation

and

The

differ substantially

364 8365 1

Leamer Depo

of the efficiency

had no undercompensation

257 814

Leamer Depo

163

results across

in

the expense

provide regression

percentages
at

and decided that the

now disaggregate
the model by interacting
company
now there are seven Defendant specific conduct variables

vary greatly in magnitude

Leamer Depo

shows

using the simplified

variation in the size

was

While

he had done

that I

so

variable

there is large

differences

Dr Leamers

with

Employee Class
version

is at

if

the

was reasonable

wrong conclusion about

efficiency

by

results disaggregated

of the data to identify

he weighed the importance

that results

In Appendices

impact of the challenged agreements

161

how

the gain in

This suggests

from the simplified

seven Defendants

other

to which

data for all Defendants

Technical Class the simplified

impact of disaggregation

indicators

at

structure

other than acknowledging

for reaching

When

and age and

throughout the period

estimates

the ability

very similar for the All Salaried
for the

wage

said that he did not provide

combining

specifics

challenged agreements

results are

for internal equity

examined the models defendant by defendant

have

and no explanation

pooling against

between

would depend on

demand

and that the extent

impact of the challenged agreements

gain from pooling

Defendants

Leamer acknowledged

firms

claimed

Dr Leamer

and was not just

161

Defendant because efficiency

162

Dr

of the impact

pressure will affect the general

his deposition

hypothesized

his deposition

the fact that the magnitude

the situation each firmfaces

across

At

for internal equity and the fact that

outside

on

depends

agreements

across

Defendants that would translate into different conduct effects even under his

across

demand

firms

to those

81

of

365 3366 2

65

but rather

estimated

from

negative

Dr Leamers

effects

As

Three

the estimated

impacts at the

of

Case5 11cv 02509LHK

the question

at

companies

namely whether

issue

economic

consistent with the

were more

estimates

of

Thus

further

below show

substantial

3

The

Statistical

significance

variation

of his estimated

However

data

an individuals

122
the

amount

case

the

Even

reflect

if

compensation

is

that the efficiency

to conclude

methods which

that his

I

critique

impact with some

164

Analysis

Improper

is

t values

errors and

to test the

thus to test his hypothesis

Dr Leamer

these values

is

more

of those of other

that the

assumes that

employees

then the

whether the regression

is

insufficient

is

especially

a major

is

a

under his

identifying

and

inference

the larger

the data although

observations

misleading measure

in

this

of the

an underlying relationship

in the

would differ by job type

unsubstantiated

if

of individual

highly

theory of

reliable estimates

However

variation

own

error in statistical

statistically

number

to capture

because that effect

and obvious feature of his

a critical

are estimated

years for the Defendants

informative only under the

not consistent

ignores

This

provides

the coefficients

impact

system that imposes

an assumption that

and

not independent

determined

model

by Defendant

were any

are

Dr Leamers

In calculating

common impact

statistically

there

Defendants

a

employee

the disaggregation

agreements
across

of

to evaluate

ability

164

number

is

a regression

of data with which

voluminous largely

conduct

reports standard

Dr Leamer

are correlated

compensation

All else equal

Leamer

statistical

of the challenged

coefficients

only because

significant

that his observations

how

he did

few

at a

impact of the challenged agreements on compensation are highly

the estimated

statistically

and

employee is independent

of each individual

Dr

81

or indeed any impact

Defendants in the estimated

Underlying

22 Dr Leamer

as

of

were not meaningful

his estimates

across

as the result

Page69

few employees

to claim

senseless

regression specification

conduct reduced employee compensation
the compensation

is

call a

Doing so simply allowed

Framework

20 and

In his Figures

common impact

to cold

failure

it

precise even though

overcompensated

employees

how

Filed1007 13

2

or not there is

harm

appropriate

is

Dr Leamers own

120

statistical

theory

translates into class wide

gain from pooling data

121

Document518

wrong

impact of the challenged
Analyses disaggregated

assumption of a rigid

formulaic adjustments across all types of jobs locations etc within a firm
with the evidence

66

Case5 11cv 02509LHK

between

and the labor empirical

Dr Leamers

123

sample contains

and
167

labor economics

employeryear so

a

his statistical

in

literature

Page70

the econometrics

81

literature

Dr Leamers

problem

of this

of

165,166

particular

in

severe version

observations but fewer than 60 unique combinations

thus effectively fewer than

This means that

Filed1007 13

2

widely recognized

over 500,000 individual

conduct variable

Dr

60

from which to estimate

observations

Leamer has almost 10,000

analysis greatly overstates

his

per group

observations

of

per

sample size and the

the effective

of his estimates

resulting precision

Dr Leamer

treats

each Defendants

employees

as

an employer But the supposed

rigid compensation

he

or she provides

completely
determined at

is

and thus lack of independence

structure

economic framework on which he

feature of the

if

about the underlying structure by which compensation

independent information

critical

is

regression suffers from

employer and year

124

This problem

the variables

generally

Document518

for his conclusion

relies

is

a

that the

challenged agreements to reduce cold calling reduced price discovery which rippled through the

members

compensation

of all

compensation

system He

influence

their

This problem

for all Pixar

is

well known

and James MacKinnon
within a given dataset
ignore

. is

it

and use

when

166

Dr

of ignoring

thought that the within group

OLS standard

errors

OLS

are

many

H Econometric
D and Jrn

Angrist Joshua

Leamers

because he lacks

University

conduct
revenue

firm are affected

movie

In

their

correlation

per

Press

regression

statistical

between
it

that

in large and

econometrics textbook

. is small

that

commonfactors

demand

in the

test

for

PCs and

Russell

across

the

Davidson

observations

may be tempting

This can be a serious

to

mistake unless

can be drastic underestimates even with small values of .

sample

group Davidson

size

is

per group

much

Pischke

is

smaller

Russell and James

large

The

correlation

than the actual

G MacKinnon

sample

of
size

Econometric

p 305

Analysis 6th Edition Chapter 9.3.3
Steffen

by

rigid

Pixar can result

at

the correlation

severe when the number of observations

observations

Defendants

performing his

covariance matrix

and Methods Oxford University Press Inc 2004

Jersey Princeton
167

the

particularly

is

Greene William

2008

it is

a

econometric literature

the potential error

terms within groups means that the effective

there

Theory

If

at

of the

or a short term reduction

estimation with the usual

zero since

actually

The problem
the error

OLS

employees

in the

describe

employees

eg a highly successful

compensation

unusual bonuses

when

failed to take into account

from the challenged agreements

aside

165

of the proposed class because

Mostly Harmless

New

Jersey Pearson

Econometrics

Prentice

Chapter 8.2

Hall

New

2009
includes

data for Lucasfilm

data for seven companies

over nine years

and Pixar for some years his regression

employer years

67

2003 2011 However

includes

only 55

Case5 11cv 02509LHK

Document518

microprocessors that power them can cause a decline
Intel

to

impose

a

wage

Dr Leamers

125

compensation

independence

move

together

statistical

inferences

informing his estimates

estimates

must take

A generally

factors

this

Put differently

groups

estimated

the price per ounce

information

the dairy

and

Dr

errors

or present

Leamer

the underlying nature

a regression

to

explain

package

pints as

sizes at

a

in a single

within a
for in

estimates

are
that

from the regression

of his data

well and

is

a

the price of all package

spike

to

price distance

commonly

is

at

they provided

the cost of gasoline

The

this

or

but he did not even

tstatistics

used for

by

state

It

is

were
as

if

Dr

using data on

grocery stores in each

state

completely independent

recognize that a store that
rises

sells

high priced

at that store

gallons

say because

it

needed to deliver the milk to the

power

from a dairy store quality etc

68

year

to affect compensation

and gallons sold

sizes will increase

common

certain

were not meaningful

the price of gallons

if

in

chance

if

used to

not only failed to implement

the price of milk per ounce

store as

would have

by

errors and resulting

standard

to result from

unlikely

and there

the impact of gasoline

receives

the information

inference

that are affected

impacts of variables hypothesized

A proper analysis

likely sells high priced

store then

company

of pints quarts half gallons

and treating the various

from

of observations

a particular

to address

the estimated

significant

regression

on employee compensation
statistical

who

be accounted

that must

Dr Leamers

affect

employees

to take into account the fact that observations

in his report that his reported

testing whether

is far

method

contains

any other methodology

although

observations

clustering the standard

referred to as

Leamer had

would

calling

such as

into account

such as those affecting

statistically

cold

way

in a

of the

Dr Leamer a shock

of individual

much more limited and any

is

accepted

a regression

acknowledge

and

through

obtained

are related or correlated

based on over 500,000 individual

estimate

to Plaintiffs

members

of all

generally and the impact would not be limited to only the employee

Defendant and within a year

making

with his claims of a rigid

inconsistent

call Statistically this means that compensation

the cold

126

According

81

of

2009

is

in informationabout compensation

compensation

is

assumption

Page71

revenue and profitability and lead

in Intels

and with Plaintiffs claim that compensation

structure

proposed class would
an increase

such as occurred in

freeze

Filed1007 13

2

of the regression
is

not enhanced

to identify

by

Case5 11cv 02509LHK

including

because

individual

127

Dr Leamer

cases

such as this

standard
that

it

errors

recognizes that
168

was

When

standard

for

shows

that

factors unrelated

compensation

169

170

of

171

at

Angrist Joshua
University

Standard

counting

about whether his

your wealth

New

in

describe

to

assumption in
to cluster his

failure

Dr

small change
having

commonly

used

170

in

lots

Leamer admitted

of individuals

but

In Exhibits

21A

which are used

Dr Leamers

determine

to

estimates

no meaningful

estimates

show

coefficient

once properly

line

in the table is

22A and 22B

statistical

I

significance

22 and

in his Figures
estimates

4

24.171

This

for any employer

computed

or

standard

are completely consistent with there

conduct and his estimates

Thus

I

variable

errors In Exhibits

undercompensation

Leamers

21B

and

Figure 20 and 23 regressions except that the

undercompensation

Dr

studies of labor markets and widely

on employer year The conduct

p values

conduct

members

of

Leamer Depo

Chapter 9.3.3

his deposition

Dr Leamers

from

imply that

to that

Leamer Depo 374

Princeton

at

significant at conventional levels under the properly

regression provides

81

sizes at a particular store

improper to rely on the independence

such as this

being no true effect of the desired

168

is

errors is

clustered

none

p values

of

firm

at a

Leamers

is statistically

errors The

it

significant under these proper standard

Dr

Page72

169

show t statistics and

calculated
exhibit

details

now

errors are

not statistically
further

to

as necessary in analyses

and other

prices of four different package

use of that language

Clustering standard

estimates

year

asked

equivalent

only having one experiment

accepted

Filed1007 13

2

same underlying information

seems like appropriate

128

on

observations
the

all reflect

Document518

resulting entirely from

analyzed

Dr

Leamers

random

conduct

evidence that the challenged agreements reduced

of the proposed class

618

375 19 376 5

D

and Jrn

Steffen

Pischke

Mostly Harmless

Jersey

errors for the

annual

undercompensation

estimates

69

Chapter 8.2

Econometrics

pp 308 315 and Greene William
Pearson Prentice Hall 2008 p 188

Press 2009

H Econometric

are calculated

New

Jersey

Analysis 6th Edition

using

a bootstrap

method

Case5 11cv 02509LHK

4 Dr Leamer Does
Whether

129

Dr Leamer

several

it

he wrote

analyses

sensitivity

reporting

130

and

with

Consistent

Q Would

we

would have

to

is

see in the

I’ve

done

document

sturdy but

there’s

models One

As

to

the

at

or

a consequence

perform their

own

complete and more honest

172

as follows

someone who

the conduct regression

on

judgment

is

evaluating

its reliability

until

saw

that evaluator

analysis

would help

to

determine reliability

your regression analysis

know

record of econometrics
something

practiced

responded at his deposition

Dr Leamer that

sensitive

First

much more

is

purposes

researchers

think that the sensitivity

I

81

of

related to the regression

analyses

A would
Q And how
A
you need

it

of statistical

for individual

inferences

to reserve

as

art

for reporting

view Dr Leamer

this

econometric

perhaps thousands

ought to be demanding

you agree

sensitivity

the

that

are selected

efficient

of claimed

of the fragility

analysis

finds pleasing

much more

is

many

fitting

that the researcher

Page73

Robust or Fragile

many years ago

involves

Filed1007 13

2

not Report any Sensitivity Tests from which to Evaluate

his Results are

wrote

computer terminal

Document518

did not carry out a complete sensitivity analysis

I

how

that discusses

But

I

have estimated

and there

is

this

should be carried out and

more than one model more than one

some dimensions

some dimensions

I

of variability

in

which

in which

it’s

not sensitive

the changes

have a

this isn’t

that

and

you

it’s

can be

173

substantial

He

testified

further that

there are

conclusions with change
the coherence
the

model

of the

which damages can be

172

Edward

E Leamer

that are

this

Lets

Take

the

I

174

Leamer Depo

at

my suggested

at

of variability in which

model

Con

Out of Econometrics

358 818

70

the

and economic judgments about
that are implied

that demonstrates the

73 The American

by

method by

174

356 120

Leamer Depo

directions

the accuracy of the estimates

1983
173

some

made econometric

produced

one as

computed

variability

And

substantially

models

and selected

some

Economic Review 1

Case5 11cv 02509LHK

131

Dr Leamers

provided
models

Dr

Leamer

tried

177

to

fit

of any sensitivity
to the data

He

also said that

He

Thus

those other directions

is

it

analyses

175

a

acknowledged
176

one

but its

unclear

what

it

substantial
alternative

is

of

81

the

which has

178

he

but

models

models
sic

sensitive

what alternative

unclear

specification

examin ing

and

Page74

and computer programs he

Thus

out an extensive

Ive carried

not a complete

different directions

Filed1007 13

2

data documentation

with data by a defendant

with disaggregation

defendant

and the backup

report

no evidence

contain

Document518

to

do

defendant by

analysis in several

any

did not describe

Dr Leamer

to

tried

fit

of

to the

data

132

The

results

Defendants
reported

provided

I

clearly

show

above

allowing the impact of the conduct to

the fragility

of the single regression specification

common way

Another

conclusions that can be

drawn

period for which the regression

of testing the robustness

is

estimated

Dr Leamer

to changes

bases his conclusion

challenged agreements reduced compensation on a regression that compares
during the class period
and

2004 and

after

essentially

2010 and

2011 An

only the before period or only the
regression

and

test

2005 2009

after

to

the

alternative

combined periods
specification

period as the

that

across

Dr Leamer
and of the

of a regression specification

verify that the results are robust

is to

differ

that the

compensation

before effectively

to test robustness

control or benchmark

whether the challenged agreements affected

in the time

period

is

2003

to use

in the

compensation of members of the

proposed class

133

Exhibit

differences

23 shows

in the estimation

Leamers conduct

regression

undercompensation

175

176

177

178

179

Dr Leamers

that

period

179

implies

model

at

generally

at

at

at

22 and 24

365 1516

Leamer Depo

in his Figures

360 1618

Leamer Depo

366 79

Detailed regression

outputs

are

it

is

but very different

substantial

359 6

Leamer Depo

a test of whether

robust

Using only the pre period as the benchmark

percentages than reported

Leamer Depo

fails

provided in Appendices12A

71

to

12D

to

Dr

estimated

almost twice as large

Case5 11cv 02509LHK

Defendants but substantially

for several

Dr Leamers

benchmark
instead

conduct

overcompensation

Another sensitivity

using data outside
compensation

of

virtually

same magnitude

Dr Leamers

during the conduct period

as Exhibit

and

compensation

If

81

of

as the

no undercompensation

though opposite

model

is

to first estimate

Dr Leamers

higher

24 shows the

least

some

model

than actual

in

but

sign from

the

is

to

predict

robust one would expect the

compensation during the conduct

overcompensation

of the years

his conduct regression

estimates

compensation

predicted

therefore implying

years and five Defendants in at
sensitivity

Using only the post period

his conduct periods and then use the coefficient

However

period

implies

Page75

24

compensation to be generally

predicted

actual

regression

22 and

test

Filed1007 13

2

lower for Intel

of roughly the

effects he reports in Figures

134

Document518

180

levels are in fact lower than

two Defendants

at

Dr Leamers

model again

in all

fails

the

test

5 Dr Leamers

Regression

Model Does Not Explain Changes

in

Compensation

Over Time

135
from

The

analysis presented

Dr Leamers

correlated

regression
of his data

nature

firm
level compensation
relies

on comparison

predict

obtaining

showed

above

model

that are not accounted
level

a reliable prediction

critical

136

Exhibits

the impact

to estimating

25A

and

if

implies

we

that can be

for in his

of compensation

model

factors that drive

Given that his methodology

to the level

that his

model would

absent the challenged agreements

that the factors that

Leamer does

not account for are

by company and year between

compensation

earned by a firms employees and the average level of compensation

180

conduct

Detailed regression

plot the difference

Dr

important

Dr Leamers

regression

outputs

are

Figures

is

those agreements

quantitatively

I

drawn

account for the

that there are important

of compensation

any of

25B show

conclusions

are fundamentally different once

That correlation

of the actual

that the statistical

20 and

23

provided in Appendices

72

The

13A

and

exhibit

13B

shows

the average

predicted

that these prediction

by

Case5 11cv 02509LHK

errors are substantial

in

Document518

Dr Leamers

an

virtually all the factors that explain

Dr Leamer
Rather

the

is

terms economically

years and Defendants

distributed across

some

predicts very poorly in

years for

I

and predicted

actual

in

compensation

2003 and 2004 and

performed

a standard

firm
level compensation

Pixar in

statistical

test

be

are too large to

explained

for whether

solely

errors

A consequence

Leamers

181

Dr

estimated

Leamer used

evaluating
residuals

results

the

year level

at

model

companies

in

between

model

This

test

by company and year

the

regression

and asks whether those average residuals
test

factors

resoundingly rejects the

and

other

in

establishes the need to use

ways

statistically significant

Critically the

which

that contrary

implies

for important factors that determine

important

compensation

determinants of firm
level compensation

conduct effects will capture

the

term

economically

estimates

the impact

numerous

significant

one should focus on economic

value predicted

in his

just

not controlled

significance

Employee Class model Figure

ranges from 39 percent

provide accurate
residual

his

or correct for that correlation

of omitting

in his All Salaried

compensation
above

which means that

at

time

the Defendants over

138

Defendant

182

standard

claims Dr Leamer has

his

by

to

there are important factors explaining

by sampling error The

average residuals are economically not
to

one of which according

with the most extreme examples

Dr Leamers

no such omitted firm
specific

that there are

clustered

not explained

capturing

2004

that are omitted from

in compensation

hypothesis

model were

Large average differences

are evident

81

of

are not evenly

different Defendant

at

essentially examines the average residuals from his regression
variation

the

some companies

outcomes

different years have been left out of the regression

137

and

whether the Defendant had a challenged agreement with another

model

being Google

if

compensation

individuals

Page76

significant181

would be expected

as

important factors that are unique to compensation

employees

Filed1007 13

2

by

his

of the

below

regression

the

A model that

model is 16.5 percent

roughly plus or minus

indicating

33

percent

economically

by Dr Leamers

predicts

variable

that his

times in his deposition

20 are

overall

The

root

regression

compensation

mean square

The average

Actual real total

significant

to

49 percent

so poorly cannot
error

of the average
at

using a 95 percent confidence level

The test Ftest results are F39 5047711319.6 for Dr Leamers
Figure 20 regression and
F39,292367 832.09 for his Figure 23 regression Pvalues for both tests are virtually zero

73

Dr

and stated that in

model measures average compensation

182

that

other than the

and not just statistical significance

value predicted

impact of his conduct

of variables

is

the

company

Case5 11cv 02509LHK

challenged agreements that

To

the potential

illustrate

Document518

between

systematically

differ

problem

I

Filed1007 13

2

Page77

non conduct

the conduct and

considered what would happen

if I

81

of

periods

simply add a variable

measuring the performance of the stock market from his regression which potentially would
measure

general

acknowledges

shows

economic and

financial performance

from adding the change

the results

conduct regression

184

results the addition

P 500 index

Leamers

much

Dr Leamer

that

overcompensation

183

26

Exhibit

as an explanatory variable

Figure 22 and 24

smaller

2005 2009 Thus

Class throughout

economy

S

in the

of this variable yields

AllSalaried Employee Class and
the Technical

Dr

In contrast to

the

see his Figure 8 and related discussion

compensation

likely affect

in

in his

undercompensation

undercompensation
for all Defendants

estimates

except

for the

Google

for

of economically significant

the existence

classwide
by

factors not captured

be unreliable

measures

his

of

Dr

model causes

damages and

Figure 20 and 23 regression estimates

Leamers

unreliable

as

a method

to

of demonstrating common

impact

6 Dr Leamers

Conduct

Variable Cannot Capture the Impact of the Challenged

Agreements

Dr Leamers

139

conduct

variable

challenged agreements between

reflects

refers to these throughout

Plaintiffs

do

not claim

in their

non compete agreements

his report as

Complaint

or in their

Motion

was

183

184

the

independent
185

14A

Appendices

and 14B show

expected

See Declaration

allege

with

Intuit

through a cold call

Defendant

Evidence

I

that these

185

positive

detailed

outputs

regression

The

sign and is statistically significant

of Jeff Vijungco
I

am

Apple See

p6

on

unaware

of

coefficient

under

Adobe and Apple continued

any requisitions that went

also Declaration

of Chris

has agreed to not cold call employees

have never been instructed

presented

Dr

estimate

Leamers

basis

above

on the change in

SP

assumption

to refrain

at

Galy

Google

from making cold

pp

45

to recruit

unfilled
I

no cold

as

any other company and

call

understand that plaintiffs in this case

To my knowledge

calls to

and hire from each other

because of the

Google employees

no such agreement

To the contrary I have made cold calls to Google employees
am aware that other recruiters at Intuit have also done

so

74

exists

I

and have never given any

such instruction to anyone else at Intuit

same

as long as

observations

during the Class Period
agreement

understand that

98

Leamer Report

500 shows

not identified or recruited

I

for Class Certification

agreements prevented a Defendant from hiring applicants from another
that applicant

Dr

employees for a period of time Although

Defendants to avoid cold calling each others

Leamer

pairs of

on the

Case5 11cv 02509LHK

Dr Leamers

and that underlies

Document518

Filed1007 13

2

demonstrates that during the period of the challenged agreements
of other Defendants even

amount

when

cold calling were an important

if

and

hiring from other Defendants

by

the challenged

variable

conduct that he

is

call

way

Defendants hired employees

during the conduct period

eliminated

would have

as

of recruiting from other Defendants

calling activity were

can identify and measure

leading

or substantially

some

restrictions

extended to other firms as well

186

of the challenged

to

reduced

agreements only

the impact of those agreements properly

of which

were

typically

the impact

Dr Leamer

cites

not limited to the other Defendant

the

the

reasons

1 shows

his Figure

support

to

if

represents

His conduct variable cannot do so for several

trying to evaluate

First evidence I reviewed

cold

cold

his regression that represents

in

7

in his Figure

agreements

Dr Leamer

140

81

of

they had agreed not to cold call those employees and that the

of hiring from other Defendants did not decline

occurred

movers

earned by

analysis of compensation

Page78

identified in Figure

that

1 but

In part this reflects the fact that the motivation for these

ofinterest
agreements generally does not appear to be holding

undercompensating
potential

senior executives

calling

on

or perceived

of another

employees

the willingness

agreements

commercial

a

but not during the

include

the impact

non Defendant

corresponding

See

e

non conduct

If

DNCC

unchallenged

same

period

biasing

his estimated

effect

he can

separate

their

Dr Leamer

upwards

If

as

he can measure

impacts from the

policies

Yahoo EBay

PayPal

and Bizrate See

176 177 178 179 182 183
Bentley

of cold

agreements or

by

firms lead to undercompensation

if

of

as the challenged

periods then the effect estimated

agreements only

impact of unchallenged

g 231APPLE041662

conflicts

or concerns about the impact

For example Google had restrictions on hiring from Genentech

Geshuri Exhibits

or

from concerns about

existed during the

of those other policies

the impact of the challenged

187

arose

arrangements

of partners to collaborate

involving

compensation of

from membership on one companys Board of Directors

Leamer claims DNCC agreements between

186

but instead

187

unilateral policies

would

a firms

down

Depo 61 21622 251618

75

Lambert

Depo 21 78

Dr

Case5 11cv 02509LHK

141

The Consent Decrees

made

clear that restrictions

circumstances
such as

Document518

that the Defendants

on

recruiting other

when they

including

are

development technology

joint

signed with the

the function

integration

in Section

not to consider

IV

from employees

applications

ventures

projects

hire employees of another

person provided

any other person adopt enforce

142

of

the Plaintiffs theory

legal alternative

to

the challenged

how

not

permissible

an individual

where

restrictions

understand what collaborations

188

190

in

Final Judgment

Final Judgment

in

in

still

there

of

190

same way
agreements

and

for

Thus

affected

the

affected

Dr Leamers
by

would only be

the

by

affected

legal restrictions

which employees were involved

legal restrictions

the price

in

analysis to

calling

etc

a unilateral policy
Directors

is

to

and

Proposed

and

Proposed

avoid cold calling another

CEO

of

a competitor
of

non

discovery process would have been affected
and the impact and loss from the challenged

amount above

76

that

and

which employees were involved

on cold

Board of

and

would have been

employees

during the Class period to not cold call employees

challenged agreements

the incremental

the

and

world for

v Adobe Systems Inc et al 317 2011 p 7
v Lucasfilm Ltd 592011 p 6

theory

pressuring

opportunities

but for

This requires individualized

example when a member of

that

hiring prohibitions

United States of America

employees

was

from

v Adobe Systems Inc et al 317 2011 p 6
v Lucasfilm Ltd 592011 p 5

United States of America

then under
it

requesting

United States of America

were unilateral policies at Defendants

Defendants

would have

existed during the class period

United States of America
in

from

or

recruit

any less restrictive

limiting cold calling

Similarly nothing prevents an employer from implementing

Defendants

determined

recruiting the other Defendants

would have been

that there

Final Judgment

Final Judgment
189

on

as on balance procompetitive

the likelihood

is

some employees

determination

call

cold

policy

policy

of employees involved in those collaborations

requiring

collaborations

eg

collaborations

purposes of measuring impact and loss has
others

solicit

to

a

189

compensation

agreements

only to employees involved in actual
compensation

person or

further that

state

to adopt

or maintain such a policy and are prohibited
or maintain such a

including

The Consent Decrees

that Defendants are prohibited

any other person to adopt enforce

Under

joint

from unilaterally deciding

of another

of Justice

agreement

collaboration

188

81

of

are legal under certain

of a legitimate

joint

shall prohibit a Defendant

Page79

US Department

companies employees

teaming agreements and the shared use of facilities

n othing

Filed1007 13

2

caused by legal agreements

in the

If

Case5 11cv 02509LHK

143

Dr Leamers

Second

cold calling but

treats

conduct

variable

any agreement between

agreements between

multiple

Document518

a Defendant

Filed1007 13

2

cannot measure

Apple

pairs of Defendants

participated

Leamers

in

according

challenged agreements with several

to

as having

the

better and

in

Plaintiffs

and

Dr Leamer

Apple

Dr
that is

Rather such models would show

more

as

in a single challenged

other Defendants simultaneously

of price discovery

they apply that more information results

on

same impact

price discovery framework does not imply that the amount of information

restricted is irrelevant to the process

81

For example Adobe and Apple

and other Defendants

while

of

the intensity of restrictions

were assigned the same conduct values every year but Adobe engaged
agreement with one Defendant

Page80

where

rapid price discovery than less

agreements should have a larger impact than a single

information and thus that multiple

agreement

7

Estimated

Persistence Effects

Discovery

Model and

his

are Inconsistent with

Claim

Defendants

that

Dr Leamers

had

Price

Rigid Compensation

Structures

144

Dr Leamer
The

describes

percent

in

some year
on

for

finding

is

that effect lingers

frameworks

this

meaning when

very persistent effects

indicates

However

on which he

relies

that

many

makes him or

all

years

inconsistent

or virtually

all

levels gained

Defendant affects
only a weak

191

basis for

why

all

gets

a bump up

in

coworkers

If this

past compensation

144

77

agreements and

a significant and

wide spread

claims that economic theory combined

employees

employees compensation

the challenged

would cause

members Dr Leamer

by one Defendants

persistence between

Leamer Report

year and the

with the price discovery and internal equity

as the theoretical

class

a worker

her better off than comparable

with Defendants rigid compensation structures shows that
compensation

in the previous

191

resulting reduced flow of information to employees

impact on

as follows

that two for each employer The fact that these numbers sum to around 90

compensation

145

estimates

variables are the levels of total compensation

persistence

year before

his persistence

new

information on appropriate

from cold

calls

were true however

levels and current

from another
then there should

ones

a

mover who

be
is

Case5 11cv 02509LHK

induced to

move

a 30

receives

to Defendant

percent

Dr Leamers

Defendant

irrespective

is

determined

that any increase

in

persistent increase

B by

a

cold

his

model

implies

restored more than 95 percent
remain for

Adobe Apple

information in

of

damages imply

reconcile

with the

employment

per

that any effect

and
a highly

is

slow

as well as with his

own

it

is

of

members

of the

flow has been fully

the flow of information is

after

has the lowest estimated
not just

of the reduced

employees have obtained

after

on compensation would

of reduced information

rates of hiring at these firms

year

the firm generates

in

the information

after

that even five years

Google

model

Yet

compensation

and Pixar employees while more than 60 percent

Dr Leamers

As

individuals

no reason why aggregate compensation

of the impact

remain for Google employees

A and

compensation

proposed class would remain depressed so long

resumed Indeed

twoyears

compensation relative to others

his estimates

provides

81

in all Defendant

by demonstrating that an

informationflow persists strongly for an extended time even

more information Yet he

of

employees previous compensation

of those other

in that individuals

model and

Page81

call from Defendant

any year largely by his previous

in

an individuals

Dr Leamers

Filed1007 13

2

should cause a change

regression rules out such an impact

compensation

146

A from

in compensation

increase

employees compensation

Document518

persistence

glacial

averaging

of the effect

level The flow

Such slow adjustment
over

would

percent

is

of

hard

to

of average

theory

price discovery

8 Summary
147

Dr Leamers

undercompensation
are

made

regression

model

of undercompensation

and

percentages by company and year are invalid

to permit a test of his theory

there is

no evidence

of

his derived

When

common

estimates

of annual

necessary corrections

impact from the

challenged conduct no evidence of average impact across members of the proposed class and no
basis for his estimates

of undercompensation

Kevin

M Murphy

November

78

12

2012

Case5 11 cv 02509 LHK

Filed10 07 13

3

Exhibit
Hires

Document518

1A

and Separations at Defendant Companies
Hires

Page1

of

46

From Other Defendants vs
To

Separations

Hires

Overall

Separations

Year

2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011

2001 2004 Avg
2005 2009 Avg
2010 2011 Avg
2001 2011 Avg

2001 2004

Total

2005 2009

Total

2010 2011

Total

2001 2011

Total

Notes

This analysis

excludes hires indicated as acquisitions hires showing the same

the hiring and separations

employment
Source

Dr

in

that appear as immediately

each year

Leamer’s employee

data

rehired by the same

defendant

defendant

company

company

as their immediate previous

within one year

Number

employer within one year of

of employees

is

calculated

as average

Case5 11 cv 02509 LHK

3

Exhibit
Hires

Document518

1B

and Separations at Defendant Companies
Hires

Filed10 07 13

From
To

Page2

of

46

Other DNCC Defendants vs Overall

Separations

Hires

Separations

Year

2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011

20012004 Avg
20052009 Avg
20102011 Avg
20012011 Avg
2001 2004

Total

2005 2009

Total

2010 2011

Total

2001 2011

Total

Notes

This analysis

excludes

the hiring and separations

employment
Source

in

hires

indicated as acquisitions

that appear as immediately

each year

Dr Leamer’s

employee

data

hires

showing the same defendant

rehired by the

same defendant company

company

as their immediate previous

within one year

Number

employer within one year of

of employees

is

calculated

as average

Case5 11 cv 02509 LHK

Document518

Filed10 07 13

3

Page3

of

46

Exhibit 2A

Number of Employees by Defendant and
All

Adobe

Apple

Salaried Employee Class

Google

2001

2,503

5,096

2002

2,226

5,255

2003

2,291

5,424

1,329

2004

2,508

5,684

2005

3,791

6,474

2006

3,663

6,993

2007

3,951

2008

4,203

2009

Year

Intel

Intuit

Lucasfilm

Pixar

All

Defendants

210

3,169

66,242

542

3,982

63,569

4,311

62,439

2,346

4,247

64,172

4,117

4,418

73,556

6,873

4,498

74,045

7,951

8,768

5,069

73,247

9,135

10,983

5,081

75,205

4,928

10,005

11,175

4,683

75,166

2010

5,010

11,655

13,988

4,605

80,193

2011

5,385

13,226

18,179

4,770

90,070

Source Dr Leamer’s

backup

data

and materials

Case5 11 cv 02509 LHK

Document518

Filed10 07 13

3

Page4

of

46

Exhibit 2B

Number of Employees by Defendant and
Technical Creative and
Adobe

Apple

Google

Intel

RD

Intuit

Year

Class

Lucasfilm

Pixar

All

Defendants

2001

1,582

2,670

101

1,557

34,484

2002

1,441

2,866

207

1,977

33,881

2003

1,450

2,954

509

1,907

33,517

2004

1,579

2,942

1,026

1,829

33,592

2005

2,205

3,358

2,258

1,814

40,479

2006

2,218

3,677

3,776

1,863

41,216

2007

2,277

4,248

5,290

2,244

42,550

2008

2,400

4,950

6,388

2,349

44,243

2009

2,552

5,589

6,825

2,237

45,453

2010

2,489

6,663

8,693

2,308

48,994

2011

2,639

7,582

11,139

2,457

55,338

Source Dr Leamer’s

backup

data

and materials

Case5 11 cv 02509 LHK

Document518

Exhibit

3

Filed10 07 13

Page5

of

46

3

Top 20 Previous Employersof Hires by Defendant Companies

Adobe

Note

Hires through

transfers

acquisitions

are excluded

This analysis uses

Adobe’s

compensation data and may not include

all internal

Case5 11 cv 02509 LHK

Document518

Exhibit

3

Filed10 07 13

Page6

of

46

3

Top 20 Previous Employersof Hires by Defendant Companies
Apple

Note

Analysis restricted

to

hires for

job codes provided

in

the compensation data

Case5 11 cv 02509 LHK

Document518

Exhibit

3

Filed10 07 13

Page7

of

46

3

Top 20 Previous Employersof Hires by Defendant Companies
Google

Case5 11 cv 02509 LHK

Document518

Exhibit

3

Filed10 07 13

Page8

of

46

3

Top 20 Previous Employersof Hires by Defendant Companies

Intel

Case5 11 cv 02509 LHK

Document518

Exhibit

3

Filed10 07 13

Page9

of

46

3

Top 20 Previous Employersof Hires by Defendant Companies

Intuit

Case5 11cv 02509LHK

Filed1007 13

Document5183

Exhibit

Page10

of

46

3

Top 20 Previous Employersof Hires by Defendant Companies

Lucasfilm
Number
Rank

Previous

of Hires

Percentage

of

Total Hires

2008Q2 2012Q1
26

Employer

LUCASFILM

2008Q2 2012Q1

7.1

1

ELECTRONIC ARTS

20

5.5

2

IMAGEMOVERS

8

2.2

3

WALT

6

1.6

4

ACTIVISION

5

1.4

5

ORPHANAGE

5

1.4

6

2K

4

1.1

7

CBS

4

8

DIGITAL

1.1
1.1

9

DIGITAL

DISNEY

INC

GAMES
DOMAIN

4

PDI

4

10

SONY

4

1.1
1.1

11

APPLE

3

0.8

12

DOUBLE

3

0.8

13

DREAMWORKS

3

0.8

14

MICROSOFT

3

0.8

15

PIXAR

3

0.8

16

ZYNGA

3

17

CRYSTAL DYNAMICS

2

0.8
0.5

18

MUNKYFUN

2

0.5

19

ADOBE

1

0.3

20

EBAY

1

0.3

3

0.8

61

16.7

187

51.2

0

0.0

7

1.9

365

100

Self

FINE

PRODUCTIONS

INC

EmployedUnemployed

Unknown
Other

NonDefendants

Other Defendants
All

Lucasfilm

Defendants excluding

Total

Lucasfilm

Case5 11cv 02509LHK

Document5183

Exhibit

Filed1007 13

Page11

of

46

3

Top 20 Previous Employersof Hires by Defendant Companies

Pixar
Number

of Hires

Percentage

of

Total Hires

2001 2012Q2

2001 2012Q2

PIXAR

5

0.6

1

LUCASFILM

22

2.5

2

BLUE SKY STUDIO

18

2.1

3

WALT

16

1.8

4

PDI

10

1.1

5

TIPPETT

10

1.1

6

APPLE

8

0.9

7

DREAMWORKS
RHYTHM
HUES

6

8

0.7
0.7

Rank

Previous

Employer

DISNEY

6

9

UC BERKELEY

5

10

WDFA

5

0.6
0.6

11

ELECTRONIC ARTS

4

0.5

12

ESC ENTERTAINMENT

4

0.5

13

MICROSOFT

4

0.5

14

SONY

4

0.5

15

BRIGHAM

3

0.3

16

FRAMESTORE

3

17

GOOGLE

3

0.3
0.3

18

TAMU

3

0.3

19

WARNER BRO

3

0.3

ACTIVISION

2

0.2

7

0.8

420

48.2

294

33.7

7

0.8

40

4.6

872

100

20

Self

YOUNG UNIV

EmployedUnemployed

Unknown
Other

NonDefendants

Other Defendants
All

Defendants excluding

Pixar

Pixar Total

Note The

lengths

of

the periods

analyzed vary by

Sources Recruiting data from Apple Google

company based on

Intel Intuit Lucasfilm

data availability

and Pixar

Compensation data from Adobe and Apple

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page12

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page13

of

46

Case5 11cv 02509LHK

Filed1007 13

Document5183

Page14

of

46

Exhibit 5

Employment

of Software Engineers

of Industries of Defendant Companies
Industries of

Defendant

Adobe

Year

Apple

Google

Intel

Intuit

LucasFilm

Pixar

Defendant

Defendant

Companies

Companies

Companies

79,910

2002

1,165

1,263

8,065

2003

1,167

1,228

7,811

101,470

2004

1,258

1,207

8,317

105,160

2005

1,694

1,336

10,656

106,890

2006

1,728

1,333

11,742

2007

1,880

1,411

13,907

108,650

2008

1,958

1,425

15,404

122,130

2009

1,984

1,282

16,301

127,860

2010

1,865

1,361

18,728

124,910

2011

1,939

1,475

22,318

134,150

96,440

Adobe

Apple

Google

Intel

Intuit

LucasFilm

Pixar

1.5
1.2

1.6
1.2

1.2
1.6

1.1
1.2

1.8
1.7
1.6
1.6

1.4
1.3
1.2
1.0

12.2

1.5
1.4

1.1
1.1

15.0

10.1

7.7
7.9
10.0
12.8
12.6
12.7
16.6
2002 2004 Average

8.6

2005 2009 Average

12.1

2010 2011 Average

15.8

of All Industries

Defendant
Industries

Adobe

584,020

0.2
0.2

0.2
0.2

1.4
1.2

0.2
0.2
0.2
0.2

0.2
0.2
0.2
0.2

1.2
1.4
1.5
1.7

868,210

All

0.2
0.2
0.2

0.2
0.2
0.2

1.8
1.9
2.2

921,500

0.2

0.2

2.4

651,740
717,420
758,050

Source Defendant

employment

numbers

are based on

data as well as classification of software engineers

Employment
Industry

of industries of

Specific

Defendant

Data for the following

Dr Leamer’s employee

performed by

companies

my

834,850
staff

based on BLS OES National

NAICS codes

764,430

based on CapIQ company

information

851,850
852,670

Apple

Google

Intel

Intuit

LucasFilm

Pixar

334100

Computer

519100

Other Information

334400

Semiconductor

511200

Software Publishers

2005 2009 Average

512100

Motion Picture

2010 2011 Average

and Peripheral

Equipment

Manufacturing

Companies

Services

and Other Electronic

and Video Industries

Component Manufacturing

2002 2004 Average

1.2
1.7
2.3

Case5 11cv 02509LHK

Document518

Filed1007 13

3

Page15

of

46

Exhibit 6

Age

Distribution of

New

Hires

2001 through 2011
Adobe

Apple

Google
All

31 to 35
36 to 40
41 and over

7

6

17
24
22
30

24
30
22
17

36 to 40
41 and over

Source Dr Leamer’s

RD

8

6

7

17
26
22
29

27
33
21
12

backup

data

and materials

19
30
24
14
13

Class

20
24
21
27

25 and under

31 to 35

Creative and

Pixar

Class

7

Technical

26 to 30

Employee

Lucasfilm

19
24
22
28

25 and under
26 to 30

Salaried

Intuit

Intel

18
32
24
15
10

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page16

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page17

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page18

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page19

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page20

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page21

of

46

Case5 11 cv 02509 LHK

Document518

3

Filed10 07 13

Page22

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page23

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page24

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page25

of

46

Exhibit 12

RSquareds

in

Dr Leamer’s

Compensation Structure Regressions Are Mostly

Attributable to

AllSalaried

R Squareds

in

Dr Leamer’s
Figure

2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011

Source Dr Leamer’s

11

95
94
94
93
93
92
91
92
92
90
92
Figure 11

Employee

Employer and Job Indicators
Technical

Class

Including Only

Excluding

Employer and Job

Employer and Job

Indicators

Indicators

94
93
93
93
92
92
91
91
92
90
91
and 13 regressions

21
21
22
19
20
21
21
20
20
22
24

RSquareds
Dr

in

Leamer’s

Figure 13

89
89
88
88
88
87
85
86
88
84
88

Creative

and

RD

Class

Including Only

Excluding

Employer and Job

Employer and Job

Indicators

Indicators

89
88
88
88
87
87
85
86
88
84
87

15
16
16
18
16
19
17
19
17
18
21

Case5 11cv 02509LHK

Document518

Filed1007 13

3

Page26

46

of

Exhibit 13A

Named

PlaintiffsActual Total Compensation

by Dr Leamer’s

Comp

Actual Total

Named

Plaintiff

Employer

Year

Predicted

by Dr

Comp

Leamer’s

Model

1
Brandon Marshall

Predictions

Model

Figure 12

Total

vs

2

Difference

3

ADOBE

2006

73,895

61,035

12

Difference

31

12,860

17.4

Michael

Devine

ADOBE

2006

131,222

124,424

6,798

Michael

Devine

ADOBE

2007

146,540

135,001

11,539

5.2
7.9

Mark Fichtner

INTEL

2001

151,712

133,620

18,091

11.9

Mark Fichtner

INTEL

2002

124,426

120,980

3,446

Mark Fichtner

INTEL

2003

109,352

109,349

Mark Fichtner

INTEL

2004

123,374

120,221

3,153

Mark Fichtner

INTEL

2005

133,431

135,403

1,972

Mark Fichtner

INTEL

2008

122,013

133,469

11,456

Mark Fichtner

INTEL

2009

138,501

139,125

Mark Fichtner

INTEL

2010

152,238

141,816

Daniel Stover

INTUIT

2006

79,129

Daniel Stover

INTUIT

2007

Daniel Stover

INTUIT

Daniel Stover

INTUIT

Siddharth

Hariharan

LUCASFILM

Dr Leamer’s

Figure 12 regressions

Source

10,422

2.8
0.0
2.6
1.5
9.4
0.5
6.8

91,136

12,007

15.2

103,265

105,061

1,796

1.7

2008

175,177

108,817

66,361

37.9

2009

132,553

121,416

11,137

8.4

2007

102,000

90,819

11,182

11.0

3

624

Case5 11 cv 02509 LHK

Filed10 07 13

Document5183

Page27

of

46

Exhibit 13B

Named

Plaintiffs

by

Actual Total Compensation

Dr Leamer’s

Actual

Named

Plaintiff

Employer

Year

Total

Comp

1
Brandon Marshall

Predictions

Model

Figure 14

Total

vs

Comp

Predicted

by Dr

Leamer’s

Model

2

Difference

3

ADOBE

2006

73,895

60,754

12

Difference

31

13,141

17.8

Michael

Devine

ADOBE

2006

131,222

124,661

6,561

Michael

Devine

ADOBE

2007

146,540

134,724

11,816

5.0
8.1

Mark Fichtner

INTEL

2001

151,712

135,177

16,534

10.9

Mark Fichtner

INTEL

2002

124,426

121,965

2,461

2.0

Mark Fichtner

INTEL

2003

109,352

109,866

Mark Fichtner

INTEL

2004

123,374

119,152

Mark Fichtner

INTEL

2005

133,431

134,261

Mark Fichtner

INTEL

2008

122,013

132,988

Mark Fichtner

INTEL

2009

138,501

139,074

Mark Fichtner

INTEL

2010

152,238

141,186

Daniel Stover

INTUIT

2007

103,265

105,025

1,760

1.7

Daniel Stover

INTUIT

2008

175,177

108,866

66,311

37.9

Daniel Stover

INTUIT

2009

132,553

122,644

9,909

7.5

Siddharth

LUCASFILM

2007

102,000

89,439

12,561

12.3

Source

Hariharan

Dr

Leamer’s Figure 14 regressions

514
4,222

830
10,974

573
11,052

0.5

3.4
0.6
9.0
0.4

7.3

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page28

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page29

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page30

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page31

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page32

of

46

Exhibit 16

Dr Leamer’s Model

Implies

Very Large Differences Over Time

Individuals with Identical Characteristics and Starting
Simulations Based

Adobe

on Dr

Leamer’s

15
32

90th Percentile

31
67

Compensation Levels

Intel

Difference in Compensation after
Average

46

Two

Intuit

All Firms

Years

11
22

100

the Compensation of

Conduct Regression

Google

Apple

in

16
33

24
56

22
46

37
86

Difference in Compensation after Five Years

29
61

Average
90th Percentile

53

62

111

16
34

135

Notes

1
2
3
4

Compensation
Percent
Based

using

on 50,000 simulations of compensation

Lucasfilm

Source

differences are constructed

differences are defined as differences

Dr

and Pixar are excluded because

Leamer’s

backup

data

in

and

residuals

from Dr Leamer’s

Figure 20 regression

logs

growth

there

and materials

coefficients

from 2004 through

is insufficient

data to

2009

for

each firm

do simulations

in all

years

model

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page33

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page34

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page35

of

46

Case5 11 cv 02509 LHK

3

Document518

Exhibit

Filed10 07 13

Page36

of

46

19

Average Percent Change in Total Compensation

Dr Leamer’s Figure

19 Disaggregated by Company

vs

Dr Leamer’s
Figure

Average Change
Year

Adobe

2002

Compensation

27.8

2003

in Total

Apple

Google

Lucasfilm

Pixar

Pooled

Intel

Intuit

27.2

0.6
1.5

2.1
5.1

8.5

4.7
2.3

13.1

8.3

10.3

9.8
6.9

1.3

2006

10.6

5.6
13.9

0.5
9.1

2007

11.2

4.5

8.8

2008

6.9

12.0

8.8

2009

7.5

2.9

0.1

7.4
6.8
7.4

2010

3.0
11.1

7.9
8.7

12.7

2011

2004
2005

EstimatedOverpayment Underpayment
Year

Adobe

Apple

Google

2005

3.4
0.6

4.2
8.8

4.9
0.0

14.5

16.4

2008

0.0

2009

0.0

0.0

2006
2007

6.5
9.7

1.8

Initial

Lucasfilm

Pixar

Pooled

Intel

Intuit

8.7

12.2

0.6

0.4

8.9

2.8
8.5

35.6

17.2

26.8

9.5
0.9

0.0

6.4
0.0

3.8
0.0

3.8
0.0

9.0
0.0

2.6
0.0

0.0

0.0

0.0

0.0

0.0

0.0

Pixar

Pooled

EstimatedOverpayment Underpayment

Cumulative

Year

Adobe

Apple

Google

Intel

Intuit

2005

3.4

4.2

8.7

12.2

0.6

2.8

35.6

9.5

2006

13.0

25.9

12.5

9.5

11.4

62.3

10.3

27.5

13.3

15.1

71.4

12.9

27.5

9.5
9.5

18.9

2008

4.0
8.9
8.9

18.9

13.3

15.1

71.4

12.9

2009

8.9

27.5

9.5

18.9

13.3

15.1

71.4

12.9

2007

Note

This analysis

follows

Defendants even though
Source

Dr

for

Lucasfilm

Leamer’s methodology in his Figure 19 of treating 2005 as the first year of the agreements for

Intuit Lucasfilm and Pixar the first alleged agreements started in other years

Leamer Report backup data and programs

all

19

Case5 11cv 02509LHK

Document518

Exhibit
Undercompensation

Defendant

Estimates Using

Conduct Variables and Other Defendant
Effects

in

Dr

Specific

Adobe

Apple

Google

2005

1.82

2.54

12.73

2006

4.37

0.72

26.90

0.68

2.65

19.16

6.26

6.45

2008

2.19

4.06

5.70

8.01

10.24

2009

20.26

1.53

5.43

8.96

10.02

RD

Apple

Google

2005

1.92

2.01

11.08

1.71

2006

5.82

2.95

22.47

0.62

2007

0.05

5.23

13.12

3.03

6.93

2008

1.29

7.33

0.88

3.44

8.59

2009

22.60

6.28

10.56

4.67

7.47

Leamer Figure 20 and 23 regressions including
indicators

Pixar revenue

and

Dr Leamer’s

vs

conduct age and

data after 2005 are included

Figures

22 and 24

Lucasfilm

Pixar

Year

Adobe

Apple

Google

25.47

2005

1.61

1.59

1.78

1.67

30.64

2006

4.28

4.43

4.44

4.70

14.63

12.44

13.95

28.52

2007

6.64

6.94

6.39

7.46

3.24

17.24

14.28

14.15

36.96

2008

9.08

9.56

8.40

10.05

5.64

19.94

15.76

13.79

31.11

2009

9.15

9.73

7.51

9.95

5.70

20.12

14.65

Intel

Technical Creative and

Class

Adobe

company

Undercompensation Estimates in Dr Leamer’s

Interactive

9.59

Year

Intel

46

1.70

Intuit

0.51

Technical Creative and

of

AllSalaried Employee Class

1.89

2007

Intel

Page37

20

Employee Class

Year

Source

Filed1007 13

Specific

Leamer’s Regression

AllSalaried

3

Intuit

Lucasfilm

12.13

RD

Pixar

10.56

Class

hiring

Pixar

Year

Adobe

Apple

Google

28.18

2005

1.56

1.90

3.07

1.64

17.23

interactions

Lucasfilm

6.60

Intuit

30.70

2006

4.29

4.96

7.23

3.06

14.77

10.47

23.38

36.34

2007

6.48

7.79

9.36

3.38

3.41

18.08

10.61

24.38

34.92

2008

8.80

10.64

11.20

4.76

5.21

20.44

11.87

24.05

28.33

2009

8.44

10.51

9.00

4.19

4.96

20.54

9.62

between

rate variables

Intel

Intuit

Lucasfilm

10.80

Pixar

9.28

Case5 11 cv 02509 LHK

Document518

Exhibit

Dr

Page38

of

46

21A

Leamer’s Figure 20 RegressionUsing Corrected Standard Errors

Employee Class

AllSalaried

Dependant

Filed10 07 13

3

Variable

Log Total Annual Compensation CPI

Variable

Estimate

St

Error

T Value

Conduct

Age

0.0067

0.0031

2.18

Conduct

Age 2

0.0001

0.0000

2.45

Conduct

Log Number

0.0028

0.0247

0.12

0.1647

0.1269

1.30

0.6949

0.0608

11.42

0.7404

0.0587

12.62

0.4945

0.0530

9.33

0.6690

0.0351

19.06

0.7090

0.0458

15.48

0.6944

0.1840

3.77

0.8131

0.1069

7.61

0.2963

0.0461

6.43

0.2610

0.0407

6.41

0.3732

0.0453

8.25

0.3001

0.0389

7.71

0.2551

0.0433

5.89

0.1983

0.0780

2.54

0.1779

0.0979

1.82

Log Age Years

0.3591

0.1799

2.00

Log Age 2

0.0394

0.0233

1.69

Log Company Tenure Months

0.0107

0.0415

0.26

Log Company Tenure

0.0012

0.0043

0.28

0.0027

0.0020

1.37

San Jose

1.4353

0.3827

3.75

Among Defendants

0.0961

0.0456

2.11

0.0038

0.0076

0.50

0.0154

0.0214

0.72

0.2485

0.0568

4.37

0.1070

0.0785

1.36

of

New

Hires In the

Firm
Number

of

Employees

1

Conduct

1
APPLE
Log Total Annual Compensation CPI 1
GOOGLE
Log Total Annual Compensation CPI 1
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

INTEL

Log Total Annual Compensation CPI

INTUIT

Log Total Annual Compensation CPI

PIXAR

1

1
1

Log Total Annual Compensation CPI

LUCASFILM

1

2
APPLE
Log Total Annual Compensation CPI 2
GOOGLE
Log Total Annual Compensation CPI 2
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

INTEL

Log Total Annual Compensation CPI

INTUIT

Log Total Annual Compensation CPI

PIXAR

2

2
2

Log Total Annual Compensation CPI

LUCASFILM

2

2

Male

DLog

Information Sector Employment

Log Total Number
Year

of

Transfers

in

trend

Log Number

of

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLog Firm Revenue

Firm
Number

of

Employees

1

Hires

1
EmployeeCPI 1

EmployeeCPI

0.2170

0.0814

2.67

APPLE

0.0627

0.2642

0.24

GOOGLE

1.0364

0.3351

3.09

INTEL

0.1522

0.2431

0.63

INTUIT

0.1462

0.2151

0.68

PIXAR

0.7251

0.6673

1.09

LUCASFILM

0.1352

0.2762

0.49

State

Location

Per

YES

Indicators

Constant

YES

RSquare

0.926

Observations

Note
Source

Significant

Dr Leamer’s

504,897

at

1

level

Significant

at

backup data and materials

5

level

Standard

Significant

at

10 level

errors clustered on employer year

Case5 11 cv 02509 LHK

Document518

Exhibit

Dr

Filed10 07 13

3

of

46

21B

Leamer’s Figure 23 RegressionUsing Corrected Standard Errors
Technical Creative and

Dependant

Page39

Variable

RD

Class

Log Total Annual Compensation CPI

Variable

Estimate

St

Error

T Value

Conduct

Age

0.0079

0.0033

2.38

Conduct

Age 2

0.0001

0.0000

2.71

Conduct

Log Number

0.0121

0.0281

0.43

0.2196

0.1362

1.61

0.6744

0.0650

10.38

0.7234

0.0570

12.70

0.4367

0.0672

6.50

0.6401

0.0325

19.67

0.6703

0.0486

13.81

0.6491

0.2295

2.83

0.8462

0.0911

9.29

0.3053

0.0523

5.83

0.2538

0.0391

6.49

0.3659

0.0476

7.68

0.3179

0.0353

9.00

0.2857

0.0439

6.51

0.1045

0.0896

1.17

0.1448

0.0805

1.80

Log Age Years

0.5894

0.1877

3.14

Log Age 2

0.0696

0.0239

2.92

Log Company Tenure Months

0.0297

0.0477

0.62

Log Company Tenure

0.0025

0.0049

0.52

0.0065

0.0024

2.64

San Jose

1.4378

0.4146

3.47

Among Defendants

0.0973

0.0493

1.98

0.0008

0.0080

0.10

0.0240

0.0241

0.99

0.2720

0.0617

4.41

0.0661

0.0853

0.78

of

New

Hires In the

Firm
Number

of

Employees

1

Conduct

1
APPLE
Log Total Annual Compensation CPI 1
GOOGLE
Log Total Annual Compensation CPI 1
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

INTEL

Log Total Annual Compensation CPI

INTUIT

Log Total Annual Compensation CPI

PIXAR

1

1
1

Log Total Annual Compensation CPI

LUCASFILM

1

2
APPLE
Log Total Annual Compensation CPI 2
GOOGLE
Log Total Annual Compensation CPI 2
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

INTEL

Log Total Annual Compensation CPI

INTUIT

Log Total Annual Compensation CPI

PIXAR

2

2
2

Log Total Annual Compensation CPI

LUCASFILM

2

2

Male

DLog

Information Sector Employment

Log Total Number
Year

of

Transfers

in

trend

Log Number

of

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLog Firm Revenue

Firm
Number

of

Employees

1

Hires

1
EmployeeCPI 1

EmployeeCPI

0.2068

0.0869

2.38

APPLE

0.1220

0.2718

0.45

GOOGLE

1.3682

0.4309

3.18

INTEL

0.1569

0.2761

0.57

INTUIT

0.1393

0.2268

0.61

PIXAR

1.5864

1.0458

1.52

LUCASFILM

0.0127

0.3184

0.04

State

Location

Per

YES

Indicators

Constant

YES

RSquare

0.874

Observations

Note
Source

Significant

Dr Leamer’s

292,489

at

1

level

Significant

at

backup data and materials

5

level

Standard

Significant

at

10 level

errors clustered on employer year

Case5 11cv 02509LHK

Document518

Filed1007 13

3

Page40

of

46

Exhibit 22A

Dr Leamer’s

Estimates of Undercompensation Are Not StatisticallySignificant
AllSalaried Employee

Adobe

Google

Apple

Dr

Class

Intel

Intuit

Lucasfilm

Pixar

Leamer’s Annual Undercompensation Estimates Figure 22

2005

1.61

1.59

1.78

1.67

12.13

10.56

2006

4.28

4.43

4.44

4.70

14.63

12.44

2007

6.64

6.94

6.39

7.46

17.24

14.28

2008

9.08

9.56

8.40

10.05

5.64

19.94

15.76

2009

9.15

9.73

7.51

9.95

5.70

20.12

14.65

T Statistics

for

3.24

Annual Undercompensation Estimates

2005

0.94

0.74

0.47

0.96

1.17

0.91

2006

0.88

0.81

0.49

1.49

0.98

0.86

2007

0.90

0.80

0.55

1.62

0.86

0.93

0.88

2008

0.90

0.80

0.60

1.63

0.99

0.95

0.79

2009

0.94

0.82

0.64

1.62

1.04

0.96

0.72

P Values

for

Annual Undercompensation Estimates

2005

35.3

46.5

64.1

34.0

24.9

36.8

2006

38.2

42.3

62.7

14.2

33.0

39.3

2007

37.1

42.6

58.7

11.1

39.4

35.5

38.4

2008

37.0

42.6

55.1

10.8

32.6

34.4

43.2

2009

35.0

41.7

52.3

11.2

30.1

34.3

47.7

Notes

1
2

Estimates with
Standard

tstatistics

below

errors are clustered

1.96

in

absolute value

on employer

Source Dr Leamer’s Figure 20 regression data

and

year

or equivalently

with

pvalues

greater than

5

are not

statistically

significant

at

the

95 level

Case5 11cv 02509LHK

Document518

Filed1007 13

3

Page41

of

46

Exhibit 22B

Dr Leamer’s

Estimates of Undercompensation Are Not StatisticallySignificant
Technical Creative and

Adobe

Google

Apple

Dr

RD

Class

Intel

Intuit

Lucasfilm

Pixar

Leamer’s Annual Undercompensation Estimates Figure 24

2005

1.56

1.90

3.07

1.64

10.80

9.28

2006

4.29

4.96

7.23

3.06

14.77

10.47

2007

6.48

7.79

9.36

3.38

3.41

18.08

10.61

2008

8.80

10.64

11.20

4.76

5.21

20.44

11.87

2009

8.44

10.51

9.00

4.19

4.96

20.54

9.62

T Statistics

for

Annual Undercompensation Estimates

2005

0.81

0.77

0.71

0.83

0.91

0.78

2006

0.78

0.79

0.72

0.94

0.85

0.72

2007

0.79

0.80

0.75

0.76

0.79

0.83

0.67

2008

0.79

0.80

0.77

0.81

0.83

0.83

0.61

2009

0.79

0.81

0.80

0.72

0.84

0.83

0.49

P Values

for

Annual Undercompensation Estimates

2005

42.4

44.7

48.2

40.8

2006

43.7

43.0

47.5

35.0

39.9

47.4

2007

43.6

43.0

45.6

44.8

43.1

41.0

50.7

2008

43.5

42.8

44.3

42.4

40.9

41.0

54.1

2009

43.1

42.4

42.8

47.8

40.4

41.2

62.7

Notes

1
2

Estimates with
Standard

tstatistics

below

errors are clustered

1.96

in

absolute value

on employer

Source Dr Leamer’s Figure 23 regression data

and

year

or equivalently

with

pvalues

greater than

36.8

5

are not

statistically

significant

44.1

at

the

95 level

Case5 11cv 02509LHK

Document518

Exhibit
Undercompensation
as

Benchmark

Estimates
Using
in

Dr

AllSalaried

Apple

2005

2.71

3.61

6.33

2006

7.94

9.12

15.64

2007

12.15

14.47

20.77

1.56

2008

16.55

19.95

25.25

2009

15.87

19.92

22.16

Apple

Google

of

46

23
Undercompensation
as

Benchmark

Estimates
Using Post Conduct Period
in

Dr

Leamer’s Regression

AllSalaried Employee Class

Lucasfilm

Pixar

Year

Adobe

Apple

Google

Intel

2.81

14.56

16.52

2005

2.35

2.55

2.76

3.65

22.11

19.53

2006

6.66

6.74

6.80

6.18

27.43

19.88

2007

10.43

10.54

9.43

6.72

2.74

9.00

30.44

23.69

2008

14.40

14.43

11.85

1.37

8.34

30.04

20.65

2009

14.55

14.49

10.20

Intel

Technical Creative and
Adobe

Page42

Employee Class

Adobe

Year

Filed1007 13

vs

Leamer’s Regression

Year

Google

PreConduct Period

3

Intel

Intuit

RD

Pixar

2.29

14.80

12.66

5.08

19.72

15.17

4.83

24.07

16.81

9.43

8.35

27.74

19.25

9.05

8.51

28.06

17.56

Technical Creative and

Class

Intuit

Lucasfilm

Lucasfilm

Pixar

Year

Adobe

Apple

Google

Intel

Intuit

RD

Class

Intuit

Lucasfilm

Pixar

2005

3.46

4.70

8.39

3.54

16.57

18.91

2005

2.33

2.26

1.81

2.25

16.28

11.56

2006

10.10

11.69

20.04

3.90

25.84

21.64

2006

6.47

6.08

4.52

5.96

20.36

13.40

2007

15.29
20.74

18.40
25.15

25.38

7.90
10.96

31.64
34.10

2007

24.35

2008

10.17
14.00

9.38
12.71

6.50
8.46

9.12
12.50

4.58
8.08

24.38

29.55

0.43
1.63

20.55

2008

28.54

14.99
16.28

2009

19.53

24.64

23.64

0.33

9.96

32.41

19.40

2009

14.25

12.62

7.12

12.37

8.24

29.30

14.15

Source

Leamer Figure 20 and 23 regressions estimated

and pre conduct

period

data only

using conduct

Source

Leamer Figure 20 and 23 regressions estimated

and post conduct

period

data only

using conduct

Case5 11cv 02509LHK

Document518

Exhibit
Undercompensation

Estimates Predicted Using

Conduct Period Data

in

AllSalaried

Dr

Apple

Google

Intel

vs

2005

5.01

0.84

0.72

2006

2.65

5.79

2007

4.26

2008

4.67

2009

1.00

Google

1.61

1.59

1.78

1.67

12.13

10.56

2006

4.28

4.43

4.44

4.70

14.63

12.44

4.45

2007

6.64

6.94

6.39

7.46

3.24

17.24

14.28

29.03

2008

9.08

9.56

8.40

10.05

5.64

19.94

15.76

2009

9.15

9.73

7.51

9.95

5.70

20.12

14.65

2.96

2.48

4.52

2005

5.61

2.73

5.99

16.84

12.56

2.34

8.78

6.72

3.78

0.10

18.53

7.36

10.78

3.88

2.21

3.13

7.87

12.05

3.93

32.40

Adobe

Apple

Google

Intel

2005

5.83

0.97

1.89

3.43

2006

2.05

4.03

12.09

1.29

2007

5.83

9.57

7.59

5.47

6.76

2008

5.18

4.33

25.03

2.56

8.81

2009

1.46

2.26

6.45

3.09

10.53

Leamer Figure 20 and 23 regressions estimated

Undercompensation calculated

using residuals

data after 2005 are included

predicted

3.07

1.64

2006

4.29

4.96

7.23

3.06

14.77

10.47

2007

6.48

7.79

9.36

3.38

3.41

18.08

10.61

16.70

2008

8.80

10.64

11.20

4.76

5.21

20.44

11.87

23.03

2009

8.44

10.51

9.00

4.19

4.96

20.54

9.62

2005

24.15

1.52

6.44

1.86
1.90

the conduct

period

Class

1.90

11.66

period

Pixar

1.56

Year

non conduct

RD

Lucasfilm

Google

Pixar

data

Adobe

Intuit

Apple

Lucasfilm

6.07

for

Intel

Technical Creative and

3.05

Intuit

using

Adobe

Class

Year

22 and 24

Apple

Year

RD

46

AllSalaried Employee Class
Pixar

Technical Creative and

of

Figures

Lucasfilm

Intuit

Page43

24

Employee Class

Adobe

Pixar revenue

Filed1007 13

Undercompensation Estimates in Dr Leamer’s

Non

Leamer’s Regression

Year

Source

3

Intel

Intuit

Lucasfilm

10.80

Pixar

9.28

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page44

of

46

Case5 11cv 02509LHK

Document518

3

Filed1007 13

Page45

of

46

Case5 11cv 02509LHK

Document518

Exhibit
Undercompensation Estimates Including Change

SP 500

in

Dr Leamer’s

AllSalaried

Adobe

Apple

Google

2005

0.11

0.06

0.17

0.23

0.27

0.43

0.39

0.44

0.68

1.70

2008

0.55

0.62

1.01

2009

0.66

0.66

1.01

Figures

Google

1.61

1.59

1.78

1.67

12.13

10.56

4.28

4.43

4.44

4.70

14.63

12.44

2007

6.64

6.94

6.39

7.46

3.24

17.24

14.28

2.25

2008

9.08

9.56

8.40

10.05

5.64

19.94

15.76

2.14

2009

9.15

9.73

7.51

9.95

5.70

20.12

14.65

1.64

2005

1.83

1.83

2006

0.22

1.96

2.23

2.22

0.55

2.28

2.32

0.61

2.31

Adobe

Apple

Google

2005

0.48

0.19

0.84

2006

1.20

0.69

1.82

2007
2008

1.93
2.64

1.00
1.32

1.87
1.74

4.26
5.59

2009

2.81

1.40

1.15

5.76

Lucasfilm

Intuit

RD

Class

1.90

3.07

1.64

10.80

9.28

4.29

4.96

7.23

3.06

14.77

10.47

6.48
8.80

7.79

2008

10.64

9.36
11.20

3.38
4.76

3.41
5.21

18.08
20.44

10.61
11.87

2009

8.44

10.51

9.00

4.19

4.96

20.54

9.62

3.49

1.29

2005

3.17

1.43

2006

0.71
1.59

3.38
4.37

2.21
1.86

2007

1.74

4.57

1.65

P 500

Pixar

1.56

0.41

S

RD

Lucasfilm

Google

2.12

Adobe

Intuit

Apple

Year

in

Intel

Technical Creative and
Pixar

Intuit

Leamer Figure 20 and 23 regressions including change

Adobe

Class

Lucasfilm

Intel

22 and 24

Apple

1.90

Year

Bloomberg

vs

Year

Technical Creative and

46

AllSalaried Employee Class

0.84

2007

of

26

Pixar

Intel

Page46

Undercompensation Estimates in Dr Leamer’s

in

Regression

0.17

2006

Source

Filed1007 13

Employee Class

Year

Net Total Return Index

3

Intel

Intuit

Lucasfilm

Pixar

Case5 11 cv 02509 LHK

Document518

4

Filed10 07 13

Page1

of

73

CurriculumVitae

Kevin

M Murphy

October 2012

Home

Address

Business

University

Address

of Chicago

1810 Pennington Court
Booth School of Business

New

5807 South Woodlawn Avenue
Chicago

Illinois

60637

murphy chicagobooth

email kevin

Lenox

Illinois

60451

Phone [phone]
Fax [phone]
edu

Current Positions

July

2005 Present George J

Department

Faculty

of Economics

Research

Stigler

Distinguished

Service Professor

of

and Booth School of Business University

Bureau of Economic

Associate National

Economics

of Chicago

Research

Education

University

of California Los Angeles

University

of

Thesis

Previous

Topic

Chicago

Ph

Specialization

D
and

Research and Academic

2002 2005 George J

Stigler

A B Economics

1981

1986

Human

Capital

Positions

Professor

of

Economics Department

of

Economics and

Booth School of Business University of Chicago

1993

2002 George

Relations

1989

University

1993

Pratt Shultz

Professor

of Business

Economics

and Industrial

of Chicago

Professor

of Business

Associate

Professor

Economics and

Industrial Relations

University

Chicago

1988

1989

University

of Chicago

of Business

Economics

and Industrial Relations

of

Case5 11 cv 02509 LHK

1986

1988

University

Document518

Assistant Professor

of Business

Filed10 07 13

4

Economics and

Page2

of

73

Relations

Industrial

of Chicago

1983

1986 Lecturer Booth School

1982

1983 Teaching

Associate Department of Economics University of Chicago

1979

1981 Research

Assistant

of Business University

Unicon

of Chicago

Corporation Santa Monica

Research

California

Honors and Awards
2008 John von Neumann

Lecture

2007 Kenneth J Arrow Award

1998

Rajk

with Robert

October 2005 Garfield Research

September

Award

Prize

College Corvinus University

Budapest

H Topel

with Robert

H Topel

2005 MacArthur Foundation Fellow

Elected

Academy

to the American

1997 John Bates

of Arts

Sciences

Clark Medalist

1993 Fellow of The Econometric

Society

1989

1991

Sloan

1983

1984

Earhart

1981

1983 Fellowship Friedman Fund

1980

1981 Phi Beta Kappa

1980

1981

1979

1981 Department Scholar Department

Foundation Fellowship University of Chicago

Earhart

Foundation Fellowship University

University

University

of Chicago

of California

Los

Foundation Fellowship University

of

of Chicago

Angeles

of California

Economics

Los

University

Angeles

of California

Los Angeles

Publications

Books

Social

Economics Market Behavior

Cambridge

MA Harvard

University

in a Social Environment
Press

Measuring the Gains from Medical Research
with Robert

H Topel

Chicago

University

with Gary

S Becker

2000
An

Economic Approach

of Chicago Press

edited

volume

2003

2

Case5 11 cv 02509 LHK

Document518

Filed10 07 13

4

Page3

73

of

Articles

Government

of

Journal

and Inference

Estimation
Journal

of Cigarette Health

Regulation

Schneider 24

Lynne

Law

of

Two Step

in

Unemployment

Risk and Earnings

with Benjamin

Klein and

1981
Models

Econometric

with Robert

H Topel

Testing

Wage

for Equalizing

Differences

H

Market with Robert
Topel in Unemployment and the Structure of
Markets pp 103139 ed Kevin Lang and Jonathan S Leonard London Basil
Labor

3

1985

and Economic Statistics 370

Business

Information

and Economics 575

the

in

Labor
Blackwell

1987
The

Evolution

Topel

MA MIT

Unemployment

of

NBER

in

in the United

States

pp 11 58

Annual

Macroeconomics

1968 1985

ed

H

with Robert

Stanley Fischer

Cambridge

1987

Press

Cohort Size and Earnings in the United States with Mark Plant and Finis Welch
Economics
Ronald

Age

of Changing

D Lee W

The Family and

A Theory

Distributions

in

Brian Arthur and Gerry

State with Gary

the

Addiction

of Rational

Developed

Countries

pp

39 58 ed
1988

Rodgers Oxford Clarendon Press

S Becker

with Gary

31

Journal

S Becker

of

96

Law

Journal

Political

1988

1

and Economics

of

in

Economy 675

1988
Vertical Restraints and Contract
and Economics 265

Income
Robert

Wage

Distribution

W Vishny 104

Premiums for
Welch 18

with Finis

of

Building

Robert

Political

Quarterly

of

Economics

537

Researcher

1003

Push with
1989

17

W Vishny

of

Law

in

NBER

Andrei

Macroeconomic

the Theory and Measurement

Weiss and Gideon Fishelson

with Andrei Shleifer

and

1989

Shleifer

Annual

MA MIT

London

of

and Robert

Models

Macmillan

ed

Olivier Jean

1989

with Robert

Unemployment

W Vishny 97

with Andrei Shleifer and

pp 247 87
Press

Wages Reconsidered Theory and Evidence
in

Journal

1989

Blocks of Market Clearing Business Cycle

Efficiency

Klein 31

Graduates Recent Growth and Possible Explanations

College

Educational

Economy

Journal

Blanchard and Stanley Fischer Cambridge

Advances

with Benjamin

Market Size and Industrialization

Industrialization and the Big
Journal

Enforcement

1988

H Topel

pp 204 240 ed

in

Yoram

1990

3

Case5 11 cv 02509 LHK

Age Earnings

Empirical

Document518

Filed10 07 13

4

Profiles with Finis Welch

8

Journal

of

Page4

Labor

of

73

Economics

202

1990
Human

Capital

Tamura 98

Accounting

Political

of

Economy

Slowdown

for the

The Role

Changing

Has

Rational
Michael

Wages

106

Enterprise

Quarterly

Journal

for

Grossman 81

American

Economic

Graduates

of College

121 40 ed
Publishers

E Becker

William

Increased

in

over

Activity

Growth with

and the Effect of Price on

Addiction

F

with Chinhui Juhn
the United

in

Enterprise

Institute

Welch

Finis

States

in

Workers

Marvin Kosters

1991

Institute

Time

H

with Robert

1991

75

Andrei Shleifer and Robert

1991

Economics 503

of

and Robert

pp 39 69 ed

States

Papers on Economic

Brookings

Patterns

with

Differentials

the United

of Talent Implications

Allocation

W Vishny

in

Rate of Unemployment

the Natural

Topel and Chinhui Juhn 2

The

Patterns

DC American

Washington

Why

Trade in Wage

of International

Convergence

DC American

Marvin Kosters Washington

Wages

and Their

Black White Wage

in

S Becker

Gary

1990

S12

and Their Wages Changing

and Brooks Pierce in Workers

pp 107143 ed
1991

Growth with

and Economic

Fertility

Journal

Consumption

S Becker

with Gary

and

1991

Review 237

pp

The Economics of American Higher Education
Lewis Boston Kluwer Academic

R

1992

Changes

Katz 107

in

The
The

Transition

Journal

Wages

Structure of

Shleifer

The

to

and Robert

Division

Quarterly

of

Journal

Industrial

Wages 1963 1987

Relative

Quarterly

of

Wage
101

of

107

Pitfalls

Quarterly

Quarterly

of Partial

Journal

of

Journal

Economics 1137

the Rising
in

Planning

Importance

Economy

of Skill

in

410

1992

Reform with
1992

Andrei

with

pp 101 132 ed

with Finis

Gary

Welch

Peter Gottschalk

Sage Foundation Publications

and the Rise

Political

Economics 285

of

S Becker

107

1992

America

Russell

F

Economics 889

Labor Coordination Costs and Knowledge

York

Inequality

Journal

107

with Lawrence

1992

with Finis Welch

Economy

Demand Factors

Supply and

Economics 35

W Vishny

Change and

New

of

a Market

Tides Rising Inequality
Danziger

and Darrell

in

Uneven

and Sheldon

1993

Returns to Skill with Chinhui Juhn and Brooks Pierce

1993

4

Case5 11 cv 02509 LHK

Change and

Occupational

Document518

for Skill

1940

1990

Page5

with Finis

of

73

Welch 83

1993

Wages

and Relative

Inequality

Demand

the

Economic Review 122

American

Filed10 07 13

4

with Finis

Welch 83

Economic Review 104

American

1993
Why

Is

Rent Seeking

Vishny 83

A Simple
Journal

of

American

So

Costly to

Growth

Theory of Advertising as

a

with Andrei Shleifer

W

and Robert

1993

Review 409

Economic

Good

or

Bad

with Gary

S Becker

108

Quarterly

1993

Economics 941

Relative Wages and Skill Demand 1940 1990 with Chinhui Juhn in Labor Markets
Employment Policy and Job Creation pp 34360 ed Lewis
Solmon and Alec

C

Levenson The Milken
Westview

Cattle

Empirical

Analysis of Cigarette

Grossman 84

in

with Chinhui

Labor Market Outcomes

Juhn 1

in

Wage

Political

of

The

and Michael

Contrasting

the 1980s and Earlier

Decades

1995

Review 26

Hike

with

Donald

R Deere

and

1995

on Minimum Wages and Employment with Donald
Effects

of the

Finis

R Deere

MinimumWage on Employment pp 26 54 ed
1996

H Kosters Washington DC The AEI Press

of

Political

Economy

Inequality

Economics 72

Quality and

Wage

Policy

Review 232

Economic

Social Status Education
Journal

Journal

1994

1990 91 Minimum Wage

the Evidence

and Finis Welch
Marvin

Economic

the

American

Examining

102

S Becker

Addiction with Gary

Economic Review 396

American

Employment and
Welch 85

A Scheinkman

Jose

1994

468

Inequality

Boulder

1994

Press

Cycles with Sherwin Rosen and

Economy

An

Economics and Education

Series in

Institute

R
CO

Growth with

Chaim Fershtman and Yoram Weissm 104

1996

and Family Labor

Supply

with Chinhui

Juhn 15

Journal

of

Labor

1997

Trade with

Inequality

Economics 72

and
108

Andrei Shleifer

and Family Labor

53

Supply

Journal

of

Development

with Chinhui

Economics

Juhn 15

Journal

of

1

1997

Labor

1997

Vertical Integration

Klein 87 American

as a SelfEnforcing Contractual

Economic

Review 415

Arrangement

with Benjamin

1997

5

Case5 11 cv 02509 LHK

Unemployment

Nonemployment

Skills and Technology

M Romen

and Paul

on

Perspectives

the United

in

Page6

73

of

87 American

Economic

Growth
145 1999

Economic Review

A Competitive Perspective
Economic Review 184

Bush School

W

with

Craig Riddell

pp 283

Growth

1998

Press

Solutions

with Finis

Welch 88

on

with Gary

S Becker

Explorer

Internet

and Edward Glaeser 89

with Steven

J

Davis 90 American

2000

Change and

Industrial

Consequences

MA MIT

Canada

and Economic

1998

and Economic

Population

States and

the Social Security Crisis and Proposed

Economic Review 142

American

H Topel

with Robert

in General Purpose Technologies

309 ed Elhanan Helpman Cambridge

American

Filed10 07 13

4

1997

Review 295

Wages

and

Document518

the

Demand

of Increasing

for Skill

Economics of

Series in the

Welch in The Causes and
Welch Volume II in the

with Finis

pp 26384 ed

Inequality

Public

Finis

Policy Chicago University

of Chicago

2001

Press

Wage

Differentials

Welch

the 1990s

Is the Glass Half Full or Half

The Causes and Consequences

in

Volume

in

II

in the

Bush School

of Chicago Press

University

on

Software

in

pp 341 64

Inequality

Economics of

the

Public

ed

Finis

Finis

Welch

Policy Chicago

2001

Economic

Perspectives

with Steven

Economy

Increasing

of

Series

Empty with

J Davis

Design PC Operating Systems and Platforms

and Jack MacCrisken

Selected Essays

in Microsoft

pp 361 420 ed

Davis

Antitrust

S Evans

and the

Boston

New

MA Kluwer

2001
Current

Unemployment

Chinhui Juhn 1

The Economics

Papers on Economic

Brookings

of Copyright Fair

and Benjamin Klein 92 American

The Economic Value

M Murphy

Use

in

Economic

of Medical

Gains from Medical Research
and Kevin

Contemplated

Historically

An

Economics 299

Entrepreneurial
Japanese
Review

of

and

2002

A Networked World with

Andres Lerner

2002

Research with Robert
Economic Approach

Chicago University

H Topel

in

pp 41 73 ed
2003

Measuring the
Robert

H Topel

of Chicago Press

Market

with Sam Peltzman

22

Journal

of

2003

ability

and market selection in an infant industry

cotton spinning
Economic

79

Review 205

School Performance and the Youth Labor
Labor

Activity

H Topel

with Robert

Dynamics

industry with Atsushi
354

Ohyama and

evidence

Serguey

from the

Braguinsky

7

2004

6

Case5 11 cv 02509 LHK

Entry Pricing

Document518

and Product Design

Davis and Robert

H Topel

112

an

in

Journal

of

Filed10 07 13

4

Initially

S188

Diminishing Returns The Costs and benefits of Increased

Topel 46

Perspectives

Persuasion

in

in

Biology

Longevity

of

73

with Steven

with Robert

J

H

2004

S108

and Medicine

Market
2004

Monopolized

Economy

Political

Page7

Politics with Andrei Shleifer

94 American

May

Economic Review 435

2004
Black White
Topel 48
The

Differences

Perspectives

Equilibrium

in

in the

Biology

Distribution of

Becker and IvÆn Werning 113

The Market

for Illegal

Grossman 114

Journal

Two

Competition in

Fees

Interchange
Journal

Economic

Goods The

of

Political

of

Economy

Sided Markets

the Market for

38

The

Drugs with
2006

Antitrust

Status

S

with Gary

2005

Economy 282

Political

Case of

H

2005

Income and

Journal

Health with Robert

Value of Improving

S176

and Medicine

Gary

S Becker

and Michael

Economics of Payment Card

with Benjamin Klein Kevin Green and Lacey Place 73

Antitrust

Law

2006

571

The Value

of Health

Economy 871

and Longevity with Robert

H Topel

2006

Social Value and the Speed of
Review 433

Innovation with Robert

Marketplace

Why
Capital

Fertility

a

1 The Journal of

Journal

in

the Whole Foods

Dealing

in

97 American

the Household

Human

Capital

with Isaac Ehrlich

Case with

Robert

Economic

1

Compared

Winter

9

The Journal

H Topel

Competition for Distribution

Intensifies

Law Journal Vol 75 October

Decline

The Market

Papers

Political

to

2007

of

Human

3

2 GCP

2008

March

Robert Tamura

Women

Need

Capital

Loss Analysis

Exclusive
Antitrust

H Topel

of

2007

1 Winter

Critical

S Becker

with Gary

Does Human

Magazine

Journal

2007

Education and Consumption The Effects of Education
the

114

2

the

Baby

The Journal

for College

with Gary
Proceedings

Boom

and Economic

Human

Capital

3

Growth with
Fall 2008

Graduates and the Worldwide

S Becker
229

of

with Benjamin

Klein

2008

and William

Boom

H J Hubbard

in

Curtis

Simon and

Higher Education of

100 American

Economic Review

May 2010

7

Case5 11 cv 02509 LHK

H J Hubbard

William

Boom

the Worldwide

Explaining

Document518

Journal

Higher Education of

in

Human

of

Filed10 07 13

4

Page8

Women

Capital University

of

with Gary

73

S Becker

of Chicago Press

vol

43 203 2010
How

Exclusivity

with Benjamin

Used

is

Klein 77

to Intensify Competition for Distribution Reply to
Antitrust

Law

Maximum Long Run Growth

Achieving

City Proceedings

of

the

Annual

Hole

Jackson

No

Journal

forthcoming

Conference

Zenger

2011

2

the Federal

in

Bank

Reserve

of

Kansas

2011

Selected Working Papers

Gauging

the

Working

War

Economic

Paper

October

NBER

Working

Estimating

the Effect

Unpublished

Working

The

Interaction

Unpublished

Persuasion

The Value
Philipson

On the

Weighing

No 12092

Growth

Costs

Its

with Steven J Davis and Robert

2006
Steve

Levitt and Roland

Fryer

and

Income

with Gary

S Becker

2006
2007

and Indoctrination with Gary Becker

of Life Near

Unpublished

2006

in Population

Paper

the

March

Epidemic with

of the Crack

End

and Terminal

Care with

Gary

S Becker

and Tomas

2007
Economics of

Working

Selected

of

Paper

Paper September

Working

S Becker

2001

In Iraq Versus Containment

H Topel

11th with Gary

Impact of September

Paper

No

Climate

234

Policy

January

with

Gary

S Becker

2010 Revised September

and Robert

H Topel

2010

Comments

Comment on
Reserve

Bank

Comment
Reform

Causes of Changing
of Kansas

City

Comment on

Answers

Social

of Investment

Feldstein

Equality by Robert

pp 175 81 ed

Andrew J

of Chicago Press

Federal

Debate Medicare
and Thomas R

Rettenmaier

2000

Security and Demographic

Uncertainty

ed
2001

Based Social Security Reform

Chicago University

Z Lawrence

1998

Asking the Right Questions in the Medicare Reform

Issues and

Saving Chicago University

Aspects

Earnings

of Chicago Press

John

by Henning

Y Campbell

Bohn

in Risk

and Martin

8

Case5 11 cv 02509 LHK

Comment on High

Filed10 07 13

4

Page9

of

73

R Varian

by Hal

Technology Industries and Market Structure

2001

Reserve Bank of Kansas City

Federal

Popular

Document518

Press Articles

Rap

The Education Gap

The American

Antitrust with Gary

Rethinking

MarchApril 1990 pp 62

Enterprise

S Becker

Wall

26 2001 pp

February

Journal

Street

pA22
Out

Prosperity Will Rise

S Becker

Gary

Wall

Street

Journal

29 2001 pp pA22

October

The Economics
the request

Articles

Ashes with

of the

NFL Team

of

of the National

Ownership

Football

with Robert

League Players

H Topel

Association

report prepared

at

2009

January

About Murphy

Higher Learning

Clearly

March 12 1989 Jobs

the Middle Class Is

Extensive

reference

Unequal
Business

One

pp 1

Section

Long

article

Kleiman Chicago

Carol

The

about

Tribune

Wages

Structure of

with

Murphy

picture of

Why

Means Higher Earning by

to

Pay Widespread

Day

section

in

pp 1

17 1992

by Louis

S Richman
to

Is

Business

on income

piece

York Times August

14 1990

inequality

Rut of Poverty by Sylvia Nasar New York

Anothers

pp 1

Section

May 21 1990 pp 106

Fortune

education

US by Louis Uchitelle New

Long

Rags to Riches

Studys

Times June

Anxious

Murphy’s work on returns

Long

piece

on

the income

inequality

research

Nobels

Pile

November

Up

for Chicago

4 1993

Business

but

Is

Section

the Glory

pp 1

Featured

a photo of five of the

Murphy

and a paragraph about Murphys

This Sin Tax

is

Win Win

Commentary section

brightest stars

by Christopher

refers to

Murphy

Gone

Long

by Sylvia Nasar New York Times

on Chicago School

piece

on

the economics

faculty

of economics
including

research

Farrell Business

Becker

Week

April

11 1994 pp 30

and Grossmans work on

rational

addiction

Growing

and the economics

inequality

Sunday Globe August
details

21 1994 pp

about Murphy and

replaced

the old in

his

of fragmentation

A1 Two page

research

article

part of a series

by David Warsh Boston

with picture and biographical

about

how

the

new

generation

economics

9

Case5 11cv 02509LHK

A Pay

Impact by

Raises

Section

pp 1

Articles

featuring

January

Filed1007 13

4

Page10

of

73

New York Times January 12 1995 Business
minimum wage

of proposed increase in the

Murphy’s comments on the minimum wage appeared in numerous
including the Chicago Tribune

Murphy was

in addition

on

interviewed

26 1995
American Wall

The Undereducated
in

Louis Uchitelle

about consequences

Article

other publications

CNN

Document518

Street

Journal August

19 1996 pp A12

Changes

the rate of returns to education

M Murphy Winner

In Honor of Kevin

Welch 14

Journal

Testimony Reports and Depositions

Canadian

Vehicles
District

15 16

January

2008

M Murphy
Inc

Appliances

v

February

1 2008

Tyco Healthcare

M Murphy

of Kevin

Declaration

Corporation

v

Eastern

Inc

v

the Central District

v

Inc
Inc

Electronics

its

Inc

Trustee

States District

v

Inc

et

al

v

Electronics

its

Inc

Trustee

States District

Initial

Inc
Jaco

Hynix

Hynix

22 2008

States District

Inc
Inc

et

v

Court for the Northern

al

NA

in

District

Court for

the Matter of Sun Microsystems

Inc

Semiconductor

Inc

Hynix

District

v

Inc
et

et

al

Hynix Semiconductor

al

v

Inc

et

All American

DRAM Claims Liquidation

Semiconductor

et

al The United

of California San Francisco

Division

24 2008 in the Matter of Sun Microsystems
Inc et al Consolidated Unisys Corporation v

Jaco

Inc

Electronics

Semiconductor

Bank

Court for the Northern

Submission of Kevin

United States

al Consolidated Unisys Corporation

Semiconductor

Fargo

in the Matter of Allied Orthopedic

April

Hynix

Hynix

Wells

et

7 2008

Electronics

Bank

M Murphy
v

United States District

in the Matter of Novelis

LP The

Semiconductor

Fargo

Hynix Semiconductor

Semiconductor
Trust by

al

Court for the Northern

Hynix Semiconductor

al Edge

et

v

Wells

Deposition of Kevin

LP The

District

M Murphy March

Hynix Semiconductor

Semiconductor

Group

of California Western

Hynix Semiconductor

Trust by

Court for the

District

28 2008

February

Tyco Healthcare

Expert Report of Kevin

al Edge

States District

Motor

Division

M Murphy

Deposition of Kevin
Appliances

February

Anheuser Busch Inc The United

Ohio

of

al

New

the Matter of Allied

in

Group

Court for the Central District of California Western

et

the Matter of

The United

Export Antitrust Litigation

Orthopedic

Inc

in

Maine

of

Expert Report of Kevin

District

by Finis

2000

Last 4 Years

M Murphy

Deposition of Kevin

Medal

of the John Bates Clark

193

Economic Perspectives

of

NA

Inc

Hynix

District

v

Inc
et

al

et

Hynix Semiconductor

al

DRAM Claims Liquidation

Semiconductor

et

al The United

of California San Francisco

M Murphy October

Inc

All American

6 2008 in the

2006

Division

MSA

Adjustment

Proceeding

10

et

Case5 11cv 02509LHK

Filed1007 13

4

M

Murphy October 29 2008
myFICO Consumer Services Inc vs

Expert Report of Kevin
Corporation and

Document518

LLC Experian Information
Solutions LLC and Does I through

Information Services
VantageScore
District

Systems Inc

v News

Court for the

District

v News

FSI Inc

America

America Marketing

In Store

LLC The United
No 07 706645

Services

Deposition of Kevin
Corporation and

VantageScore

21 2008

Equifax

United States

LLC
Court

District

ak

a

News

Inc

Services

in the Matter of Valassis

a k a News

America Incorporated

America Marketing

America Marketing FSI

aa

a

News

LLC and

M Murphy December 12 2008
Services

News

In Store

American Marketing

States Third Circuit Court of Michigan Detroit

myFICO Consumer

Information Services

Division

in the Matter of Fair Issac

Inc vs Equifax Inc Equifax

LLC Experian Information
LLC and Does I through

Solutions

Solutions

X The

Inc TransUnion

United States

LLC

District

Court

of Minnesota

Deposition of Kevin

v News

Rebuttal

X The

Inc

Inc TransUnion

21 2008 in the Matter of Insignia
In Store Inc The United States District

M Murphy November

Group News

District

Solutions

of Minnesota

Communications Inc

Inc

in the Matter of Fair Issac

Equifax

M Murphy November

America Marketing

Expert Report of Kevin

District

73

of

of Minnesota

Expert Report of Kevin

Case

Page11

M Murphy December 15 2008

America Marketing

In Store Inc The

in the Matter of Insignia Systems

United States

District

Court for the

of Minnesota

Expert Report of Kevin

Valassis Communications

Marketing

Group News

Inc

M Murphy December

v News

26 2008

America Incorporated

America FSI Inc

ak

a

News

aa

Division Case

a

the Matter of

News

America Marketing

News America Marketing In Store Services Inc
In Store Services LLC The United States Third Circuit
and

in

ak

a

News

America

FSI LLC

American Marketing

Court of Michigan Detroit

No 07 706645

Final Submission of Kevin

M Murphy

January

16 2009

in

the 2006

MSA Adjustment

Proceeding

Expert Report of Kevin

v
of

M Murphy

January

Corp et al The United
New York Report submitted on behalf

Amerada Hess

of Kevin

Declaration

Inc

v News

District

M Murphy

America Marketing

January

23 2009

in

States District

the Matter of City of

Court for the Southern

of Citgo Petroleum

29 2009

In Store Inc The

in

New

York

District

Corporation

the Matter of Insignia Systems

United States

District

Court for the

of Minnesota

11

Case5 11cv 02509LHK

Document518

M Murphy

Deposition of Kevin

Communications Inc

v News

Group News

FSI Inc

America

America Marketing

Case

In Store

LLC The

Services

v

Services

News

a

a k a News

Inc

America Marketing

LLC and

America Marketing FSI

aa

News

a

News

In Store

American Marketing

Division

M Murphy

Citgos

Center on behalf
States District

all

others

Court for the Eastern

Center on behalf

supply

M Murphy March

of itself and

Deposition of Kevin

States District

RFG

share of total

Expert Report of Kevin

all

Case

Declaration

of Kevin

Declaration

of Kevin

M Murphy
v

District

17 2009

April

v

6 2009

v

April

16 2009

in

April

23 2009

in

of California San Francisco

Expert Report of Kevin

M Murphy May 11 2009

Judicial

District

v

Inc

Meritor

LLC

States District

Meritor

LLC and

States District

v

Division

Inc

Sun Microsystems

a Korean

al The

corporation

United States

Hynix

District

Division

in the Matter of Jim

Corporation

Microsoft

Sun Microsystems
al The United States

et

the Matter of

a California corporation et

State of Mississippi

Expert Report of Professor

the Western

United

the Matter of

Hynix Semiconductor

District

Petroleum Corporation

ZF

The United

of California San Francisco

Court Northern

First

ZF

The United

in the Matter of

Eaton Corporation

Inc

rel

Medical

Inc The

Division

the Matter of

in

United

Division

CR Bard

Eaton Corporation

Semiconductor America

Hinds County

vs

Medical

Inc The

in the Matter of St Francis
situated

Hynix Semiconductor

M Murphy

a California corporation

General ex

CR Bard

No 06 CV 623

a California corporation

Court Northern

vs

No 06 CV 623

M Murphy

Case

the Matter of St Francis

situated

of Missouri Southeastern

District

v

in

York Harbor

of Missouri Southeastern

similarly

M Murphy March

TransmissionCorporation

Court of Delaware

District

New

the Matter of City of

in

New

3 2009

District

and Meritor TransmissionCorporation

Deposition of Kevin

the

similarly

others

Court for the Eastern

Court of Delaware

at

M Murphy March 6 2009

of itself and

Expert Report of Kevin

Meritor

13 2009

February

Amerada Hess Corp et al The United States District Court for the Southern
of New York Report submitted
on behalf of Citgo Petroleum Corporation

regarding

Inc

in the Matter of Valassis

America Incorporated

ak

73

of

No 07 706645

District

Inc

10 2009

February

Page12

United States Third Circuit Court of Michigan Detroit

Expert Report of Kevin
York

Filed1007 13

4

Hood

The Chancery

Attorney

Court of

District

Kevin

M Murphy

Ranger Enterprises

June

12 2009

Inc The

in the Matter of

CITGO

United States District Court for

of Wisconsin

12

Case5 11cv 02509LHK

Expert Report of Kevin

v

Incorporated
District

Trial

Filed1007 13

4

M Murphy June 24 2009

Page13

of

73

in the Matter of Novell

The United

Corporation

Microsoft

Court Northern

States District

of Maryland

M Murphy

Testimony of Kevin

Communications Inc

Group News

v News

July

16 2009

the Matter of Valassis

in

a k a News

America Incorporated

ak

America Marketing

FSI Inc
a News America Marketing FSI LLC and News
In Store Services Inc
a a News American Marketing In Store

America

America Marketing

LLC The United
No 07 706645

Services

Case

Document518

a

States Third Circuit Court of Michigan Detroit

M Murphy August 14 2009

Declaration

of Kevin

Litigation

The United

Declaration

submitted

of

EBay

Antitrust

Seller

District of California

Ebay Inc s motion

of defendant

support

in

in the Matter

Court for the Northern

States District

Division

for

summary

judgment

Expert Report of Kevin

Inc

and

S Jerrold

M Murphy August 21 2009

Kaplan

v

in the Matter of

The

Corporation

Microsoft

Superior

Go Computer

Court for the

State

of California for the City and County of San Francisco

Deposition of Kevin

v

Incorporated
District

M Murphy

Microsoft

16 2009 in the Matter of Novell
The United States District Court Northern

September

Corporation

of Maryland

Deposition of Kevin

M Murphy

September

21 2009

The United

States District

California Deposition in support

of defendant

Antitrust Litigation

Ebay

in the Matter of

Court for the Northern

Ebay Inc s motion

for

Seller

of

District

summary

judgment

Expert Report of Kevin

M Murphy September

Temperature Sales Litigation The United States

Trial

Testimony of Kevin

M Murphy October

and Meritor TransmissionCorporation
Court of Delaware

Declaration

of Kevin

California Declaration

District

1 2009

the Matter of

Motor

Fuel

Court of Kansas

ZF

in the Matter of

Eaton Corporation

The United

LLC

Meritor

States District

No 06 CV 623

Case

Antitrust Litigation

v

29 2009 in

M Murphy October
The United
in

16 2009

States District

further support

in the Matter of

Ebay

Court for the Northern

of defendant

Seller

District

Ebay Inc s motion

for

of

summary

judgment

Expert Report of Kevin
Devices
Intel

Inc

and

Kabushiki

AMD

M Murphy October
International

Kaisha The United

Sales

20 2009
Service

States District

in the Matter of

LTD v

Intel

Court for the

Advanced Micro

Corporation and

District

of

Delaware

13

Case5 11cv 02509LHK

M Murphy

Deposition of Kevin

S Jerrold

and

Document518

v

Kaplan

Filed1007 13

4

24 2009

October

Court

Superior

73

of

Go Computer Inc

the Matter of

in

The

Corporation

Microsoft

Page14

the State of

for

California for the City and County of San Francisco

M Murphy

Deposition of Kevin

Expert Report of Kevin

Discount

New

Court for the Eastern District of

Valassis Communications

Group News

Inc

Card

States District

ak

ak

America FSI Inc

a

News

a

in

the Matter of

News

a

News

America

FSI LLC

America Marketing

American Marketing

Court of Michigan Detroit

No 07 706645

Testimony of Kevin
and

21 2009

America Incorporated

aa

Inc

The United

M Murphy December

v News

News America Marketing In Store Services Inc
In Store Services LLC The United States Third Circuit

Trial

Fuel

the Matter of Payment

in

Antitrust Litigation

and

Division Case

Motor

Court of Kansas

York

Supplemental Expert Report of Kevin

Marketing

the Matter of

in

States District

M Murphy December 14 2009

Fee and Merchant

Interchange

26 2009

October

The United

Temperature Sales Litigation

S Jerrold

M Murphy

Kaplan

v

11 2010

January

The

Corporation

Microsoft

in

the Matter of

Go

Computer

Court for the

Superior

State

of California for the City and County of San Francisco

Supplemental Rebuttal

America Marketing

LLC

News

and

Marketing
Detroit

Group News

v

Services

Division Case

LLC

Communications Inc

v News

Group News

FSI Inc

America

America Marketing
Services

Case

In Store

LLC The

a k a News
Inc

Services

The United

aa

a

the

in
a

News

America Marketing

News

States Third Circuit

FSI

American

Court of Michigan

26 2010

January

in

the Matter of Valassis

America Incorporated

ak

News

a

Services

Inc

a k a News

America Marketing

America Marketing FSI

aa

a

News

LLC and

American Marketing

United States Third Circuit Court of Michigan Detroit

News

In Store
Division

No 07 706645

Declaration

of Kevin

Antitrust Cases I and
the

In Store

ak

No 07 706645

M Murphy

Deposition of Kevin

14 2010

January

News America Incorporated

America FSI Inc

America Marketing

In Store

M Murphy

Expert Report of Kevin

Matter of Valassis Communications Inc

County

Declaration

M Murphy
II The

28 2010

in

the Matter of Automobile

Court of the State of California for

of San Francisco

of Kevin

M Murphy

Determination of Interim

al The United

Court for the Eastern

Court Southern

M Murphy

Fee and Merchant
District

2 2010

April

April

Discount
of

in

the Matter of the Application

Fees for The Cromwell

License

States District

Deposition of Kevin
Interchange

January

United States Superior

New

13 14

District

2010

in

of

Group Inc

New

et

York

the Matter of Payment

Antitrust Litigation

for the

and Affiliates

The United

Card

States District

York

14

Case5 11cv 02509LHK

Document518

M Murphy June 1 2010

Supplemental Expert Report of Kevin

Inc

Insignia Systems

2010

The United

v

News America Marketing

M Murphy June 21 2010

Comcast Corporation General
to

In Store

Assign Licenses or Transfer

Page15

73

of

in the Matter of

Inc

8

June

corrected

Court for the District of Minnesota

States District

Expert Report of Kevin

Filed1007 13

4

in the Matter of Applications

NBC Universal

Company and

Electric

of Licensees

Control

Inc

of

for Consent

Communications

Federal

Commission
Supplement to Expert Report of Kevin
Payment

Card Interchange

M Murphy June 24 2010

Fee and Merchant

Discount

Court for the Eastern District of

States District

Inc

Insignia Systems
District

Court for the

Deposition of Kevin

v

The United

New York

M Murphy

Second Supplemental Expert Report of Kevin

in the Matter of

Antitrust Litigation

July

6 2010

in

the Matter of

News America Marketing In Store Inc The United

District

States

of Minnesota

M Murphy

8 2010

July

in the Matter of Insignia Systems

News America Marketing In Store Inc The United

Inc

Court for the

States District

v

District

of Minnesota

M Murphy 28 2010
by Thomas W Corbett Jr

Expert Report of Kevin
Pennsylvania

July

of Pennsylvania

Commonwealth

Court of Pennsylvania

Response of Kevin

19 2010

M Murphy

of Licensees

NBC Universal Inc
Federal

Louis et al

v

of

al

et

in

the

MD 2004
Israel

and Michael Katz August

Comcast Corporation General

Electric

Control

or Transfer

Commission

M Murphy September

14 2010 in

the Matter of City of

Co et al The Circuit Court of

American Tobacco

of

St

St Louis

the City of

of Missouri

State

Deposition of Kevin
et

212

Inc

Products

for Consent to Assign Licenses

Communications

Expert Report of Kevin

No

Commonwealth
General of the

as Attorney

Pharmaceutical

Reply Report of Mark

to

in the Matter of Applications

Company and

the Matter of

his capacity

in

v TAP

Commonwealth

in

al

v

M Murphy

American Tobacco

Co

et

24 2010

September

al The

in

St Louis

the Matter of City of

Court of the City of St Louis State of

Circuit

Missouri

Supplemental Expert Report of Kevin

Commonwealth
al

in the

Commonwealth

Expert Report of Kevin
Hampshire

v

by Thomas

of Pennsylvania

General of the Commonwealth

M Murphy September

W Corbett

of Pennsylvania

v TAP

Court of Pennsylvania

M Murphy October

Hess Corporation

et

al The

No

1 2010

State of

30 2010

Jr in his

capacity

Pharmaceutical

212

in the Matter of

Inc

MD 2004

in the Matter of State

New

as Attorney

Products

of

Hampshire Superior

New

Court

15

et

Case5 11cv 02509LHK

Expert Report of Kevin
between

Cordis

Conflict

Prevention

M Murphy October

Corporation and Abbott

Cordis

Conflict

M Murphy

Prevention

of Kevin

M Murphy November
v BP

Court for the Northern

CPR

Expert Report of Kevin

8 2010

v BP

Inc

Analysis of Kevin

Dodd Frank

Comments

of Kevin

L Katz

and Michael

of

International

RWJ
States

Division

in the Matter of

RWJ

America Inc The United

19 2010

States

Division

al

in the Matter of Craft et

v

a corporation and Philip Morris Incorporated a

M Murphy

Bank

Judicial

District City of

St Louis

to Guide Interpretation

of Debit Interchange

or Transfer

on

the

November 10 2010 Report

November 24 2010

Control

v NFL

of Provisions of the

Fees November 23 2010

of America Corporation

M Murphy

Expert Report of Kevin

in

of Licensees

the Matter of Applications

NBC Universal Inc

Federal

Lockout

29 2010
Loans The

Drs Mark
of

Israel

Comcast

for Consent to Assign

Communications

M Murphy November

Lockout Insurance

of

Commission

in the Matter of Reggie

United States

District

White

Court

of Minnesota

Deposition of Kevin

v NFL

District

CPR

the Matter of

of Illinois Eastern

District

Corporation General Electric Company and

al

in

15 2010

Products North

Act Regarding Regulation

submission on behalf

al

the Matter of the

00200406 02

Economic

District

for

Institute

America Inc The United

corporation Missouri Circuit Court Twenty Second

et

in

Vascular

12 2010

M Murphy November

Philip Morris Companies

Licenses

International

of Illinois Eastern

District

Court for the Northern

No

for

Institute

the Matter of the Arbitration

in

Products North

M Murphy November

Management Company Inc

Case

International

Resolution

Prevention

Expert Report of Kevin

District

7 2010

Vascular

M Murphy November

Management Company Inc
District

73

of

in the Matter of the Arbitration

CPR

Cordis Corporation and Abbott

for Conflict

Declaration

Page16

Resolution

Testimony of Kevin

Institute

4 2010

Vascular

October

Corporation and Abbott

Arbitration between

Filed1007 13

4

Resolution

Deposition of Kevin
between

Trial

Document518

Lockout

M Murphy December 3 2010
Insurance

Lockout

in the Matter of Reggie

Loans The

United States

District

White

et

Court

of Minnesota

Deposition of Kevin

Company Inc
the Northern

v BP

District

M Murphy December 13 2010
Products North

in the Matter of

America Inc The United

of Illinois Eastern

RWJ

Management

States District

Court for

Division

16

Case5 11cv 02509LHK

Deposition of Kevin

Document518

M Murphy
Inc

Philip Morris Companies

Filed1007 13

4

17 18

January

2011

No

73

of

v

al

the Matter of Craft et

a corporation and Philip Morris Incorporated a

corporation Missouri Circuit Court Twenty Second
Case

in

Page17

St Louis

District City of

Judicial

00200406 02

Report of Kevin

M Murphy
16

Corporation on February

Consumer

15 2011

February

submitted

by

TCF Financial

2011 to the Subcommittee on Financial

Credit of the Committee

on

Financial Services of the

and

Institutions

US House

of

Representatives

Declaration

Ben

of Kevin

S Bernanke

M Murphy March

Janet

L Yellen

Kevin

2 2011

in

M Warsh

Elizabeth

and Sarah Bloom Raskin the Board of Governors
capacities and John

official

Expert Report of Kevin

LTD

M Murphy

Declaration

Court Northern

of Kevin

Board on behalf

Entertainment Consoles

Judicial

June

District

System in

Reserve

in his official

the Matter of Datel

Microsoft

v

Corporation

their

capacity

Holdings

The United

14 2011

v

filed

with

the National

Labor Relations

Players Association

in

the Matter of Datel

Corporation

Microsoft

LTD

Holdings

The United

States

of California

M Murphy

July

1 2011

in

the Matter of Certain

Gaming and

and Components Thereof The United States

Related Software

Trade Commission

Expert Report of Kevin

Inc

Basketball

M Murphy

Expert Report of Kevin

Airlines

v

M Murphy May 26 2011

Court Northern

International

of the Currency

in

Bank

National

A Duke Daniel K Tarullo

of the Federal

11 2011

TCF

of California

Development Inc

and Datel Design
District

District

of the National

Deposition of Kevin

April

Development Inc

and Datel Design

States District

Walsh Comptroller

the Matter of

v

Sabre

M Murphy August 17 2011

Inc

et

al The

Judicial

District

in the Matter of American
of Tarrant

County Texas

67th

District

Expert Report of Kevin

M Murphy August 19 2011

Temperature Sales Litigation

The United

in the Matter of

States District

Court for the

Motor

Fuel
of

District

Kansas

Deposition of Kevin

M Murphy

September

6 2011

and Entertainment Consoles Related Software
States International

v

Intel

the Matter of Certain

Gaming

Trade Commission

Expert Report of Kevin
York

in

and Components Thereof The United

M Murphy September

Corporation

The United

9 2011

States District

in the Matter of State of

Court for the

District

of

New

Delaware

17

Case5 11cv 02509LHK

Deposition of Kevin

Document518

M Murphy

Temperature Sales Litigation

14 2011

September

The United

Filed1007 13

4

Page18

in the Matter of

Court for the

States District

73

of

Motor

Fuel

District

of

Kansas

Direct Testimony

M Murphy

of Kevin

Gaming and Entertainment Consoles

York

v

Report of Kevin

NRLC

and

M Murphy

M Murphy October

railroad

810

October

The United

Corporation

Intel

2011

States District

10 2011

employees National

Hearing Testimony of Kevin

NRLC

and

Thereof The

in connection

railroad

before

M Murphy October
employees National

of

District

A13569 A13570

Emergency Board

13 2011

in

No

243

connection with dispute

Mediation Board Case

A13570 A13572 A13573 A13574 A13575 A13592

Delaware

between

with dispute

Nos

New

of

in the Matter of State

Court for the

Mediation Board Case

A13572 A13573 A13574 A13575 A13592
between

in the Matter of Certain

Software and Components

Trade Commission

United States International

Deposition of Kevin

27 2011

September

Related

before

Nos

Emergency

A13569
No

Board

243

Expert Report of Kevin
Hampshire

Declaration

v

M Murphy October

Hess Corporation

of Kevin

et

al The

17 2011

M Murphy December 1 2011

Temperature Sales Litigation

The United

Hampshire Superior

the Matter of

States District

New

in the Matter of State of

New

State of

Court

Motor Fuel

Court for the

District

of

Kansas

Expert Report of Kevin

M Murphy December 5 2011
v

the Matter of Retractable

in

Technologies

Inc and Thomas Shaw

States District

Court for the Eastern District of Texas Marshall

Trial

Testimony of Kevin

Incorporated
District

Trial

v

Microsoft

Corporation

The United

in

Company The

United

Division

the Matter of Novell

States District

Court Northern

of Maryland

Testimony of Kevin

M Murphy December
v BP

Court for the Northern

M Murphy

Technologies

Inc and Thomas Shaw

The United

States District

Court for the Eastern

M Murphy

January

and Entertainment Consoles Related Software
States International

the Matter of

RWJ

America Inc The United

of Illinois Eastern

District

Retractable

Testimony of Kevin

29 2011 in

Products North

Supplemental Expert Report of Kevin

Trial

and

M Murphy December 7 8 2011

Management Company Inc
District

Becton Dickinson

January

v

States

Division

15 2012

in

the Matter of

Becton Dickinson and Company

District

18 2012

of Texas Marshall

in

Division

the Matter of Certain

Gaming

and Components Thereof The United

Trade Commission

18

Case5 11cv 02509LHK

Document518

M Murphy February

Supplemental Expert Report of Kevin
State

New

of

v

Hampshire

Filed1007 13

4

Page19

23 2012

the Matter of

in

New

Hess Corporation et al The State of

73

of

Hampshire

Court

Superior

M Murphy March

Affidavit of Kevin

12 2012

Fruth Individually and on Behalf

Michael

of Sharon

in the Matter

of Others Similarly

Incorporated The United States Circuit Court Third

Situated

Price and

vs

Philip Morris

Court Madison

Judicial

County

Illinois

of Kevin

Declaration

M Murphy May 3 2012
v

in

the Matter

Inc and Thomas Shaw

States District

Court for the Eastern District of Texas Marshall

Comments

of Kevin

M Murphy

of the Commissions

Revision

DIRECTV Group Inc
Authority

Assignees

Subsidiaries

Sabre
Judicial

Inc

Sabre Holdings

of Kevin

Capital Partners

LLC

July

Corp and

July

The United

20 2012

67th

21 2012

July

v

Motorola Inc The United

Washington

at

the Matter of American Airlines

Ltd The

International

Court

23 2012

Corporation

Commission

in the Matter of Kirk Dahl

in

v

United States

in

District

v

District

Bain

of Massachusetts

the Matter of Kirk Dahl

Court

24 2012

July

and or

Corporation and

District

Judicial

States District

M Murphy

in

Sabre Travel

States District

M Murphy
The United

Expert Report of Kevin

Corporation and the

Time Warner Cable Inc

to

Communications

Federal

M Murphy

LLC

Expert Report of Kevin
Capital Partners

al

County Texas

Tarrant

District

Declaration

et

M Murphy

Expert Report of Kevin

the Matter of

for Consent to the Assignment

Communications

Debtorsin Possession Assignors

Subsidiaries

in

Transferors and Liberty Media Corporation Transferee for

of Control of Licenses Adelphia

Transfer

United

Division

2012

News

Program Access Rules

Company The

and

LLC June 22

DirecTV

Control Applications

Transfer

to

of

Becton Dickinson

of Retractable

Technologies

v

Bain

of Massachusetts

the Matter of Microsoft

Seattle

Deposition of Kevin

M Murphy

August

States District Court Western

22 2012

Corporation

v

Motorola Inc The United

Washington

at

in

District of

the Matter of Microsoft

Seattle

Economic
Evidence

Analysis of the Impact

from San

Diego

Commissions Program

Inc

Transferors

August

Control Applications

Federal

31 2012
News

submitted

in

District of

of Carrying an

the Matter of Revision

Corporation and the

and or

RSN
of the

DIRECTV Group

Corporation Transferee for Authority
Transfer

to Transfer

of Control

of

Corporation and Subsidiaries Debtorsin

Assignors to Time Warner

Communications

Court Western

Subscribership

for Consent to the Assignment

Licenses Adelphia Communications
Possession

on DIRECTVs

Access Rules

and Liberty Media

States District

Cable Inc

Subsidiaries

Assignees

et

al

Commission

19

Case5 11cv 02509LHK

Expert Report of Kevin

Brown

et al v

Document518

M Murphy September

The American Tobacco

California for the County of San

Deposition of Kevin
et

of San

Deposition of Kevin

Inc

Inc

v Sabre

State

of Texas

Court for the

Sabre Holdings

et

73

of

in the Matter of Willard

al Superior

in

al Superior

September

Corp and

District

24 2012

Court for the

R

State

of

the Matter of Willard

R Brown

Court for the State of California

Properties and

in

the Matter of American Airlines

Sabre Travel International

of Tarrant

M Murphy October

Innovative

District

et

14 2012

September

Co Inc

M Murphy

for the Judicial

v 3M

7 2102

Page20

Diego

Expert Report of Kevin
Corporation

Co Inc

Filed1007 13

Diego

M Murphy

al v The American Tobacco

for the County

4

LTD

for the

County

10 2102

in the Matter of Avery Dennison

3M Company

The United

States District

of Minnesota

20

Case5 11cv 02509LHK

Appendix

B Materials Relied

Document518

Filed1007 13

4

Page21

of

73

Upon

Court Documents
Motion and Motion for Class Certification

Plaintiffs Notice

of

Support

Consolidated Amended Complaint

in

Re High Tech

and

Memorandum

Law

1 2012

October

Employee Antitrust

Litigation

of

in

September

2 2011

E Leamer PhD

Expert Report of Edward

1 2012

October

Leamer Backup
Plaintiffs First Set of Requests

M

Declaration

of Tina

Declaration

of

of

Documents

October

3 2011

of Chris Galy

Declaration

for Production

in Support of Opposition to Class Certification

Evangelista

Danny McKell

in Support of Defendants

Opposition

to

Plaintiffs

Motion

for

Class Certification
Declaration
Plaintiffs
Declaration

of

Donna Morris

of

Adobe Systems Inc

in Support of

Defendants Opposition

to

Motion for Class Certification
of Frank

Wagner

in

Support of Defendants Opposition to Plaintiffs Motion for

Class Certification
Declaration
Plaintiffs
Declaration

of Jeff Vijungco

of

Adobe Systems Inc

in

Support of Defendants Opposition to

Motion for Class Certification
of Lori

McAdams

Defendants Opposition

in Support of

to Plaintiffs

Motion

for

Class Certification

Mason

Declaration

of

Stubblefield

Declaration

of Michelle

Maupin

in

Support of Defendants Opposition to Plaintiffs Motion for

Class Certification
Declaration

of Steven

Burmeister

in

Support of Defendants Opposition to Plaintiffs Motion for

Class Certification
Declaration

of

Rosemary

Opposition to Plaintiffs Motion
Deposition of Lori

Keiper of Adobe Systems Inc in Support of Defendants

Arriada

for Class Certification

McAdams and

Deposition of Arnnon Geshuri and Exhibits

Deposition of Donna Morris and Exhibits

Deposition of Jeffrey Vijungco

3 2012
October 5 2012

August

Deposition of Daniel Stover and Exhibits
Deposition of Mark Fichtner and Exhibits
Hariharan

and

24 2012

October

October
October

Exhibits

Deposition of Edward Leamer and Exhibits

23 2012

October

Exhibits

Deposition of Brandon Marshall and Exhibits

Deposition of Siddharth

2 2012

August

and Exhibits

Devine and

October

August 21 ,2012

Deposition of Mark Bentley and Exhibits
Deposition of Michael

17 2012

August

Deposition of Danielle Lambert and Exhibits

Deposition of James Morris and Exhibits

2 2012

August

Exhibits

22 2012

29 2012
15 2012

October

October

12 2012

26 2012

Case5 11cv 02509LHK

Document518

Filed1007 13

4

Deposition of Jack Gilmore and Exhibits

June
June

of

73

28 2012

Deposition of Denise Miller and Exhibits

Page22

28 2012
June

Deposition of Shawna

Dougherty

June

Deposition of John Schirm and Exhibits

Yu and

Deposition of Matthew
Deposition of Shiloh

Exhibits

Howard and

Kuz and

26 2012
29 2012

June

July

17 2012

Exhibits

26 2012

June

Exhibits

Deposition of

and

AmberGay Remaley

Exhibits

Plaintiff

Michael

March

21 2012
June 21 2012

June

and Exhibits

Cruzat and Exhibits

26 2012

June

Deposition of Mary Kathleen Galle and Exhibits
Deposition of Eleterio

26 2012

June

Deposition of Robert DeMartini and Exhibits
del Torro

17 2012

July

Deposition of Michelle Deneau and Exhibits

Deposition of Rebecca

12 2012

and Exhibits

Deposition of Mai Tran and Exhibits

Deposition of Jaime

27 2012

July

Deposition of Steven Burmeister and Exhibits

June

June

21 2012

22 2012

Answers and Objections

27 2012

Plaintiff

March
Plaintiff

Mark

Devine’s

Answers and Objections

Fichtner

to

to Defendants

Defendants

Set of Interrogatories

First

First

Set of Interrogatories

28 2012
Siddharth Hariharan’s

Plaintiff

Brandon

March

to

Defendants

First

Set of

27 2012

March

Interrogatories

Answers and Objections

27 2012

Plaintiff

March
Final

Answers and Objections

Marshall’s

to

Defendants

Daniel Stover’s Answers and Objections to Defendants

Interviews

in United

Final

States

of

Judgment in United

States

v

Adobe Systems Inc

of

America

America

v

23 2012

Jeff

August

23 2012

Donna Morris Adobe

27 2012

30 2012

August

31 2012

August

30 2012

July

25 2012
19 2012
26 2012

September

Vijungco Adobe

Interview

August

June

Set of Interrogatories

Interview

Interview
Interview

6 2012

30 2012

August

16 2012

with Steve Burmeister Apple
with Seth Williams Google

Interview

Interview

August

with Mark Bentley Apple

Interview

with Frank

with Christina

Wagner Google
Dickenson

with Danny McKell

Interview

Interview
Interview

with Chris Galy
with Michelle

with Laurie

Intel

Intel

with Mason Stubbenfeld

Intuit

Intuit

Maupin

McAdams

Lucasfilm
Pixar

et al

Lucasfilm Ltd

Conducted by Kevin Murphy

August

July

First

Set of Interrogatories

28 2012

Judgment

Proposed

July

First

March

17 2011

May 9 2011

Case5 11cv 02509LHK

Academic
Albert

Document518

Filed1007 13

4

Page23

of

73

Papers

Rees The

Role of Fairness

in

Wage Determination 11

Economics 243

of Labor

Journal

1993

Mas Pay

Alexandre

D

Angrist Joshua
Jersey Princeton

and Jrn

Vol

M

P

Ann

Science

Working Paper

J

of

Journal

of

New

Chapter 8.2

No

Bargaining with

Game Theory Aumann Robert

BV Chapter

J Borjas

and George

Borjas George

Quarterly

Mostly Harmless Econometrics

Cramton and Raymond J Deneckere

Peter

Information Handbook

Consequences

NBER

Pischke

Steffen

3 Amsterdam Elsevier

Bartel

121

Press 2009

University

Ausubel Lawrence
Incomplete

Performance

Reference Points and Police

2006

Economics 783

50

NBER

Sergiu Hart

eds

2002

Middle Age Job Mobility

161

J and

Its

Determinants

and

Working Paper Series January 1977

Job Mobility and Earnings Over the Life

Cycle Working

paper

No

233

Working Paper Series February 1978

Davidson
University

Press Inc

E Leamer

Edward

G MacKinnon Econometric

and James

Russell

Theory and Methods

Oxford

2004
Take

Lets

the

Con Out

of

Econometrics 73 The American Economic

Review 1 1983
Freeman Richard
Gary Becker

B and

Nobel Lecture

Economy 385

Political

James

Greene William

June

L Medoff

What

The Economic

Do

Way

of

Unions

Do

Looking

at

New York

Basic

Behavior 101

Books 1984

Journal

of

1993

H Econometric

Analysis

6th Edition Chapter 9.3.3

New

Jersey Pearson

Hall 2008

Prentice

Grossman
Academic

Sanford

Press revised

Hirsch Barry
Competition

Honoree
Individual

J and

T

Motty

Perry

February

Sequential

Bargaining under AsymmetricInformation

2 1986

Sluggish Institutions in a Dynamic

Coexist

Journal

Andre I and David

of

Economic

E Terpstra

and Procedural Equity

to

World Can

Perspectives vol

The

Relative

Unions and Industrial

22 1

Importance

Winter

2008

of External

Internal

Benefits

Review

Pay Satisfaction Compensation

November December 2003
Joseph Stiglitz

Economic
Robert

H Topel

Quarterly

Information and the Change in the Paradigm in

Review 460

Journal

and Michael
of

William Samuelson

Economics 92 American

2002

P Ward

Job Mobility and the Careers of

Young

Men

107 The

Economics 2 1992
Bargaining Under AsymmetricInformation Econometrica

52 1984

Websites

http online wsj comarticle SB10001424052970203750404577173031991814896
http online wsj comarticle SB124269038041932531
http techcrunch com2007 1121 facebook

html

html

stealinggooglers at an alarming
rate

http www aeaweb org honors awards clark medal php
http www dailytechcomGoogle FindsThat PerksCant KeepSomeEmployees FromLe
aving article11794 htm

Case5 11cv 02509LHK

Document518

Bates Documents

76550DOC000014

231APPLE04166
76583DOC001487

Other
Pixar Data

Pixar revenues

2005

2011 xlsx

4

Filed1007 13

Page24

of

73

Case5 11cv 02509 LHK Document5184

Filed1007 13

Page25

of

73

Appendix1A
Analysis of Hires from Other Defendants
AllSalaried Employee Class
Panel

A 2001 2012
Last Previous

Company

Company

within 1 year

Adobe

Apple

Lucasfilm

0

6

0

2

0

Pixar

3

8

6

1

2

222

218

54

293

98

37

Last Previous

Company

Percentage

12

Hiring

Google

Intel

Intuit

Lucasfilm

Pixar

Other

Total

of

Row

Adobe

Apple

Google

Intel

Total
Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

All

Defendants

Panel

35

1,369

0.00

0.44

0.00

0.15

0.00

1,367

0.22

0.59

0.44

0.07

0.15

0.88

91,971

0.24

0.24

0.06

0.32

0.11

0.04

91,014

0.73
0.04

B 2001 2004

Company

within 1 year

Adobe

Apple

Lucasfilm

0

1

0

1

0

Pixar

0

4

0

0

1

34

45

0

34

15

6

Last Previous

Company

Adobe

Apple

0

5

Percentage

3

Hiring

1,351
1,335

10

Google

Intel

Intuit

Lucasfilm

of

Row

Total

Other

Total

Adobe

Apple

Google

Intel

402

407

0.00

0.25

0.00

0.25

0.00

431

Pixar

439

0.00

0.91

0.00

0.00

0.23

0.68

23,181

0.15

0.19

0.00

0.15

0.06

0.03

Other

Total

Adobe

Apple

Google

Intel

788

799

0.00

0.63

0.00

0.13

Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

All

Defendants

Panel

Hiring

3

5

23,042

0.74
0.02

C 2005 2009

Company

Google

Intel

Intuit

within 1 year
Lucasfilm

Percentage
Pixar

of

Row

Total
Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

1

0

0

5

1

3

5

1

1

6

104

97

27

167

44

17

Last Previous

Company

Adobe

Apple

Google

0

0

Pixar
All

Defendants

Panel

Hiring

0.63

674

0.15

0.45

0.74

0.15

0.15

0.89

44,069

0.24

0.22

0.06

0.38

0.10

0.04

Other

Total

Adobe

Apple

Google

Intel

161

163

0.00

0.00

0.00

0.00

657
18

0.00

43,595

0.04

D 2010 2012

Company

Intel

Intuit

within 1 year
Lucasfilm

Percentage
Pixar

of

Row

Total
Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

Defendants

Note
Source

0

2

2

Pixar
All

0

0

This analysis

Dr

1

1

0

0

3

84

76

27

92

39

14

excludes hires indicated as acquisitions

Leamer’s employee

data

and

hires

showing the same defendant

247
12

company

24,377

as their

0.00

1.23

254

0.79

0.39

0.39

0.00

0.00

1.18

24,721

0.34

0.31

0.11

0.37

0.16

0.06

immediate previous employer within one year of the hiring

0.05

Case5 11cv 02509 LHK Document5184

Filed1007 13

Page26

of

73

Appendix 1B
Analysis of Separations Going to Other Defendants
AllSalaried Employee Class
Panel

A 2001 2012
Next

Separation

Company

Adobe

Apple

Google

Company

Intel

within 1 year

Intuit

Percentage of

Lucasfilm

Pixar

Other

Total

Row

Total

Adobe

Apple

Google

Intel

0.00

0.59

0.98

0.07

0.00

0.00

1.46

0.80

0.27

0.00

0.93

0.17

0.45

0.46

0.05

0.10

0.02

Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

0

9

15

1

0

Pixar

0

11

6

2

0

7

122

326

336

35

74

15

All

Defendants

Panel

12

1,490

1,527

726
31

72,287

752
73,226

0.79
0.04

B 2001 2004
Next

Separation

Company

Adobe

Apple

Google

Company

Intel

within 1 year

Intuit

Percentage of

Row

Total

Pixar

Other

Total

Adobe

Apple

Google

Intel

4

Lucasfilm

580

589

0.00

0.51

0.34

0.00

0.00

229

235

0.00

0.85

0.43

0.00

0.00

1.28

0.11

0.22

0.09

0.01

0.09

0.02

Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

0

3

2

0

0

Pixar

0

2

1

0

0

3

28

55

24

3

22

5

Adobe

Apple

Google

All

Defendants

Panel

0.68

9

25,399

Pixar

Other

Total

Adobe

Apple

Google

Intel

5

655

669

0.00

0.45

0.75

0.15

0.00

340

0.00

1.18

0.88

0.59

0.00

35,375

0.20

0.42

0.51

0.05

0.11

0.02

Pixar

Other

Total

Adobe

Apple

Google

Intel

3

255

269

0.00

1.12

2.97

0.00

0.00

177

0.00

2.82

1.13

0.00

0.00

1.13

6

11,513

0.20

1.01

1.10

0.13

0.11

0.02

0.04

0.59

16

25,545

C 2005 2009
Next

Separation

Company

Company

Intel

within 1 year

Intuit

Percentage of

Lucasfilm

Row

Total
Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

0

1

0

0

4

3

2

0

2

151

182

17

39

8

Adobe

Apple

Google

Defendants

Panel

5

70

Pixar
All

3

329

35,858

0.75
0.04

D 2010 2012
Next

Separation

Company

Company

Intel

within 1 year

Intuit

Percentage of

Lucasfilm

Row

Total
Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

0

Source

0

0

0

Defendants

Note

8

This analysis

5

2

0

0

2

24

Pixar
All

3

120

130

15

13

2

excludes separations

Dr Leamer’s

employee

data

that appear as immediately

rehired by the

same defendant

168

company

11,823

within one year

1.12
0.05

Case5 11 cv 02509 LHK

Document518

4

Filed10 07 13

Page27

of

73

Appendix 1C
Analysis of Hires from Other

DNCC Defendants

AllSalaried Employee Class
Panel

A 2001 2012
Last Previous

Company

Hiring

DNCC Defendant

Company

within 1 year

Non DNCC Defendant

Percentage

of

Row

Total

Non

DNCCDefendant

Total

DNCC Defendant

1,369

1.17

98.83

1,367

1.54

98.46

91,971

0.79

99.21

Adobe
Apple
Google
Intel
Intuit

16

Lucasfilm

21

Pixar
All

1,353

Defendants

Panel

91,246

B 2001 2004
Last Previous

Company

within 1 year

Percentage

of

Row

Total

DNCC Defendant

Non DNCC Defendant

Total

DNCC Defendant

4

Company

Hiring

1,346

725

403

407

0.98

99.02

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

7

All

Defendants

Panel

432

439

1.59

98.41

110

Pixar

23,071

23,181

0.47

99.53

C 2005 2009
Last Previous

Company

within 1 year

Percentage

of

Row

Total

DNCC Defendant

Non DNCC Defendant

Total

DNCC Defendant

Lucasfilm

10

789

799

1.25

98.75

Pixar

10

664

674

1.48

98.52

346

43,723

44,069

0.79

99.21

Company

Hiring

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

All

Defendants

Panel

D 2010 2012
Last Previous

Company

within 1 year

Percentage

of

Row

Total

DNCC Defendant

Non DNCC Defendant

Total

DNCC Defendant

Lucasfilm

2

161

163

1.23

98.77

Pixar

4

250

254

1.57

98.43

269

24,452

24,721

1.09

98.91

Company

Hiring

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

All

Defendants

Notes
This analysis

excludes hires indicated as acquisitions

and

hires

showing the same defendant company as

year of the hiring

Adobe
Apple

Google

had DNCC agreements with Adobe Google

allegedly

Lucasfilm

Source

had DNCC agreements with Apple Google and Pixar
had DNCC agreements with Apple and Google

allegedly

Pixar allegedly

Dr

Intel Intuit Lucasfilm

had DNCC agreements with Apple Intel and Intuit

allegedly

Intel allegedly
Intuit

had a DNCC agreement with Apple

allegedly
allegedly

had

had DNCC agreements with Apple and Pixar

DNCC

agreements with Apple Intel and Lucasfilm

Leamer’s employee

data

and Pixar

their

immediate previous employer within one

Case5 11 cv 02509 LHK

Document518

4

Filed10 07 13

Page28

of

73

Appendix 1D
Analysis of Separations Going to Other

DNCC Defendants

AllSalaried Employee Class
Panel

A 2001 2012
Next

Separation

Company

Company

DNCC Defendant

Non

within 1 year

Percentage of

DNCCDefendant

Total

DNCC Defendant

Row

Total

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

21

Pixar

20

All

Defendants

Panel

1,506

712

1.38

72,514

73,226

97.34

0.97

752

98.62

2.66

1,527

732

99.03

B 2001 2004
Next

Separation

Company

Company
Non

within 1 year

Percentage of

Row

Total

DNCCDefendant

Total

DNCC Defendant

582

DNCC Defendant

589

1.19

98.81

235

2.13

97.87

0.45

99.55

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

7

Lucasfilm

5

Pixar
All

Defendants

Panel

230

116

25,429

25,545

C 2005 2009
Next

Separation

Company

Company

DNCC Defendant

Non

within 1 year

Percentage of

DNCCDefendant

Total

DNCC Defendant

Row

Total

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

8

661

669

1.20

98.80

Pixar

8

332

340

2.35

97.65

0.98

99.02

All

Defendants

Panel

350

35,508

35,858

D 2010 2012
Next

Separation

Company

Company
Non

within 1 year

Percentage of

Row

Total

DNCCDefendant

Total

DNCC Defendant

263

DNCC Defendant

269

2.23

97.77

177

3.95

96.05

2.08

97.92

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

6

Lucasfilm

7

Pixar
All

Defendants

170

246

11,577

11,823

Notes
This analysis

Adobe
Apple

excludes separations

had DNCC agreements with Apple Intel and Intuit

allegedly

Intel allegedly
Intuit allegedly

had DNCC agreements with Apple Google and Pixar
had DNCC agreements with Apple

Lucasfilm allegedly
Pixar

allegedly

Source

Dr

rehired by the

same

had DNCC agreements with Adobe Google Intel Intuit Lucasfilm

allegedly

Google

that appear as immediately

defendant

had a DNCC agreement with Apple

allegedly

had

and Google

had DNCC agreements with Apple

DNCC

agreements with Apple

Leamer’s employee

data

and Pixar

Intel

and Lucasfilm

and Pixar

company

within one year

Case5 11cv 02509 LHK Document5184

Filed1007 13

Page29

of

73

Appendix2A
Analysis of Hires from Other Defendants
Technical Creative and
Panel

Class

A 2001 2012
Last Previous

Company

Company

within 1 year

Adobe

Apple

Lucasfilm

0

5

0

0

0

Pixar

2

7

3

1

2

159

150

29

191

59

24

Last Previous

Company

Percentage

8

Hiring

RD

Google

Intel

Intuit

Lucasfilm

of

Row

Total

Other

Total

Adobe

Apple

Google

Intel

532

543

0.00

0.92

0.00

0.00

0.00

762

Pixar

785

0.25

0.89

0.38

0.13

0.25

1.02

53,747

0.30

0.28

0.05

0.36

0.11

0.04

Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

All

Defendants

Panel

6

25

0.05

B 2001 2004

Company

within 1 year

Adobe

Apple

Lucasfilm

0

0

0

0

0

Pixar

0

3

0

0

1

17

32

0

17

7

3

Last Previous

Company

Adobe

Apple

0

5

Percentage

1

Hiring

53,110

1.10

Google

Intel

Intuit

Lucasfilm

of

Row

Total

Total

Adobe

Apple

Google

Intel

56

57

0.00

0.00

0.00

0.00

0.00

234

Pixar

239

0.00

1.26

0.00

0.00

0.42

0.42

12,349

0.14

0.26

0.00

0.14

0.06

0.02

Other

Total

Adobe

Apple

Google

Intel

387

397

0.00

1.26

0.00

0.00

Other

Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

All

Defendants

Panel

Hiring

1

2

12,271

1.75
0.02

C 2005 2009

Company

Google

Intel

Intuit

within 1 year
Lucasfilm

Percentage
Pixar

of

Row

Total
Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

0

0

0

5

0

3

3

1

1

4

81

65

15

99

29

10

Last Previous

Company

Adobe

Apple

Google

0

0

Pixar
All

Defendants

Panel

Hiring

1.26

406

0.00

0.74

0.74

0.25

0.25

0.99

26,035

0.31

0.25

0.06

0.38

0.11

0.04

Other

Total

Adobe

Apple

Google

Intel

89

89

0.00

0.00

0.00

0.00

394
18

0.00

25,718

0.07

D 2010 2012

Company

Intel

Intuit

within 1 year
Lucasfilm

Percentage
Pixar

of

Row

Total
Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

Defendants

Note
Source

0

0

2

Pixar
All

0

0

This analysis

Dr

1

0

0

0

3

61

53

14

75

23

11

excludes hires indicated as acquisitions

Leamer’s employee

data

and

hires

showing the same defendant

134
5

15,121

company

as their

0.00

0.00

140

1.43

0.71

0.00

0.00

0.00

2.14

15,363

0.40

0.34

0.09

0.49

0.15

0.07

immediate previous employer within one year of the hiring

0.03

Case5 11cv 02509 LHK Document5184

Filed1007 13

Page30

of

73

Appendix 2B
Analysis of Separations Going to Other Defendants
Technical Creative and
Panel

RD

Class

A 2001 2012
Next

Separation

Company

Adobe

Apple

Google

Company

Intel

within 1 year

Intuit

Percentage of

Row

Total

Pixar

Other

Total

Adobe

Apple

Google

Intel

5

Lucasfilm

333

349

0.00

0.86

2.01

0.29

0.00

378

397

0.00

1.76

1.26

0.50

0.00

1.26

0.20

0.60

0.70

0.06

0.10

0.02

Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

0

3

7

1

0

Pixar

0

7

5

2

0

5

74

223

259

23

37

9

All

Defendants

Panel

18

36,356

36,999

1.43
0.05

B 2001 2004
Next

Separation

Company

Adobe

Apple

Google

Company

Intel

within 1 year

Intuit

Percentage of

Lucasfilm

Total

Total

Adobe

Apple

Google

Intel

7

7

0.00

0.00

0.00

0.00

0.00

106

Pixar

Other

Row

111

0.00

0.90

0.90

0.00

0.00

2.70

0.19

0.23

0.11

0.01

0.10

0.03

Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

0

0

0

0

0

Pixar

0

1

1

0

0

3

21

25

12

1

11

3

Adobe

Apple

Google

All

Defendants

Panel

0

0.00

3

11,001

Pixar

Other

Total

Adobe

Apple

Google

Intel

2

197

201

0.00

0.00

0.50

0.50

0.00

186

0.00

2.15

1.61

1.08

0.00

18,863

0.21

0.53

0.74

0.06

0.10

0.03

Pixar

Other

Total

Adobe

Apple

Google

Intel

3

129

141

0.00

2.13

4.26

0.00

0.00

100

0.00

2.00

1.00

0.00

0.00

0.00

0.18

1.43

1.55

0.15

0.09

0.00

0.03

1.08

9

11,077

C 2005 2009
Next

Separation

Company

Company

Intel

within 1 year

Intuit

Percentage of

Lucasfilm

Row

Total
Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

0

1

0

0

4

3

2

0

2

102

143

12

20

6

Adobe

Apple

Google

Defendants

Panel

1

41

Pixar
All

0

175

19,196

1.00
0.05

D 2010 2012
Next

Separation

Company

Company

Intel

within 1 year

Intuit

Percentage of

Lucasfilm

Row

Total
Intuit

Lucasfilm

Pixar

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

0

Source

0

0

0

Defendants

Note

6

This analysis

2

1

0

0

0

12

Pixar
All

3

96

104

10

6

0

excludes separations

Dr Leamer’s

employee

data

that appear as immediately

rehired by the

same defendant

97

6

company

6,492

within one year

6,726

2.13
0.09

Case5 11 cv 02509 LHK

Document518

4

Filed10 07 13

Page31

of

73

Appendix 2C
Analysis of Hires from Other DNCC Defendants
Technical Creative and
Panel

Class

A 2001 2012
Company

Last Previous

DNCC

Company

Hiring

RD

within 1 year

Percentage

of

Row

Total

DNCC Defendant

Total

DNCC Defendant

532

Non

Defendant

543

2.03

97.97

785

2.04

97.96

0.90

99.10

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

11

Lucasfilm

16

All

Defendants

Panel

769

482

Pixar

53,265

B 2001 2004
Company

Last Previous

DNCC

Company

Hiring

53,747

Non

Defendant

within 1 year

DNCC Defendant

Percentage
Total

of

DNCC Defendant

Row

Total

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

1

56

57

1.75

98.25

Pixar

4

235

239

1.67

98.33

61

12,288

0.49

99.51

All

Defendants

Panel

C 2005 2009
Company

Last Previous

DNCC

Company

Hiring

12,349

Non

Defendant

within 1 year

DNCC Defendant

Percentage
Total

of

DNCC Defendant

Row

Total

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

10

387

397

2.52

97.48

8

398

406

1.97

98.03

228

Lucasfilm

25,807

0.88

99.12

Pixar
All

Defendants

Panel

D 2010 2012
Company

Last Previous

DNCC

Company

Hiring

26,035

Defendant

Non

within 1 year

DNCC Defendant

Percentage

of

Row

Total

Non

DNCCDefendant

Total

DNCC Defendant

89

0.00

140

2.86

97.14

1.26

98.74

Adobe
Apple
Google
Intel
Intuit

0

Lucasfilm

4

136

193

15,170

Pixar
All

89

Defendants

15,363

100.00

Notes
This analysis

excludes

hires

indicated as acquisitions and hires showing the

same defendant company

one year of the hiring
Adobe

had a

allegedly

Intuit

allegedly

had

allegedly had

Source

Adobe Google Intel Intuit

DNCC agreements

with Apple

Google

and

Pixar

had DNCC agreements with Apple and Google

Lucasfilm allegedly
Pixar

with

had DNCC agreements with Apple Intel and Intuit

allegedly

Intel allegedly

agreement with Apple

DNCC agreements

Apple allegedly had
Google

DNCC

had DNCC agreements with Apple and Pixar

DNCC agreements

Dr Leamer’s

employee

data

with Apple

Intel and Lucasfilm

Lucasfilm

and Pixar

as their immediate previous

employer within

Case5 11 cv 02509 LHK

Document518

4

Filed10 07 13

Page32

of

73

Appendix 2D
Analysis of Separations Going to Other
Technical Creative and
Panel

RD

DNCC Defendants

Class

A 2001 2012
Next

Company

within 1 year

Percentage of

Row

Total

Total

DNCC Defendant

8

Company

DNCCDefendant

341

349

2.29

97.71

14

Separation

383

397

3.53

96.47

1.35

98.65

DNCC Defendant

Non

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

Lucasfilm
Pixar
All

Defendants

Panel

498

36,501

36,999

B 2001 2004
Next

Separation

Company

Company
Non

within 1 year

Percentage of

Row

Total

DNCCDefendant

Total

DNCC Defendant

7

DNCC Defendant

7

0.00

100.00

111

3.60

96.40

0.55

99.45

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

0

Lucasfilm

4

Pixar
All

Defendants

Panel

107

61

11,016

11,077

C 2005 2009
Next

Separation

Company

Company

DNCC Defendant

Non

within 1 year

Percentage of

DNCCDefendant

Total

DNCC Defendant

Row

Total

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

Lucasfilm

2

199

201

1.00

99.00

Pixar

8

178

186

4.30

95.70

1.29

98.71

All

Defendants

Panel

248

18,948

19,196

D 2010 2012
Next

Separation

Company

Company
Non

within 1 year

Percentage of

Row

Total

DNCCDefendant

Total

DNCC Defendant

135

DNCC Defendant

141

4.26

95.74

100

2.00

98.00

2.81

97.19

Non

DNCCDefendant

Adobe
Apple
Google
Intel
Intuit

6

Lucasfilm

2

Pixar
All

Defendants

98

189

6,537

6,726

Notes
This analysis

Adobe
Apple

excludes separations

had DNCC agreements with Apple Intel and Intuit

allegedly

Intel allegedly
Intuit allegedly

had DNCC agreements with Apple Google and Pixar
had DNCC agreements with Apple

Lucasfilm allegedly
Pixar

allegedly

Source

Dr

rehired by the

same

had DNCC agreements with Adobe Google Intel Intuit Lucasfilm

allegedly

Google

that appear as immediately

defendant

had a DNCC agreement with Apple

allegedly

had

and Google

had DNCC agreements with Apple

DNCC

agreements with Apple

Leamer’s employee

data

and Pixar

Intel

and Lucasfilm

and Pixar

company

within one year

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page33

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page34

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page35

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page36

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page37

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page38

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page39

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page40

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page41

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page42

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page43

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page44

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page45

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page46

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page47

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page48

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page49

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page50

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page51

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page52

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page53

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page54

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page55

of

73

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page56

of

73

Appendix 9A
Dr

Leamer’s Figure 20 RegressionIncluding Defendant

Defendant

Specific

Specific

Conduct Variables and Other

Interactive Effects

AllSalaried Employee Class
Dependant

Variable Log Total Annual Compensation CPI

Variable

ADOBE
APPLE

Conduct
Conduct

GOOGLE
INTEL

PIXAR

Age

Conduct

LUCASFILM

APPLE

ADOBE
APPLE

GOOGLE
INTEL

Conduct

INTUIT
PIXAR

ADOBE

Log Number
Log Number

Conduct

New

of
of

New

New

New

Log Number

of

Hires in the

New

Log Age

ADOBE
APPLE

Log Age

GOOGLE
INTEL
INTUIT
PIXAR

Log Age

Log Age
Log Age
Log Age

3.52

0.0001

0.19
22.24

0.0250

12.57

of

0.3453

0.0061

56.20

0.0323

0.0020

16.45

0.0213

0.0127

1.67

0.1142

0.0342

3.34

0.0664

0.0169

3.92

0.0976

19.15

0.0549

13.46

0.0380

6.84

0.0132

1.81

0.0576

2.46

0.1164

0.24

0.1636

1.48

0.0056

125.95

0.0027

272.85

0.0017

294.66

0.0023

286.66

0.7202

0.0059

121.40

0.6619

0.0056

117.60

0.8067

0.0360

22.42

0.2868

0.0055

52.13

0.2828

0.0028

102.17

0.3466

0.0017

207.40

0.2964

0.0023

129.91

0.2541

0.0057

44.21

0.1743

2

0.0376

0.3141

0.6721

1

0.8370

0.5121

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

0.0001

0.7265

Log Total Annual Compensation CPI

LUCASFILM

0.78

0.7079

Log Total Annual Compensation CPI

2
2
GOOGLE Log Total Annual Compensation CPI 2
INTEL Log Total Annual Compensation CPI
2
INTUIT
Log Total Annual Compensation CPI 2
PIXAR
Log Total Annual Compensation CPI 2
APPLE

6.83

0.0000

0.2427

Log Total Annual Compensation CPI

ADOBE

0.0000

0.0277

Conduct

Log Total Annual Compensation CPI

3.44

0.1416

Conduct
Conduct

LUCASFILM

0.0000

0.0240

1
1
GOOGLE Log Total Annual Compensation CPI 1
INTEL Log Total Annual Compensation CPI
1
INTUIT
Log Total Annual Compensation CPI 1
PIXAR
Log Total Annual Compensation CPI 1
APPLE

5.58

0.2602

Hires in the

Conduct

ADOBE

1.26

0.0000

0.7391

of

FirmNumber
FirmNumber

Conduct

LUCASFILM

0.0000

Employees

0.0053

32.60

0.1922

0.0365

5.26

0.4727

0.2194

2.15

1.0913

0.1256

8.69

1.0010

0.1547

6.47

0.2981

0.0485

6.15

0.8571

0.1696

5.05

0.0441

0.4413

0.10

of

of

FirmNumber

FirmNumber

Hires in the
Hires in the

1
1
Employees 1

Employees

0.37

1.8691

Hires in the

Conduct

INTUIT
PIXAR

of

of

Firm
Number
Firm
Number

Conduct

GOOGLE
INTEL

Log Number

Conduct
Conduct

LUCASFILM

APPLE

Log Number

Conduct

Hires in the
Hires in the

0.0074

0.0000

New

3.59

0.0002

New

0.0042

0.0000

of

0.75

0.0000

of

5.78

0.0024

0.0001

Log Number

0.0006

0.0001

Log Number

Conduct
Conduct

3.38

0.0000

Conduct

0.0020

0.0027

Conduct

5.34

0.0152
Age

1.79

0.0015

0.0018

Age

0.0026

0.0032

Age

Age 2
Age 2
GOOGLE Conduct
Age 2
INTEL Conduct
Age 2
INTUIT
Conduct
Age 2
PIXAR
Conduct
Age 2
LUCASFILM
Conduct
Age 2
ADOBE

T Value

0.0067

Age

Conduct

Error

0.0079

Age

Conduct

St

0.0047

Age

Conduct

Conduct

INTUIT

Estimate

1
Employees 1
Employees 1

Employees

of

of

Firm
Number

of

Employees

1

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page57

of

73

Appendix 9A
Log Age

LUCASFILM

0.0240

0.8306

0.03

0.0695

0.0297

2.34

0.1235

0.0170

7.24

0.1483

0.0214

6.92

0.0348

0.0066

5.30

0.1010

0.0229

4.41

0.0166

0.0605

0.27

0.0085

0.1115

0.08

Log Company Tenure Months

0.0167

0.0050

3.36

Log Company Tenure

0.0017

0.0005

3.14

0.0025

0.0005

4.62

in San Jose

1.5574

0.0183

85.30

Among Defendants

0.0770

0.0018

42.53

0.0025

0.0003

7.90

0.0441

0.0095

4.63

0.0461

0.0066

6.94

0.2261

0.0026

0.0049

0.0013

0.0808

0.0046

0.1603

0.0308

0.0217

0.0154

0.2292

0.0026

89.66

0.0915

0.0043

21.15

0.1646

0.0033

50.39

APPLE

3.3227

0.4646

7.15

GOOGLE

0.0066

0.4898

0.01

INTEL

1.6772

0.4130

4.06

INTUIT

2.9576

0.5094

5.81

PIXAR

1.3942

0.9009

1.55

LUCASFILM

0.9044

1.5907

0.57

Log Age

ADOBE

Log Age

APPLE

2

2

Log Age 2

GOOGLE

Log Age 2

INTEL

Log Age 2

INTUIT

Log Age 2

PIXAR

Log Age

LUCASFILM

2

2

Male

DLogInformation
Log Total Number
Year

Sector Employment
of Transfers

trend

ADOBE

Log Number
Log Number

APPLE

GOOGLE
INTEL

PIXAR

of

Log Number
Log Number

of

New

New

Hires in the

New

of

Log Firm Revenue

Per EmployeeCPI

DLogFirm

Revenue

State

Location

Per

of

of

FirmNumber
FirmNumber

Hires in the

Log Total Number

of

Firm
Number

FirmNumber

Hires in the
Hires in the

of

Firm
Number
Firm
Number

Hires in the

Hires in the

New

Log Number

LUCASFILM

Hires in the

New

New

of
of

New

New

of

Log Number

Log Number

INTUIT

of

1
Employees 1
Employees

of

1

1
1
Employees 1

Employees

of

of

Employees

Employees

Firm
Number

of

Employees

1

Hires

EmployeeCPI

1

1

YES

Indicators

YES

Constant

R Square

0.928

508,969

Observations

Note
Source

Significant

Dr

at

1

level

Significant

at

Leamer’s backup data and materials

5

level

Pixar

Significant

revenue

at

10 level

data after 2005 are included

86.41
3.77
17.61
5.20
1.41

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page58

of

73

Appendix 9B
Dr

Leamer’s Figure 23 RegressionIncluding Defendant

Defendant

Specific

RD

APPLE

Conduct
Conduct

GOOGLE
INTEL

PIXAR

Age

LUCASFILM

APPLE

APPLE

GOOGLE
INTEL

Conduct

INTUIT
PIXAR

ADOBE

Log Number
Log Number

Conduct

New

of
of

New

New

New

Log Number

of

Hires in the

New

Log Age

ADOBE
APPLE

Log Age

GOOGLE
INTEL
INTUIT
PIXAR

Log Age

Log Age
Log Age
Log Age

1.92

0.0002

0.41

0.0345

of

0.3276

0.0088

37.18

0.0388

0.0026

14.83

0.0750

0.0194

3.87

0.0642

0.0440

1.46

0.0820

0.0276

2.97

0.1241
0.0747

5.79

0.0494

4.21

0.0185

2.97

0.0875

2.14

0.1508

1.37

0.3662

0.56

0.0075

89.78

0.0037

192.60

0.0022

207.91

0.0029

219.78

0.6772

0.0088

76.81

0.6202

0.0084

73.65

0.7676

0.0695

11.04

0.3112

0.0074

42.05

0.2864

0.0038

74.62

0.3478

0.0021

162.51

0.3113

0.0028

109.66

0.2930

0.0085

34.49

0.0956

2

0.0482

0.1272

0.6429

1

0.9854

0.4607

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

0.0001

0.7040

Log Total Annual Compensation CPI

LUCASFILM

0.17

0.6754

Log Total Annual Compensation CPI

2
2
GOOGLE Log Total Annual Compensation CPI 2
INTEL Log Total Annual Compensation CPI
2
INTUIT
Log Total Annual Compensation CPI 2
PIXAR
Log Total Annual Compensation CPI 2
APPLE

5.07

0.0000

0.2062

Log Total Annual Compensation CPI

ADOBE

0.0000

0.2066

Conduct

Log Total Annual Compensation CPI

3.01

0.1868

Conduct
Conduct

LUCASFILM

0.0000

0.0548

1
1
GOOGLE Log Total Annual Compensation CPI 1
INTEL Log Total Annual Compensation CPI
1
INTUIT
Log Total Annual Compensation CPI 1
PIXAR
Log Total Annual Compensation CPI 1
APPLE

4.65

0.2078

Hires in the

Conduct

ADOBE

1.37

0.0000

0.4323

of

FirmNumber
FirmNumber

Conduct

LUCASFILM

0.0000

Employees

0.0076

12.61

0.2340

0.0702

3.34

0.3557

0.2812

1.26

1.2304

0.1670

7.37

0.1880

0.1917

0.98

0.3725

0.0699

5.33

1.0874

0.2520

4.31

0.6246

0.5776

1.08

of

of

FirmNumber

FirmNumber

Hires in the
Hires in the

1
1
Employees 1

Employees

0.20

2.2161

Hires in the

Conduct

INTUIT
PIXAR

of

of

Firm
Number
Firm
Number

Conduct

GOOGLE
INTEL

Log Number

Conduct
Conduct

LUCASFILM

APPLE

Log Number

Conduct

Hires in the

0.0182

0.0001

Hires in the

1.83

0.0001

New

0.0056

0.0000

New

0.29

0.0000

of

4.42

0.0037

0.0001

of

0.0008

0.0001

Log Number

2.93

0.0001

Log Number

Conduct
Conduct

0.0025

0.0036

Conduct

4.54

0.0102
Age

1.85

0.0020

0.0011

Age

0.0033

0.0035

Age

Conduct

ADOBE

T Value

0.0074

Age 2
Age 2
GOOGLE Conduct
Age 2
INTEL Conduct
Age 2
INTUIT
Conduct
Age 2
PIXAR
Conduct
Age 2
LUCASFILM
Conduct
Age 2
ADOBE

Error

0.0090

Age

Conduct

St

0.0062

Age

Conduct
Conduct

Estimate

Age

Conduct

Conduct

INTUIT

Class

Variable Log Total Annual Compensation CPI

Variable

ADOBE

Conduct Variables and Other

Interactive Effects

Technical Creative and
Dependant

Specific

1
Employees 1
Employees 1

Employees

of

of

Firm
Number

of

Employees

1

20.45
3.68

17.85

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page59

of

73

Appendix 9B
Log Age

LUCASFILM

0.4933

1.5449

0.32

0.0547

0.0381

1.43

0.1382

0.0228

6.07

0.0387

0.0265

1.46

0.0449

0.0095

4.73

0.1305

0.0342

3.82

0.0667

0.0793

0.84

0.0634

0.2101

0.30

Log Company Tenure Months

0.0021

0.0067

0.31

Log Company Tenure

0.0003

0.0007

0.47

0.0058

0.0008

7.21

in San Jose

1.6830

0.0250

67.20

Among Defendants

0.0854

0.0024

35.18

0.0004

0.0004

0.99

0.0497

0.0122

4.06

0.0349

0.0092

3.81

0.2318

0.0037

0.0041

0.0018

0.1109

0.0069

0.0495

0.0394

0.0296

0.0227

0.2643

0.0035

0.0435

0.0058

0.1532

0.0044

APPLE

3.4399

0.5998

5.73

GOOGLE

1.5131

0.6217

2.43

INTEL

1.6323

0.5322

3.07

INTUIT

3.2415

0.6919

4.68

PIXAR

0.8473

1.1715

0.72

LUCASFILM

1.4582

2.8740

0.51

Log Age

ADOBE

Log Age

APPLE

2

2

Log Age 2

GOOGLE

Log Age 2

INTEL

Log Age 2

INTUIT

Log Age 2

PIXAR

Log Age

LUCASFILM

2

2

Male

DLogInformation
Log Total Number
Year

Sector Employment
of Transfers

trend

ADOBE

Log Number
Log Number

APPLE

GOOGLE
INTEL

PIXAR

of

Log Number
Log Number

of

New

New

Hires in the

New

of

Log Firm Revenue

Per EmployeeCPI

DLogFirm

Revenue

State

Location

Per

of

of

FirmNumber
FirmNumber

Hires in the

Log Total Number

of

Firm
Number

FirmNumber

Hires in the
Hires in the

of

Firm
Number
Firm
Number

Hires in the

Hires in the

New

Log Number

LUCASFILM

Hires in the

New

New

of
of

New

New

of

Log Number

Log Number

INTUIT

of

1
Employees 1
Employees

of

1

1
1
Employees 1

Employees

of

of

Employees

Employees

Firm
Number

of

Employees

1

Hires

EmployeeCPI

1

1

YES

Indicators

YES

Constant

R Square

0.879

295,136

Observations

Note
Source

Significant

Dr

at

1

level

Significant

at

Leamer’s backup data and materials

5

level

Pixar

Significant

revenue

at

10 level

data after 2005 are included

63.00
2.34
16.17
1.26
1.31
76.33
7.45
35.02

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page60

of

73

Appendix 10A
Dr

Leamer’s Figure 20 RegressionUsing a Single Conduct Variable
AllSalaried Employee Class

Dependant

Log Total Annual Compensation CPI

Variable

Variable

Estimate

Conduct

St

Error

T Value

0.0344

0.0008

41.98

0.6978

0.0054

129.27

0.7416

0.0026

279.85

0.4943

0.0017

293.50

0.6687

0.0024

282.48

0.7117

0.0057

124.33

0.6961

0.0069

100.42

0.8118

0.0363

22.36

0.2934

0.0053

55.74

0.2595

0.0027

95.36

0.3734

0.0016

229.06

0.3005

0.0023

130.49

0.2522

0.0055

45.49

0.1992

0.0067

29.64

0.1798

0.0367

4.90

Log Age Years

0.0105

0.0328

0.32

Log Age

0.0076

0.0044

1.72

1
APPLE
Log Total Annual CompensationCPI 1
GOOGLE
Log Total Annual Compensation CPI 1
INTEL
Log Total Annual Compensation CPI 1
Log Total Annual Compensation CPI 1
INTUIT
PIXAR
Log Total Annual Compensation CPI 1
ADOBE

Log Total Annual CompensationCPI

LogTotal

LUCASFILM

1

Annual CompensationCPI

2
APPLE
Log Total Annual CompensationCPI 2
GOOGLE
Log Total Annual Compensation CPI 2
INTEL
Log Total Annual Compensation CPI 2
INTUIT
Log Total Annual Compensation CPI 2
PIXAR
Log Total Annual Compensation CPI 2
ADOBE

Log Total Annual CompensationCPI

LogTotal

LUCASFILM

Annual CompensationCPI

2

2

Log Company Tenure

Months

0.0083

0.0050

1.66

Log Company Tenure

2

0.0009

0.0006

1.66

0.0027

0.0005

5.02

San Jose

1.4135

0.0136

103.90

Among Defendants

0.0959

0.0015

63.66

0.0039

0.0003

14.53

0.0169

0.0008

21.61

0.2478

0.0021

116.78

0.1027

0.0034

30.20

0.2162

0.0033

66.49

APPLE

0.0607

0.0162

GOOGLE

1.0320

0.0174

59.42

INTEL

0.1516

0.0146

10.40

INTUIT

0.1473

0.0193

PIXAR

0.7075

0.0422

0.1256

0.0480

Male

DLog

Information Sector Employment

Log Total Number
Year

of

Transfers

trend

Log Number

of

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLog

in

Firm Revenue

Firm
Number

Employees

1

Hires

EmployeeCPI

Per

of

EmployeeCPI

1
1

LUCASFILM
Location

State Indicators

YES

Constant

YES

RSquare

0.926
504,897

Observations

Note
Source

Significant

Dr Leamer’s

at

1

level

Significant

at

backup data and materials

5

level

Significant

at

10 level

3.75

7.64
16.77
2.61

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page61

of

73

Appendix 10B

Dr

Leamer’s Figure 23 RegressionUsing a Single Conduct Variable
Technical Creative and

Dependant

RD

Class

Log Total Annual Compensation CPI

Variable

Variable

Estimate

Conduct

St

Error

T Value

0.0234

0.0011

20.94

0.6643

0.0072

91.76

0.7212

0.0037

197.36

0.4403

0.0022

203.78

0.6407

0.0030

215.53

0.6578

0.0084

78.28

0.6523

0.0106

61.69

0.8457

0.0692

12.21

0.3158

0.0071

44.58

0.2581

0.0038

68.54

0.3629

0.0021

173.68

0.3171

0.0029

110.18

0.2967

0.0081

36.48

0.1054

0.0097

10.89

0.1456

0.0694

2.10

Log Age Years

0.1807

0.0463

3.90

Log Age

0.0146

0.0063

2.32

1
APPLE
Log Total Annual CompensationCPI 1
GOOGLE
Log Total Annual Compensation CPI 1
INTEL
Log Total Annual Compensation CPI 1
Log Total Annual Compensation CPI 1
INTUIT
PIXAR
Log Total Annual Compensation CPI 1
ADOBE

Log Total Annual CompensationCPI

LogTotal

LUCASFILM

1

Annual CompensationCPI

2
APPLE
Log Total Annual CompensationCPI 2
GOOGLE
Log Total Annual Compensation CPI 2
INTEL
Log Total Annual Compensation CPI 2
INTUIT
Log Total Annual Compensation CPI 2
PIXAR
Log Total Annual Compensation CPI 2
ADOBE

Log Total Annual CompensationCPI

LogTotal

LUCASFILM

Annual CompensationCPI

2

2

Log Company Tenure

Months

0.0326

0.0068

4.78

Log Company Tenure

2

0.0028

0.0008

3.78

0.0065

0.0008

7.89

San Jose

1.5271

0.0189

80.81

Among Defendants

0.0983

0.0020

48.08

0.0009

0.0004

0.0154

0.0011

14.31

0.2724

0.0029

93.07

0.0811

0.0047

17.17

0.2127

0.0044

48.43

APPLE

0.1244

0.0245

GOOGLE

1.3816

0.0259

INTEL

0.1573

0.0219

INTUIT

0.1486

0.0315

PIXAR

1.5543

0.0771

0.0296

0.1038

Male

DLog

Information Sector Employment

Log Total Number
Year

of

Transfers

trend

Log Number

of

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLog

in

Firm Revenue

Firm
Number

Employees

1

Hires

EmployeeCPI

Per

of

EmployeeCPI

1
1

LUCASFILM
Location

State Indicators

YES

Constant

YES

RSquare

0.874
292,489

Observations

Note
Source

Significant

Dr Leamer’s

at

1

level

Significant

at

backup data and materials

5

level

Significant

at

10 level

2.52

5.08
53.33
7.19
4.71
20.17
0.29

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page62

of

73

Appendix 10C
Undercompensation Estimates Using a
Conduct Variable

in

Dr Leamer’s

AllSalaried

Year

Adobe

Apple

Google

2005

1.72

1.72

1.72

2006

4.63

4.71

2007

7.17

2008

9.80

2009

9.80

10.28

vs

Regression

Figures

Employee Class

Apple

Google

1.61

1.59

1.78

1.67

12.13

10.56

2006

4.28

4.43

4.44

4.70

14.63

12.44

14.00

2007

6.64

6.94

6.39

7.46

3.24

17.24

14.28

15.61

2008

9.08

9.56

8.40

10.05

5.64

19.94

15.76

14.52

2009

9.15

9.73

7.51

9.95

5.70

20.12

14.65

Pixar

Year

1.72

11.95

10.29

2005

4.28

4.58

14.77

12.23

7.37

6.19

7.02

3.44

17.58

10.13

8.10

9.51

5.88

20.36

7.17

9.32

5.91

20.55

Intuit

RD

Adobe

Apple

Google

2005

1.17

1.17

1.17

1.17

2006

3.12

3.19

2.86

3.09

2007

4.78

4.94

4.03

4.69

2.34

2008

6.50

6.73

5.15

6.33

3.88

2009

6.42

6.71

4.31

6.13

3.83

Intel

with age and hiring rate

Intel

Technical Creative and

Lucasfilm

Class

1.90

3.07

1.64

2006

4.29

4.96

7.23

3.06

14.77

10.47

2007

6.48

7.79

9.36

3.38

3.41

18.08

10.61

7.92

2008

8.80

10.64

11.20

4.76

5.21

20.44

11.87

6.54

2009

8.44

10.51

9.00

4.19

4.96

20.54

9.62

2005

6.85

12.27

7.45

14.22
14.40

interactions

Pixar

1.56

6.08

8.33

RD

Lucasfilm

Google

Year

Adobe

Intuit

Apple

Pixar

10.31

Intuit

Leamer Figure 20 and 23 regressions excluding conduct

Adobe

Class

Year

22 and 24

AllSalaried Employee Class
Lucasfilm

Intel

Technical Creative and

Source

Undercompensation Estimates in Dr Leamer’s

Single

Intel

Intuit

Lucasfilm

10.80

Pixar

9.28

Case5 11 cv 02509 LHK

Document518

4

Filed10 07 13

Page63

of

73

Appendix 11A
Dr

Leamer’s Figure 20 RegressionIncluding Defendant

Specific

Conduct

Variables

AllSalaried Employee Class
Variable Log Total Annual Compensation CPI

Dependant

Variable

ADOBE

Estimate

Conduct

St

Error

T Value

0.0053

0.0028

1.89

0.0139

0.0019

7.37

0.0969

0.0021

45.25

0.0304

0.0009

33.37

0.0600

0.0026

23.17

0.0396

0.0048

8.34

0.0000

0.0075

0.00

0.6855

0.0056

122.85

0.7361

0.0027

276.84

0.4858

0.0017

283.31

0.6721

0.0024

283.28

0.7173

0.0058

122.92

0.6857

0.0055

124.10

0.7984

0.0364

21.92

Log Total Annual Compensation CPI

0.3056

0.0055

56.03

Log Total Annual Compensation CPI

0.2645

0.0027

96.26

0.3741

0.0016

228.53

0.2976

0.0023

128.96

0.2466

0.0056

43.72

0.1758

0.0053

33.30

0.2003

0.0369

5.43

Log Age Years

0.0244

0.0327

0.75

Log Age

APPLE

Conduct

GOOGLE

Conduct
Conduct

INTEL

Conduct

INTUIT

PIXAR

Conduct

LUCASFILM

Conduct

1
APPLE
Log Total Annual Compensation CPI 1
GOOGLE
Log Total Annual Compensation CPI 1
INTEL
Log Total Annual Compensation CPI 1
INTUIT
Log Total Annual Compensation CPI 1
PIXAR
Log Total Annual Compensation CPI 1
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

1

2
2
GOOGLE
Log Total Annual Compensation CPI 2
INTEL
Log Total Annual Compensation CPI 2
INTUIT
Log Total Annual Compensation CPI 2
PIXAR
Log Total Annual Compensation CPI 2
ADOBE
APPLE

Log Total Annual Compensation CPI

LUCASFILM

2

0.0057

0.0044

1.28

Log Company Tenure Months

0.0128

0.0050

2.55

Log Company Tenure 2

0.0013

0.0006

2.42

0.0032

0.0005

1.4228

0.0136

104.42

0.0800

0.0015

53.90

0.0032

0.0003

12.13

0.0128

0.0008

16.20

0.2273

0.0021

108.21

0.0677

0.0033

20.55

0.1461

0.0029

50.95

APPLE

0.0492

0.0163

GOOGLE

1.0950

0.0176

62.24

INTEL

0.1587

0.0147

10.82

INTUIT

0.1818

0.0193

PIXAR

0.7905

0.0264

LUCASFILM

0.0271

0.0503

2

Male

DLog

Log Total Number
Year

of Transfers

Among Defendants

trend

Log Number

of

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLog

in San Jose

Information Sector Employment

Firm Revenue

State

Location

Firm
Number

Employees

1

Hires

EmployeeCPI

Per

of

EmployeeCPI

1
1

YES

Indicators

Constant

YES

R Square

0.926

Observations

Note
Source

Significant

Dr Leamer’s

508,969

at

1

level

Significant

backup data and materials

at

5
Pixar

level

Significant

revenue

at

10 level

data after 2005 are included

5.82

3.02

9.40
29.96
0.54

Case5 11 cv 02509 LHK

Document518

4

Filed10 07 13

Page64

of

73

Appendix 11B
Dr

Leamer’s Figure 23 RegressionIncluding Defendant

Specific

Conduct

Variables

Technical Creative and

RD

Variable Log Total Annual Compensation CPI

Dependant

Variable

ADOBE

Class

Estimate

Conduct

St

Error

T Value

0.0175

0.0036

4.80

0.0227

0.0026

8.71

0.1219

0.0029

42.51

0.0124

0.0012

10.12

0.0512

0.0040

12.96

0.0800

0.0061

13.10

0.0204

0.0130

0.6517

0.0075

86.93

0.7204

0.0036

197.54

0.4279

0.0022

195.45

0.6449

0.0030

217.17

0.6682

0.0086

77.99

0.6623

0.0081

81.28

0.7861

0.0701

11.21

Log Total Annual Compensation CPI

0.3285

0.0074

44.62

Log Total Annual Compensation CPI

0.2566

0.0038

67.66

0.3684

0.0021

175.48

0.3140

0.0029

109.24

0.2870

0.0083

34.76

0.1014

0.0075

13.58

0.2148

0.0707

3.04

Log Age Years

0.2111

0.0461

4.58

Log Age

APPLE

Conduct

GOOGLE

Conduct
Conduct

INTEL

Conduct

INTUIT

PIXAR

Conduct

LUCASFILM

Conduct

1
APPLE
Log Total Annual Compensation CPI 1
GOOGLE
Log Total Annual Compensation CPI 1
INTEL
Log Total Annual Compensation CPI 1
INTUIT
Log Total Annual Compensation CPI 1
PIXAR
Log Total Annual Compensation CPI 1
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

1

2
2
GOOGLE
Log Total Annual Compensation CPI 2
INTEL
Log Total Annual Compensation CPI 2
INTUIT
Log Total Annual Compensation CPI 2
PIXAR
Log Total Annual Compensation CPI 2
ADOBE
APPLE

Log Total Annual Compensation CPI

LUCASFILM

2

1.57

0.0187

0.0063

2.99

Log Company Tenure Months

0.0011

0.0068

0.16

Log Company Tenure 2

0.0005

0.0008

0.73

0.0067

0.0008

1.5258

0.0189

80.88

0.0805

0.0020

40.21

0.0000

0.0004

0.0145

0.0011

13.40

0.2548

0.0029

88.38

0.0402

0.0045

0.1324

0.0038

APPLE

0.1309

0.0246

GOOGLE

1.4469

0.0261

INTEL

0.1653

0.0220

INTUIT

0.1840

0.0315

PIXAR

1.3668

0.0455

LUCASFILM

0.0872

0.1064

2

Male

DLog

Log Total Number
Year

of Transfers

Among Defendants

trend

Log Number

of

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLog

in San Jose

Information Sector Employment

Firm Revenue

State

Location

Firm
Number

Employees

1

Hires

EmployeeCPI

Per

of

EmployeeCPI

1
1

YES

Indicators

Constant

YES

R Square

0.874

Observations

Note
Source

Significant

Dr Leamer’s

295,136

at

1

level

Significant

backup data and materials

at

5
Pixar

level

Significant

revenue

at

10 level

data after 2005 are included

8.24

0.08

8.91
34.60
5.32
55.52
7.53
5.83
30.03
0.82

Case5 11cv 02509LHK

Document518

4

Filed1007 13

Page65

of

73

Appendix 11C
Undercompensation

Estimates Using

Conduct Variables

in

AllSalaried

Year

Adobe

Apple

Google

2005

0.26

0.69

4.85

2006

0.71

1.90

2007

1.09

2008

1.49

2009

1.49

Dr

Defendant

vs

Leamer’s Regression

Figures

Employee Class

Apple

Google

1.61

1.59

1.78

1.67

12.13

10.56

2006

4.28

4.43

4.44

4.70

14.63

12.44

15.21

2007

6.64

6.94

6.39

7.46

3.24

17.24

14.28

16.76

2008

9.08

9.56

8.40

10.05

5.64

19.94

15.76

2009

9.15

9.73

7.51

9.95

5.70

20.12

14.65

Pixar

Year

1.52

0.01

11.48

2005

12.04

4.06

0.01

13.46

2.97

17.35

6.23

6.00

0.01

4.08

22.63

8.44

10.30

0.02

4.13

19.91

8.28

10.36

0.02

15.16

Intuit

RD

Adobe

Apple

Google

2005

0.87

1.13

6.09

0.62

2006

2.32

3.08

14.79

1.64

2007

3.55

4.78

20.76

2.50

5.12

2008

4.82

6.50

26.52

3.37

8.55

2009

4.74

6.47

22.04

3.27

8.46

Intel

hiring

rate and including company conduct
data after 2005 are included

3.07

1.64

2006

4.29

4.96

7.23

3.06

14.77

10.47

2007

6.48

7.79

9.36

3.38

3.41

18.08

10.61

27.50

2008

8.80

10.64

11.20

4.76

5.21

20.44

11.87

22.83

2009

8.44

10.51

9.00

4.19

4.96

20.54

9.62

8.71

age and

interactions

Class

1.90

2005

23.69

10.39

25.82

12.08
12.24

with

Pixar

1.56

21.01

interactions

RD

Lucasfilm

Google

Year

Adobe

Intuit

Apple

Pixar

7.02

Leamer Figure 20 and 23 regressions excluding conduct

Intel

Technical Creative and

Lucasfilm

Intuit

Source

Adobe

Class

Year

22 and 24

AllSalaried Employee Class
Lucasfilm

Intel

Technical Creative and

Pixar revenue

Undercompensation Estimates in Dr Leamer’s

Specific

Intel

Intuit

Lucasfilm

10.80

Pixar

9.28

Case5 11 cv 02509 LHK

Document518

Filed10 07 13

4

Page66

of

73

Appendix 12A
Dr

Leamer’s Figure 20 RegressionUsing

Employee Class

AllSalaried

Dependant

Variable

PreConduct Periodas Benchmark

Log Total Annual Compensation CPI

Variable

Estimate

St

Error

T Value

Conduct

Age

0.0056

0.0005

10.83

Conduct

Age 2

0.0001

0.0000

11.78

Conduct

Log Number

0.0391

0.0010

40.01

0.2432

0.0111

21.97

0.7667

0.0062

122.75

0.7374

0.0033

223.86

0.5619

0.0023

245.29

0.6743

0.0026

263.51

0.7086

0.0062

114.53

0.6957

0.0056

123.46

0.7392

0.0390

18.95

0.2167

0.0061

35.43

0.2637

0.0034

77.79

0.3504

0.0020

178.13

0.2932

0.0025

118.61

0.2459

0.0059

41.50

0.1477

0.0054

27.16

0.2434

0.0395

6.16

Log Age Years

0.4166

0.0537

7.75

Log Age 2

0.0498

0.0073

6.79

Log Company Tenure Months

0.0684

0.0057

12.04

Log Company Tenure

0.0068

0.0006

10.87

0.0030

0.0006

4.83

San Jose

1.2592

0.0166

75.70

Among Defendants

0.0789

0.0018

42.98

0.0105

0.0003

29.97

0.0197

0.0010

19.03

0.2174

0.0030

71.92

0.0928

0.0045

20.50

of

New

Hires In the

Firm
Number

of

Employees

1

Conduct

1
APPLE
Log Total Annual Compensation CPI 1
GOOGLE
Log Total Annual Compensation CPI 1
INTEL
Log Total Annual Compensation CPI 1
INTUIT
Log Total Annual Compensation CPI 1
PIXAR
Log Total Annual Compensation CPI 1
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

1

2
APPLE
Log Total Annual Compensation CPI 2
GOOGLE
Log Total Annual Compensation CPI 2
INTEL
Log Total Annual Compensation CPI 2
INTUIT
Log Total Annual Compensation CPI 2
PIXAR
Log Total Annual Compensation CPI 2
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

2

2

Male

DLog

Information Sector Employment

Log Total Number
Year

of

Transfers

in

trend

Log Number

of

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLog Firm Revenue

Firm
Number

of

Employees

1

Hires

1
EmployeeCPI 1

EmployeeCPI

0.1286

0.0033

38.95

APPLE

0.1111

0.0194

5.71

GOOGLE

0.6086

0.0217

28.00

INTEL

0.1019

0.0173

5.89

INTUIT

0.2270

0.0223

10.17

PIXAR

0.9625

0.0302

31.82

LUCASFILM

0.1298

0.0626

2.07

State

Location

Per

YES

Indicators

Constant

YES

RSquare

0.924

Observations

Note
Source

Significant

Dr Leamer’s

381,288

at

1

level

Significant

at

backup data and materials

5

level

Pixar

Significant

revenue

at

10 level

data after 2005 are included

Case5 11 cv 02509 LHK

Document518

Filed10 07 13

4

Page67

of

73

Appendix 12B
Dr

Leamer’s Figure 23 RegressionUsing

PreConduct Periodas Benchmark

Technical Creative and
Dependant

Variable

RD

Class

Log Total Annual Compensation CPI

Variable

Estimate

St

Error

T Value

Conduct

Age

0.0061

0.0008

8.05

Conduct

Age 2

0.0001

0.0000

8.90

Conduct

Log Number

0.0546

0.0013

40.90

0.2967

0.0159

18.61

0.7426

0.0083

89.58

0.7137

0.0047

151.39

0.4868

0.0031

157.85

0.6285

0.0032

195.11

0.6641

0.0093

71.55

0.6794

0.0084

81.00

0.6826

0.0827

8.25

0.2307

0.0081

28.45

0.2675

0.0049

54.82

0.3341

0.0026

129.27

0.3232

0.0031

104.05

0.2842

0.0088

32.11

0.0644

0.0078

8.27

0.2566

0.0822

3.12

Log Age Years

0.5769

0.0798

7.23

Log Age 2

0.0720

0.0109

6.59

Log Company Tenure Months

0.0994

0.0079

12.64

Log Company Tenure

0.0093

0.0009

10.65

0.0065

0.0009

6.89

San Jose

1.1685

0.0234

49.89

Among Defendants

0.0782

0.0025

30.91

0.0042

0.0005

8.83

0.0239

0.0014

16.49

0.2084

0.0043

48.83

0.1131

0.0062

18.39

of

New

Hires In the

Firm
Number

of

Employees

1

Conduct

1
APPLE
Log Total Annual Compensation CPI 1
GOOGLE
Log Total Annual Compensation CPI 1
INTEL
Log Total Annual Compensation CPI 1
INTUIT
Log Total Annual Compensation CPI 1
PIXAR
Log Total Annual Compensation CPI 1
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

1

2
APPLE
Log Total Annual Compensation CPI 2
GOOGLE
Log Total Annual Compensation CPI 2
INTEL
Log Total Annual Compensation CPI 2
INTUIT
Log Total Annual Compensation CPI 2
PIXAR
Log Total Annual Compensation CPI 2
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

2

2

Male

DLog

Information Sector Employment

Log Total Number
Year

of

Transfers

in

trend

Log Number

of

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLog Firm Revenue

Firm
Number

of

Employees

1

Hires

1
EmployeeCPI 1

EmployeeCPI

0.1164

0.0044

26.21

APPLE

0.0573

0.0292

1.96

GOOGLE

1.1501

0.0330

34.87

INTEL

0.1375

0.0256

5.38

INTUIT

0.2064

0.0364

5.67

PIXAR

1.5840

0.0521

30.41

LUCASFILM

0.0853

0.1652

0.52

State

Location

Per

YES

Indicators

Constant

YES

RSquare

0.866

Observations

Note
Source

Significant

Dr Leamer’s

216,253

at

1

level

Significant

at

backup data and materials

5

level

Pixar

Significant

revenue

at

10 level

data after 2005 are included

Case5 11 cv 02509 LHK

Document518

Filed10 07 13

4

Page68

of

73

Appendix 12C
Dr

Leamer’s Figure 20 RegressionUsing Post Conduct Period as Benchmark

Employee Class

AllSalaried

Dependant

Variable

Log Total Annual Compensation CPI

Variable

Estimate

St

Error

T Value

Conduct

Age

0.0078

0.0006

13.85

Conduct

Age 2

0.0001

0.0000

13.31

Conduct

Log Number

0.0114

0.0009

12.67

0.0973

0.0121

0.7630

0.0069

110.30

0.7349

0.0029

250.23

0.5002

0.0018

277.95

0.6763

0.0034

200.70

0.8207

0.0103

79.39

0.7036

0.0058

122.35

0.8750

0.0378

23.12

0.2528

0.0070

36.11

0.2602

0.0031

85.08

0.3684

0.0017

213.20

0.3235

0.0034

95.84

0.1548

0.0104

14.95

0.1769

0.0055

32.24

0.1143

0.0382

2.99

Log Age Years

0.6760

0.0560

12.08

Log Age 2

0.0797

0.0076

10.55

Log Company Tenure Months

0.0254

0.0058

4.39

Log Company Tenure

0.0020

0.0006

3.21

0.0021

0.0006

3.34

San Jose

0.8493

0.0541

15.70

Among Defendants

0.0287

0.0019

15.14

0.0113

0.0005

23.30

0.0325

0.0012

26.15

0.0683

0.0059

11.64

0.0268

0.0040

6.61

of

New

Hires In the

Firm
Number

of

Employees

1

Conduct

1
APPLE
Log Total Annual Compensation CPI 1
GOOGLE
Log Total Annual Compensation CPI 1
INTEL
Log Total Annual Compensation CPI 1
INTUIT
Log Total Annual Compensation CPI 1
PIXAR
Log Total Annual Compensation CPI 1
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

1

2
APPLE
Log Total Annual Compensation CPI 2
GOOGLE
Log Total Annual Compensation CPI 2
INTEL
Log Total Annual Compensation CPI 2
INTUIT
Log Total Annual Compensation CPI 2
PIXAR
Log Total Annual Compensation CPI 2
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

2

2

Male

DLog

Information Sector Employment

Log Total Number
Year

of

Transfers

in

trend

Log Number

of

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLog Firm Revenue

Firm
Number

of

Employees

1

Hires

1
EmployeeCPI 1

EmployeeCPI

8.06

0.1248

0.0032

39.43

APPLE

0.2203

0.0187

11.80

GOOGLE

1.1437

0.0196

58.31

INTEL

0.0757

0.0169

4.47

INTUIT

0.2278

0.0247

9.23

PIXAR

0.8522

0.0283

30.13

LUCASFILM

0.1705

0.0507

3.36

State

Location

Per

YES

Indicators

Constant

YES

RSquare

0.922

Observations

Note
Source

Significant

Dr Leamer’s

399,299

at

1

level

Significant

at

backup data and materials

5

level

Pixar

Significant

revenue

at

10 level

data after 2005 are included

Case5 11 cv 02509 LHK

Document518

Filed10 07 13

4

Page69

of

73

Appendix 12D
Dr

Leamer’s Figure 23 RegressionUsing Post Conduct Period as Benchmark
Technical Creative and

Dependant

Variable

RD

Class

Log Total Annual Compensation CPI

Variable

Estimate

St

Error

T Value

Conduct

Age

0.0096

0.0008

12.31

Conduct

Age 2

0.0001

0.0000

11.96

Conduct

Log Number

0.0008

0.0012

0.70

0.1544

0.0165

9.37

0.7523

0.0092

81.89

0.7161

0.0039

181.32

0.4438

0.0023

193.37

0.6464

0.0041

156.05

0.7732

0.0151

51.22

0.7071

0.0085

83.39

0.9511

0.0719

13.24

0.2530

0.0094

26.98

0.2581

0.0041

62.57

0.3655

0.0022

165.61

0.3478

0.0041

84.01

0.1837

0.0151

12.18

0.1052

0.0078

13.57

0.0413

0.0720

0.57

Log Age Years

0.9447

0.0755

12.51

Log Age 2

0.1145

0.0102

11.21

Log Company Tenure Months

0.0094

0.0078

1.21

Log Company Tenure

0.0008

0.0009

0.98

0.0065

0.0009

6.91

San Jose

0.9430

0.0718

Among Defendants

0.0088

0.0026

3.41

0.0148

0.0006

22.84

0.0367

0.0017

21.93

0.0834

0.0078

10.64

0.0112

0.0054

2.05

of

New

Hires In the

Firm
Number

of

Employees

1

Conduct

1
APPLE
Log Total Annual Compensation CPI 1
GOOGLE
Log Total Annual Compensation CPI 1
INTEL
Log Total Annual Compensation CPI 1
INTUIT
Log Total Annual Compensation CPI 1
PIXAR
Log Total Annual Compensation CPI 1
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

1

2
APPLE
Log Total Annual Compensation CPI 2
GOOGLE
Log Total Annual Compensation CPI 2
INTEL
Log Total Annual Compensation CPI 2
INTUIT
Log Total Annual Compensation CPI 2
PIXAR
Log Total Annual Compensation CPI 2
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

2

2

Male

DLog

Information Sector Employment

Log Total Number
Year

of

Transfers

in

trend

Log Number

of

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLog Firm Revenue

Firm
Number

of

Employees

1

Hires

1
EmployeeCPI 1

EmployeeCPI

13.14

0.1110

0.0042

26.40

APPLE

0.2949

0.0283

10.42

GOOGLE

1.4735

0.0292

50.43

INTEL

0.0390

0.0255

1.53

INTUIT

0.2932

0.0406

7.21

PIXAR

1.2492

0.0487

25.67

LUCASFILM

0.0692

0.1083

0.64

State

Location

Per

YES

Indicators

Constant

YES

RSquare

0.869

Observations

Note
Source

Significant

Dr Leamer’s

236,748

at

1

level

Significant

at

backup data and materials

5

level

Pixar

Significant

revenue

at

10 level

data after 2005 are included

Case5 11cv 02509LHK

Document518

Filed1007 13

4

Page70

of

73

Appendix 13A
Dr

Leamer’s Figure 20 RegressionEstimated Using

Variable

Log Total Annual Compensation CPI

Variable

ADOBE

Period Data

Employee Class

AllSalaried

Dependant

Non Conduct

Estimate

1
1
CPI
1

Log Total Annual Compensation CPI

St

Error

T Value

0.6108

0.0072

84.47

0.7408

0.0036

205.55

0.4578

0.0026

175.14

0.6685

0.0034

196.94

0.7266

0.0063

115.16

0.8377

0.0219

38.18

0.9990

0.0845

11.82

0.3441

0.0067

51.72

0.2708

0.0036

74.65

0.3957

0.0028

141.55

0.2620

0.0032

81.66

0.2413

0.0060

40.26

0.1329

0.0201

6.60

0.0161

0.0856

0.19

Log Age Years

0.0292

0.0436

0.67

Log Age 2

0.0122

0.0059

2.07

Log Company Tenure Months

0.0613

0.0071

8.59

Log Company Tenure

0.0064

0.0008

8.21

Log Total Annual Compensation CPI

APPLE

GOOGLE

Log Total Annual Compensation

Log Total Annual Compensation
CPI

INTEL

Log Total Annual Compensation CPI

INTUIT

Log Total Annual CompensationCPI

PIXAR

1

1
1

Log Total Annual Compensation CPI

LUCASFILM

1

2
APPLE
Log Total Annual Compensation CPI 2
GOOGLE Log Total Annual Compensation CPI 2
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation
CPI

INTEL

Log Total Annual Compensation CPI

INTUIT

Log Total Annual CompensationCPI

PIXAR

2

2
2

Log Total Annual Compensation CPI

LUCASFILM

2

2

Male

0.0041

0.0007

5.58

San Jose

1.3739

0.0252

54.58

Among Defendants

0.0610

0.0027

22.79

0.0028

0.0007

3.93

0.0365

0.0013

27.33

0.2303

0.0053

43.47

0.0961

0.0048

19.94

0.0715

0.0062

11.50

APPLE

0.2454

0.0216

11.37

GOOGLE

0.8453

0.0233

36.31

INTEL

0.1981

0.0195

10.18

INTUIT

0.0736

0.0242

3.04

PIXAR

0.0559

0.0473

1.18

0.2748

0.0708

3.88

DLogInformation
Log Total Number
Year

Sector Employment
of Transfers

in

trend

Log Number

of

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLogFirm

Revenue

Firm
Number

EmployeeCPI

Per

of

Employees

1

Hires

EmployeeCPI

1
1

LUCASFILM
State Indicators

Location

YES
YES

Constant

R Square

0.937
237,351

Observations

Note
Source

Significant

Dr

at

1

level

Significant

Leamer’s backup data and materials

at

5

level

Pixar

Significant

revenue data

at

after

10 level
2005 are included

Case5 11cv 02509LHK

Document518

Filed1007 13

4

Page71

of

73

Appendix 13B
Dr

Leamer’s Figure 23 RegressionEstimated Using
Technical Creative and

Dependant

Variable

RD

Non Conduct

Class

Log Total Annual Compensation CPI

Variable

ADOBE

Period Data

Estimate

1
1
CPI
1

Log Total Annual Compensation CPI

St

Error

T Value

0.5929

0.0100

59.23

0.7428

0.0049

151.07

0.4205

0.0033

129.36

0.6526

0.0043

153.41

0.7101

0.0092

76.79

0.9381

0.0359

26.12

0.9713

0.1224

7.94

0.3475

0.0092

37.69

0.2392

0.0050

48.28

0.3895

0.0036

108.96

0.2660

0.0040

66.55

0.2593

0.0087

29.69

0.0343

0.0307

1.12

0.0629

0.1247

0.50

Log Age Years

0.2740

0.0614

4.46

Log Age 2

0.0282

0.0083

3.38

Log Company Tenure Months

0.0758

0.0096

7.89

Log Company Tenure

0.0086

0.0011

8.09

Log Total Annual Compensation CPI

APPLE

GOOGLE

Log Total Annual Compensation

Log Total Annual Compensation
CPI

INTEL

Log Total Annual Compensation CPI

INTUIT

Log Total Annual CompensationCPI

PIXAR

1

1
1

Log Total Annual Compensation CPI

LUCASFILM

1

2
APPLE
Log Total Annual Compensation CPI 2
GOOGLE Log Total Annual Compensation CPI 2
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation
CPI

INTEL

Log Total Annual Compensation CPI

INTUIT

Log Total Annual CompensationCPI

PIXAR

2

2
2

Log Total Annual Compensation CPI

LUCASFILM

2

2

Male

0.0071

0.0011

6.43

San Jose

1.3635

0.0362

37.70

Among Defendants

0.0650

0.0038

17.33

0.0034

0.0011

3.16

0.0495

0.0018

26.92

0.2480

0.0078

31.98

0.0458

0.0067

6.82

0.0388

0.0086

4.51

APPLE

0.1750

0.0326

5.37

GOOGLE

0.9977

0.0343

29.13

INTEL

0.2041

0.0293

6.96

INTUIT

0.1603

0.0388

4.13

PIXAR

0.1585

0.0893

1.77

0.5484

0.1265

4.34

DLogInformation
Log Total Number
Year

Sector Employment
of Transfers

in

trend

Log Number

of

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLogFirm

Revenue

Firm
Number

EmployeeCPI

Per

of

Employees

1

Hires

EmployeeCPI

1
1

LUCASFILM
State Indicators

Location

YES
YES

Constant

R Square

0.895
137,271

Observations

Note
Source

Significant

Dr

at

1

level

Significant

Leamer’s backup data and materials

at

5

level

Pixar

Significant

revenue data

at

after

10 level
2005 are included

Case5 11 cv 02509 LHK

Document518

Filed10 07 13

4

Page72

of

73

Appendix 14A
Dr

Leamer’s Figure 20 RegressionIncluding Change

Variable

SP 500

Employee Class

AllSalaried

Dependant

in

Log Total Annual Compensation CPI

Variable

Estimate

St

Error

T Value

Conduct

Age

0.0066

0.0005

13.98

Conduct

Age 2

0.0001

0.0000

13.83

Conduct

Log Number

0.0043

0.0008

5.54

0.1309

0.0100

13.04

0.6894

0.0054

126.98

0.7449

0.0027

280.12

0.4988

0.0017

293.05

0.6678

0.0024

282.12

0.7070

0.0058

122.77

0.6943

0.0069

100.22

0.8204

0.0363

22.62

0.3023

0.0053

57.04

0.2581

0.0027

94.33

0.3694

0.0016

225.49

0.3012

0.0023

130.80

0.2567

0.0056

46.04

0.1985

0.0067

29.56

0.1737

0.0366

4.74

Log Age Years

0.3495

0.0415

8.42

Log Age 2

0.0380

0.0056

6.74

Log Company Tenure Months

0.0039

0.0050

0.78

Log Company Tenure

0.0005

0.0006

0.92

0.0027

0.0005

4.93

San Jose

1.5373

0.0151

101.59

Among Defendants

0.0566

0.0020

27.69

0.0656

0.0023

28.72

0.0026

0.0003

7.45

0.0135

0.0009

14.55

0.2182

0.0024

92.01

0.1319

0.0037

36.14

0.2371

0.0033

70.97

APPLE

0.0747

0.0162

4.62

GOOGLE

1.0592

0.0174

60.95

INTEL

0.1542

0.0146

10.59

INTUIT

0.1485

0.0193

7.71

PIXAR

0.7001

0.0422

16.60

0.1483

0.0480

3.09

of

New

Hires In the

Firm
Number

of

Employees

1

Conduct

1
APPLE
Log Total Annual Compensation CPI 1
GOOGLE
Log Total Annual Compensation CPI 1
INTEL
Log Total Annual Compensation CPI 1
INTUIT
Log Total Annual Compensation CPI 1
PIXAR
Log Total Annual Compensation CPI 1
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

1

2
APPLE
Log Total Annual Compensation CPI 2
GOOGLE
Log Total Annual Compensation CPI 2
INTEL
Log Total Annual Compensation CPI 2
INTUIT
Log Total Annual Compensation CPI 2
PIXAR
Log Total Annual Compensation CPI 2
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

2

2

Male

DLog

Information Sector Employment

Log Total Number
DLog

SP 500 Net

Year

of

Transfers

in

trend

Log Number

of

Total Return Index CPI

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLog Firm Revenue

Firm
Number

Employees

1

Hires

EmployeeCPI

Per

of

EmployeeCPI

1
1

LUCASFILM
State

Location

YES

Indicators

Constant

YES

RSquare

0.926

Observations

Note
Source

Significant

Dr Leamer’s

504,897

at

1

level

Significant

at

backup data and materials

5

level

Significant

at

10 level

Case5 11 cv 02509 LHK

Document518

Filed10 07 13

4

Page73

of

73

Appendix 14B
Dr

Leamer’s Figure 23 RegressionIncluding Change
Technical Creative and

Dependant

Variable

RD

in

SP 500

Class

Log Total Annual Compensation CPI

Variable

Estimate

St

Error

T Value

Conduct

Age

0.0077

0.0007

11.44

Conduct

Age 2

0.0001

0.0000

11.18

Conduct

Log Number

0.0099

0.0010

9.44

0.1717

0.0141

12.16

0.6662

0.0073

91.42

0.7299

0.0037

199.33

0.4425

0.0022

202.73

0.6405

0.0030

215.77

0.6672

0.0085

78.91

0.6508

0.0106

61.63

0.8548

0.0691

12.37

0.3141

0.0071

44.00

0.2505

0.0038

66.22

0.3607

0.0021

171.44

0.3177

0.0029

110.53

0.2888

0.0082

35.32

0.1053

0.0097

10.90

0.1398

0.0692

2.02

Log Age Years

0.5757

0.0587

9.80

Log Age 2

0.0676

0.0080

8.46

Log Company Tenure Months

0.0204

0.0068

3.00

Log Company Tenure

0.0016

0.0008

2.14

0.0064

0.0008

7.86

San Jose

1.5716

0.0209

75.07

Among Defendants

0.0443

0.0028

16.05

0.0881

0.0031

28.55

0.0078

0.0005

16.67

0.0213

0.0013

16.62

0.2308

0.0033

70.79

0.1028

0.0051

20.31

0.2359

0.0045

52.12

APPLE

0.1328

0.0244

5.44

GOOGLE

1.4013

0.0259

54.09

INTEL

0.1574

0.0218

7.20

INTUIT

0.1378

0.0315

4.38

PIXAR

1.5355

0.0770

19.94

0.0399

0.1036

0.38

of

New

Hires In the

Firm
Number

of

Employees

1

Conduct

1
APPLE
Log Total Annual Compensation CPI 1
GOOGLE
Log Total Annual Compensation CPI 1
INTEL
Log Total Annual Compensation CPI 1
INTUIT
Log Total Annual Compensation CPI 1
PIXAR
Log Total Annual Compensation CPI 1
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

1

2
APPLE
Log Total Annual Compensation CPI 2
GOOGLE
Log Total Annual Compensation CPI 2
INTEL
Log Total Annual Compensation CPI 2
INTUIT
Log Total Annual Compensation CPI 2
PIXAR
Log Total Annual Compensation CPI 2
ADOBE

Log Total Annual Compensation CPI

Log Total Annual Compensation CPI

LUCASFILM

2

2

Male

DLog

Information Sector Employment

Log Total Number
DLog

SP 500 Net

Year

of

Transfers

in

trend

Log Number

of

Total Return Index CPI

New

Hires In the

New

Log Total Number

of

Log Firm Revenue

Per

DLog Firm Revenue

Firm
Number

Employees

1

Hires

EmployeeCPI

Per

of

EmployeeCPI

1
1

LUCASFILM
State

Location

YES

Indicators

Constant

YES

RSquare

0.875

Observations

Note
Source

Significant

Dr Leamer’s

292,489

at

1

level

Significant

at

backup data and materials

5

level

Significant

at

10 level

4

Case5 11 cv 02509 LHK

Document424

2

Filed05 17 13

Page1

of

62

IN THE UNITED STATES DISTRICT COURT
FOR THE NORTHERN DISTRICT OF CALIFORNIA
SAN JOSE DIVISION
CONFIDENTIAL
TO BE FILED
SUBJECT TO PROTECTIVE

IN RE HIGH TECH EMPLOYEES ANTITRUST
LITIGATION

THIS

UNDER SEAL
ORDER

No 11 CV2509LHK

DOCUMENT RELATES TO

ALL ACTIONS

SUPPLEMENTAL EXPERT REPORT OF EDWARD

May

E LEAMER PH D

10 2013

REDACTED

Case5 11 cv 02509 LHK

Document424

Filed05 17 13

2

Page2

of

62

TABLE OF CONTENTS

I

Introduction Assignment and

II

Defendants Use

III

Empirical Methodologies for Exploring the

of

Summary

of

Conclusions

1
4

Compensation Structures

Somewhat

Rigid

Salary Structure

A
B
C
IV

6
6

VI

of

Regression Analysis

of Compensation

byTitle

Correlation Analysis

Title

byTitle

7

Structure

7

Multiple Regressions

of Compensation

10

Structure

14

Based Correlations and Multiple Regressions

18

Decile

Based

Correlation Analysis

18

Decile

Based

Multiple Regression Results

20

Additional Exploration of

1
2
3
4
VII

Structure

Based Correlations and Multiple Regressions 10

Title

Decile

A
B

Compensation

Correlation Analysis

Results of Title

A
B

V

Choice of Aggregation Level

Adobe

Adobe Correlations

22
22

Correlation Results

Headcount Matters

for Interpreting

24

Correlations

Correlations

25

Outliers

26

29

Internal Versus External Forces

i

Supplemental Expert Report

of

Edward

E Leamer

Ph D

Case5 11 cv 02509 LHK

Document424

Filed05 17 13

2

Page3

62

of

5 10 2013

CONFIDENTIAL

I

Summary

Introduction Assignment and

1

been asked by counsel

I have

my

RD

and

2

my

1

job

somewhat

titles

structure

would

be

to

require

p283

the parallel

movementsof

orderly and prices

move

in parallel

compensation
rigid
that

of

For

for so

for

these firms
studies

vast

also evidence

compensation

I

found

same group from two
This again

is

zinc

I

Class

forces

are

that the

not

that

this

dictated

so

do not normally

with a

somewhat

compensation

me

and found

I have

titles

and the

the

the internal forces

compared

also

empirical

to separate

that

in all

average

non technical

employees

compensation curves of these two

firms move in a

compensation curves

for

much more disparate

are evident but the external

to detect

present correlations

To

allow

parallel while the

different

that

movementsof

role in determining

Class of

highly

average compensation of each
Technical

parallel

I

typically are not

copper and

silver

saying that the internal forces

more difficult

In this Report

forces

for the Technical

groups within each firm are

4

have added

I should

Markets

titles

models

estimated regression

for all the defendants

forces are

what

specifically in favor of the hypothesis

played an important

are evident but the external

way

to

In this report I confirm this opinion with two additional

I have

of a

the existence

Could a nonrigid wage

not only to be consistent

contributions of internal and external

the

Class employees in

market conditions which

labor

numbers of

titles

Creative

report

lines I responded

reason I regard the

many

my initial

could

it

example gold

that

wage structure but
internal equity

Yes

with

unusual external

highly

been asked to focus

I have

Defendant

asked in the deposition

a hypothetical

in

to the

can be

analysis that

over time further confirming
each

at

respond

proposed Technical

to the

structure as you’ve defined it lead to parallel
thought

further

compensation of Technical

total

move together

pay

Answer When

the

and

in this case

data

Classidentified

Technical

Does

rigid

available

prior analysis

the employees belonging

Class

Question
specific

3

on

response

Conclusions

for Class Plaintiffs in this matter to

following questions regarding

conducted based on the

of

title

accommodate

that

compare

the

movement

over

time of the

with the average compensation of the firms
titles

that

cannot be accessed

on a

by

title

Page 1
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11 cv 02509 LHK

Document424

Filed05 17 13

2

Page4

62

of

5 10 2013

CONFIDENTIAL

title

due to

basis

insufficient

titles but representing just

analyzed correlations

of

data

approximately 63 percent of Technical

6 percent

relatively

of Class Period

narrow groups

computed

are

of employees

each

Class employees of that

approximately a tenth of the Technical
correlations

employee years

for all titles not just

20

They

reveal

Class

I also

comprising

firm These
that

there

is

topdown
amount

large

co movement

of

among most

of compensation

Class titles of each defendant These correlations

method in which

budgeting

receive

a

common compensation

and

possibly

by

The

evident

substantial

what

membersof

title

I previously

the firm in any given year

called

a

is

depending

common firmwide

with a

are consistent

increment which

within the

individual

what

creates

all

of the Technical

adjusted

on

by

title

circumstances

specific

component

somewhat

somewhat

of compensation

is

rigid salary structure which

allows the effects of the anti cold calling conspiracy

to spread broadly across

each firm

5

2 Do

Question

partly drove

as internal equity

opposed to only

6

Answer

I

the data show additional

external

have analyzed a model of sharing

bytitle

title

relatively

compensation of the

year byyear

I report
increases

explanatory variables
the previous years
average

title

Technical

below estimated
in

of Technical

other Technical
that

titles

title

by

Class

cannot be accessed

Class titles but representing

1 increases

in

multiple

the

regression models that explain
title

average Technical

ratio of average Technical

3 the previous
title

against

Class employees

average compensation at the

revenue divided by the average
in

movementsof

narrow groups of employees

overall

compensation

software jobs

of compensation effects

of Class Period employee years I also analyzed the

compensation of

Specifically

compensation structures as

Again to accommodate

approximately 70 percent

just 8.4 percent

such

market forces

employees compensation

7

the Defendants

within Defendant firms relative to

title

evidence that internal factors

level in terms of four

Class compensation

2

Class compensation divided by the
years

compensation

ratio of

firmwide average

4 the

percent change in

the San Jose Sunnyvale Santa Clara Metropolitan

Area hereafter San Jose

the

Statistical

MSA
Page 2
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11 cv 02509 LHK

Document424

Filed05 17 13

2

Page5

of

62

5 10 2013

CONFIDENTIAL

8

the vast majority of individuals

I find that

1 positive

within

fall

contemporaneous sharing of compensation

across time that would spread gains in compensation
This

is

with

consistent

my

previous opinion that

employees would have been impacted
Furthermore the
an

9

3 Do

No

Answer

do not

agreements
the existence

of

member compensation

of large groups

would not have been harmed by

necessarily

I

have performed the above mentioned

find persuasive

evidence

compensation structure The
this regard

roles
a

The

show

ripple

analysis

shows

case bycase basis

the collection

it is

of other

they differ in the job

4

Question
job

titles

Answer

based

No

correlations

on

there are sizeable groups whose

be candidates

statistically

titles

among

positive

An

each

of these positive

vast

rigid

performed in

across employees in very different

title

or group

Technical

statistical

is

separately on

studied

how much

titles

variability

statistical

many

strong positive

exhibit

One

Class there certainly are exceptions
with the overall

correlations

of

Class

relationships

from the class However

the thousands

corrected estimated of

correlative

majority of

titles

Technical

some

even

the true correlations

if

all

negative

for this kind of data

when

this

is

done the

of these negatives

is

Class

this is not justified

can cause

model

pooling of evidence across titles and
that

analysis I

to identify and exclude from the Technical

with negative

appropriate

and regression

somewhat

I

they contain are found to be tied closely together

for exclusion

because

estimates

when

from Defendants

or groups All these groups no matter

with the overall
titles

analyses

statistical

found that compensation almost always moves with

titles

a lack

on

to suggest that

spillover effects

possible

it

a restriction

grouped by compensation level

correlation

that

Although the

might consider
to

Is

and

of class

of employees

compensation might have been disconnected

12

Defendants

all

indicates

of class

the data show the existence

for distinct subgroups

separately

11

titles

coldcalling
memberswho

10

2 sharing

market forces

than only external

Question

or almost

non compete

the structure

and

effects

across other job

of gains over time strongly

sharing

internal sharing force driving

rather

by the

all

show

or groups that

titles

correlation

allows

are

some

analysis indicates

positive

In other words

Page 3
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11 cv 02509 LHK

Document424

Filed05 17 13

2

Page6

of

62

5 10 2013

CONFIDENTIAL

matters for interpreting the evidence about each

it

estimated correlations

13

In

sum

the

statistical

and econometric
of a somewhat

are positive

analysis I

evidence

conduct

herein

my original

in

rigid pay structure

at

II

Most

provide them information about

finding

each Defendant that would have

broadly including throughout

subscribe to services

market

prices

the imminent threat

opportunities with or without

However

these external

relevance and

the

offer

of loss of

is

jobs Such

an employee

suggest

with the information conveyed

That can ring off a loud alarm that

to

averages with limited

Regardless of what these services

compare

are intended

in line with the external

sources provide broad industry

reliability

information cannot

that

for various

information helps them keep compensation packages

outside

my original

Compensation Structures

of

not all of these defendants

if

with the economic

Class

Defendants Use
14

conjunction

reports supports

transmitted the effects of the agreements
Technical

the vast majority of

that

title

heard

all

their

by an

actual

way up

the

to the

CEO
15

The

information by an outside offer or even a cold

by management

A

affected

that

chain of

individual
threat

move

against

that

threat

similar employees

compensation

action

aware

by an

compensation

to salary increases

for a

First when

one

for

also of the implicit competitive

may

increase

Though

directly

opportunity

outside

the

feel
in

wise to

it

make a

compensation

market

does

for these

not require a

until they actually

receive

an

can minimizethe disruption to employee

might occur when an employee discovers

unfairly undercompensated
analogous

individual

bump in

for these similar individuals

offer preemptive

loyalty that

attractive

and management

to similar individuals

bump in

the specific

can transmit a

may make management

newlythreatened

outside

similarities

becomes aware of an

this

preemptive

much beyond

broadly across a firm for two reasons

single individual

management

can go

a response

can stimulate

call

A broad

preemptive

that

he or she had been

response

is

completely

that are tied to information provided

by

Page 4
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11 cv 02509 LHK

Document424

Filed05 17 13

2

Page7

62

of

5 10 2013

CONFIDENTIAL

employment

These responses

16

Similarity in

broad and not

are

worth

is

a

bump in

compensation

individual

jobs

can be broad

productivity

employees

who are

any

to treat

employees

individual

until

are likely

to have

an

perceive

that

compared

Fairness

some

with

is

feels

their

is

salary

employees

equivalent

As

doesnt

actually

increases

own views

in

their productivity

it

is

essential for

an

require

increase

in

but the force

materializes

compensation

adversely

above a

In addition

affected

similarity

if

they

unfairly high compensation

are receiving

reflect

and there

really

is

reason

no sure way to know

who got

care The

title

that

why companies

bump in

and grade

managements views of what

is fair

and

employee

of

structure
it

may

fairness beliefs

tend to follow guidelines laid out in terms of

ranges so employees can be assured

reasonable

high

contented

of job and performance

the perception of similarity that determines

the

come from

discussed in the paragraph

threat

to the employee

who doesnt

compensation may

This

encourage

that

them

compensation and

influence

outside

a matter of personal opinion

who

policies

and commitment

fairly

preemptive

and these employees can have

exactly

HR

salaried

range of achieved

levels of productivity

the contentment

fairness can necessitate

employees

17

yet the

market view of employee compensation

salary of

of

highest

can be

worker in most

of each

is

committed to the mission of the enterprise In order to

maintain or to increase

management

with accuracy

Fairness

marketbased

problem with

Firms need to use

The

levels of productivity

A critical

tied together

for a single individual

that the productivity

to determine

is difficult

strictly

is

can be

salaries

compensation

transmitted broadly across a firm

based

necessarily individual

one reason why

why

the second reason

market

regarding the compensation offered by the

services

that their

compensation

falls

within

range of their colleagues

Page 5
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11 cv 02509 LHK

Document424

Filed05 17 13

2

Page8

of

62

5 10 2013

CONFIDENTIAL

III

Somewhat

Empirical Methodologies for Exploring the

Rigid

Salary Structure

A Choice of Aggregation
18

The

data set

explore

I

on

individuals

composed

title

Based on

of compensation records of salaried

of the Defendants

the payrolls

the Defendants by

by grade

is

Level

and

some

for

These

of the Defendants

19

the

the

including

have chosen

I

data

individual

the individual

at the individual

studied

more aggregated groups
because

is likely

titles

are

grouped

the

title

in the

been

have

that

level

work

to

at

first

level or

with the

title

to be dominated by forces

which can make

level

titles

grouped by

Class titles 1

as Technical

These data could be

the

are

from counsel regarding the employees

instructions

Class except for Lucasfilm I limit the inquiry to the
identified

individuals

it

difficult

averages

that operate

at

the firm wide effects

to detect

spread of the anti cold calling agreements

some

broadly across the

titlelevel
firms

Averaging

across individuals

in a title

more transparent

thus making the firmwide effects

effects

documentary evidence shows Defendants

20

I have

internal equity

discovered

there are

some

that

titles

individual

estimate
adequate

The

title

and

which

behavior

still

for the statistical

Because Lucasfilm

inquiry

is

expanded

meaning

Technical Class for

did

the

data

analysis

works

briefly

and

well for

effect

many

but not plentiful

not provide

title

statisticians

the

idiosyncratic

that

Titles that

call

for

Apple which had

I

am

seeking to

statistically

data prior to 2006 there are insufficient

all

Defendants expect Lucasfilm

for

whom

it

applies

all

in

This

is

2005 and

of data unless the

presented below
to

annual

insignificant

estimates

years

is

have fewer

a title restructuring

Lucasfilm employees Hence the analysis

cover

but

titles

are other titles that

there

sets are too small to yield accurate

to

all

many employees

employees

masks the firmwide

work

the

structures

data set contains only eleven annual observations which

particularly troublesome

1

In addition a

used to manage their

seem much influenced by

that

observations tend to produce what
results

their

were used only

that

are sparsely populated

among

bytitle

the

individual

perspective on the compensation

analysis provides a clearer

and maintain

can average out the

is

limited

to

employees

Page 6
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11 cv 02509 LHK

Document424

Filed05 17 13

2

Page9

62

of

5 10 2013

CONFIDENTIAL

which did not provide

for Lucasfilm
include

few

a

just

and furthermore

individuals

for sparsely populated

21

To
of

the

much from

data the

the individual

with highly

titles

title

across

compensation

come and

go

I give

and

headcounts

variable

that

titles

averaging

the

of the titlebytitle data I also include

limitations

work but applied

statistical

In addition

median ages

variable

deal with

2006

can vary wildly as individuals

Adobe

of

prior to

not benefit

unlike

titles

some examples below
highly

may

individuals

titles

the ten groups of

titles

to ten groups

by ordering the

of

titles

titles

the same type

firm

in each

I have

formed

by average base compensation and

employeeyears
then

the

splitting

into

titles

ten deciles

based

on

the

number of

2

B
22

Correlation Analysis of
Economists often look
closely

different

The

relationship
question

move

compensation

23

There

are

to correlation

move

variables

from 0

absolute value

to

sign

in the

Compensation Structure

together

1 One indicates

on

the

correlation

same direction

structures

two types of

and

correlations

Correlation coefficients
perfect
indicates

I begin

how

range in

correlation zero indicates

whether or not the

my analysis

no

series in

Defendant

of

correlations

relevant

for determining

if

the
of compensation

series are similar correlation

of compensation changes

The

correlations

the levels of compensation emphasize longer run movements
correlations

statistically

with compensation correlations

compensation movementsof two
levels

to measure

coefficients

of the log of

and the

of the change in the log of the levels focus on year byyear

movements

C Regression
24

Analysis of

compensation and

For several

number

of

title

sharing

2

Correlation of

but could

effects

Defendants

certain

large

Compensation Structure

titles

also

made

class

come from

splits into

ten

compensation could come from

third variables

that

groups impractical

In

operate

on both

those cases a smaller

groups was used

Page 7
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page10

62

of

5 10 2013

CONFIDENTIAL

title

and

class

compensation

of a somewhat

confirm the existence

my

correlation

analysis

I

model which forces the
an explanation

of title

This regression

25

model

compensation

the

class

equation

increases

two

26

The

regression

class

effect

one

the

variable

includes
title

a

they

and

for the other

sense the

second

move apart

sharing

variable which

in the previous year
the

in

closely

If

the

the ratio of

is

While the

two compensation

coefficient

which the

is

on

positive

is

levels

action

first

move

taken at

this

class average

abnormally high compared

tends to get a special increase

in

the coefficient

two would be

the extent to which corrective

is

If

variables

effect

compensation

company

setting

the extent to which

market forces

in that

that following periods in

at the

regression

after controlling

for

increase

encompasses the

In the multiple

measures the extent to which

means

it

compensation
title

model

company when

as

then the employee would inherit 100 percent

sharing

first

the second measures

together

variables

By including the

at a particular defendant

after controlling

the

compensation to

sharing

regression

average real inflation adjusted

regression

variables

of the class compensation changes
is

a multiple

in class average real total

move together

were equal to

This

title

variables 3

equation the

in the

in particular

tied together

in

one of four explanatory

compensation

of this variable

revealed by

structure

compete with other

to

the increase

includes

allows us to determine

this variable

and

compensation

class

analysis of these

correlation

compensation

rigid

market forces To

example

examine company by company

explains

compensation

in class

title

as

for

compensation

compensation and

total

same time

at the

with the

compensation to bring it back

title

the

in line with

the class

27

The

regression

against

model

is

divided by the average

For each

title

title

both

two other determinants of

these other variables

3

requires

regression

I

of these sharing

title

compensation

the previous years ratio of
title

compensation

exclude from the class average

variables to
at

company

firm
wide

This variable

real total

the

compete

One

of

average revenue

allows us to determine

compensation the compensation

of

the

itself

Page 8
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page11

62

of

5 10 2013

CONFIDENTIAL

which
that

titles

on

28

share increases

and

firms

fourth variable

the percent

is

workers

there are general

2

other Technical
increase

in

coefficients

with this

the firm and

ones

are the

may

It

who would

title

increases
a title that

Figure

in software jobs in the

which

1 as

for the

is

intended

effect

two sharing

in the

compensation

received

variables

a

move

how hot

a relatively small percent

company

years This indicates

last

year

sharing

in a fashion that helps align workers

meaning

increase

if

1

Class titles

increase

and

at

relative to

tend to receive

will

a positive

In this

are positive

of other Technical

or

MSA

title 4

compensation

MSA

San Jose

to reflect

estimated for one Intel

can expect to receive

Class titles at the

in subsequent

forces

have revenue

have an accentuated

effect Both the contemporaneous and lagged coefficients
equity

might be expected

job market generally not just in the San Jose

this regression

example the two
that

growth

job market variable

cold was the technical

I illustrate

workers

overall

success

This the external

29

creative

in firm revenue

with their firms since they

relationships

the

The

any

technical

critical

sharing

if

a larger

internal equity

suggest

that internal

compensation together

with that of employees in other roles at the firm

4

As mentioned

afford

before this regression

a sufficient

Estimated

and

are

number
excluded

of

is

estimated separately

observations

from the

6 observations

coefficient

for
or

distribution

each

title

7 consecutive
calculations

and

company

Titles that

years are treated

as

do not

Not

presented in this report

Page 9
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page12

of

62

5 10 2013

CONFIDENTIAL

Figure 1
Example of Compensation

Illustrative

Intel Named

Plaintiff

Title

Sharing

Variable

RD

Total

Average Annual

0.784

0.064

12.238

0.000

1

0.251

0.098

2.562

0.051

1

0.032

0.094

0.346

0.743

Employment

0.092

0.126

0.731

0.498

0.541

0.411

0.698

Compensation

Total

Compensation

1

Avg Annual Total Comp

Avg Annual

Title

Total Compensation

Forces Variables

Log

Firm Revenue

Avg Annual

Title

DLog

San

Information

Jose

1

Per Employee

Total Compensation

Sector

Constant

Observations

10

Rsquared
1
2
3

0.986

Significant
Title

RD

1

at

All

Compensation

Defendants

A Titleby

is

is

computed

employee

are

5

level

Significant

computed as the average of
over

all

Technical

title

Creative and

10

level

RD

total

compensation

employees other than the

tilte

itself

Adjusted

Inflation

compensation

data

Title Correlation Analysis of
for all Defendants

Exhibit 2

other Defendants

but here

it

is

3 which

indicate

enough

to

at the

Below

between

of

titles

title

Compensation Structure

are reported in Exhibit
I will

summarize the

the fractions

correlations

compensation

at

employee’s annual

Based Correlations and Multiple Regressions

correlations

positive

at

Significant

Variables

Results of Title

The

level

Average Compensation

Avg Total Comp

Source

30

5

Variable

Effect

Log

IV

4

Variable

Effect

RD

DLog

Note

Pvalue

value

3

0.223

Average Annual

Title

Contemporaneous

External

T

2

Variable

DLog

Lagged

7

Std Error

Coefficient

1
Dependant

Model

Regression

SOFTWARE ENGINEER

discuss

overall

the

1

Adobe

and

results in detail

results with Figure

weighted

2 and Figure

by employee years with

compensation and Technical

same firm restricted

Adobe

to titles with

six

Class

or more annual

Page 10
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page13

of

62

5 10 2013

CONFIDENTIAL

The

observations

Although there
that

among

32

are

error

which

thousands

the

Moreover

statistical

that

of

titles

enough

produce some

to

included even

if

all

of cases

the true correlations

for each

Once

title

toward

this shrinkage

is

is

strong support

Share

are formal

on

will

all

the

titles

the results indicate

that

which
for

be positive strengthening the

movementswith

the class overall

Change

Correlations

Change

of

Compensation

mean across

the

done

the corrected results

that all titles in the class share

2 Large

were positive

based

titles

with

estimates

There

are positive

mean

probably have similar correlations These methods

estimates

of these negatives

negative

are positive

allow pooling of results across

titles

computed

are

true correlations

that

of course positive

Figure

large

is

all

the

would shrink the

conclusion

These estimates

are negative

that

methods

assumption

many

estimated correlations that does not

negative

the fact that the vast majority

for the conclusion

is

some

any true correlations

statistical

more extreme

of correlations

distribution

31

with five or fewer tend to produce a

titles

Correlation by Titles

are Positive

100

80

60

Share

40

20

0

ADOBE

GOOGLE

APPLE

INTEL

Negative

Source
Note

Distribution

of

Defendant

growth

in

Employee

Positive

Compensation

avg compensation
Weighted by class

PIXAR

INTUIT

correlation
period

Data
over

employee

Correlation

titles

with

Analysis

six

or more

years

of

data

years

Page 11
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page14

62

of

5 10 2013

CONFIDENTIAL

Figure

3 Large

Share

of Level

Compensation

Correlations

Correlation

by

are Positive

Titles

100

80

60

Share

40

20

0

ADOBE

GOOGLE

APPLE

INTEL

INTUIT

Negative

Source
Note

Distribution

Defendant

of log

avg

Weighted

33

It

is

not just

statistical

Positive

Employee Compensation
compensation

variability

correlation

by class

period

that

Data

over

with the class overall

examination of

correlations

some

of the

with

six

Adobe

Analysis

or

more

years

of

data

years

can explain

the

negative

or small

of employees within a

employees come and go can cause changes

close

Correlation

titles

employee

correlations Changes in the composition

normal correlation

PIXAR

in title

that

as

compensation and mask the

I will illustrate

titles

title

this point

below with

a

have low or negative

with the class

Page 12
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page15

of

62

5 10 2013

CONFIDENTIAL

Figure 4

Summary of Compensation Change

Positive

Employer

Sign

Not

Significant

Percent

Negative

Significant

Sign

Not

Significant

Percent

Correlation

Percent

Significant

Percent

Total
Percent

ADOBE

67

32

0

0

100

APPLE

54

35

1

10

100

GOOGLE

76

22

0

2

100

INTEL

94

6

0

1

100

INTUIT

81

14

0

5

100

PIXAR

86

13

0

1

100

Source Defendants
Note

Distribution of growth

in

employee
compensation

data Correlation Analysis

compensation

over

correlation

titles

by class period employee

Weighted

with

six

or

more years of data

years

Figure 5

Summary of Compensation Level

Positive

Employer

Sign

Not

Significant

Percent

Correlation

Negative

Significant

Not

Significant

Percent

Sign

Significant

Percent

Percent

Total
Percent

ADOBE

92

5

0

3

APPLE

78

16

1

5

100

GOOGLE

83

16

0

1

100

INTEL

85

14

0

1

100

INTUIT

45

40

2

12

100

PIXAR

84

15

0

0

100

Source Defendants
Note

Distribution of log avg

employee

compensation

Weighted

compensation
correlation

100

data Correlation Analysis
over

titles

by class period employee

with

six

or

more yearsof data

years

Page 13
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page16

of

62

5 10 2013

CONFIDENTIAL

B
34

Titleby Title Multiple Regressions
As

above

described

compensation

that

I also analyzed a multiple

the year byyear increases

explains

Class compensation

average Technical

1 increases

jobs in the San Jose

The

data set

many

is

title

words

36

seven

least

6 and

observations

Those

results

positive

show

sharing

relationship
significant

for statistical

Figure

Second

of titles

If

This model is completely appropriate

on the other hand

title

A

if

only to

would

yield a small

largely

determined by variability in compensation

estimating

makes

it

and probably insignificant

less likely

effect
of

force
title

all

of

majority

these are

To

not more likely to find a sharing

titles

by sharing

body of

relationships

significant

across

class overall

attempt to link

a

are statistically

in the context of

statistically

came from the

B

Class

or the lagged

relationship

all

In

sum

Class titles

equally

A to

across

all

at

a

titles

the class overall

unless the variability in compensation
title

with at

have

titles

a small fraction

supported

my

Technical

wins

employee years

results occur

B then

In other

titles

of

The

5

effects

class wide results for

is

for internal relationships

the sharing

were connected

results are particularly

it

contemporaneous

Many of

statistically

that

First the vast

relationship

a heavy

is

sharing

their Class Period

a general

is

2011 and

significant coefficients

of those that are negative

this analysis provides support

5

in software

revenue sharing

approximately 30 percent

Third even these negative

majority

the

significance

7 below

in either the

regression

insignificant

statistically

the following

effect

evidence that there
for the vast

statistically

more

and more than 91 percent of

37

change

number of

in the

market effects and

have

in the competition

I present in Figure

title

average revenue

percent

the

four variable

reflected

is

The

prevalent for the external
variables

A

observations

coefficients

two sharing

firm
wide

observations from 2001 to

limited to eleven annual

burden with such data which
insignificant

4

compensation

ratio of

years

firmdivided by the average

the

at

average

in

MSA

have fewer

titles

2 the previous

previous years ratio of

divided by the average

35

firm

Class compensation

3 The

compensation

the

at

compensation

in average

at the title level in terms of four explanatory variables

Technical

model of

regression

of

the class

were

put this in simple terms the model that I

am

effect

Page 14
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page17

of

62

5 10 2013

CONFIDENTIAL

firm that would tend to

make

impact

agreements

of the

common to

all

Class

members

38

Thus

the vast majority

relationship

is

6 Large

my

to support

non compete

common across
Figure

titles

with other Technical

of these results
alleged

of these

Share

of

a positive

Class titles

at

internal equity

the same firm

previous conclusion

agreements

the Technical

have

would be

The

sharing
implication

that the impact

common across

of the

the class and

Class employees in particular

Contemporaneous

Contemporaneous

Coefficients

Coefficient

by

are Positive

Titles

100

80

60

Share

40

20

0

ADOBE

GOOGLE

APPLE

INTEL

Negative

Source
Note

Distribution

of

Defendant

estimated

Employee

PIXAR

INTUIT

Positive

Compensation

Data

Regression

Analysis

contemporaneous
coefficient over
titles
with seven
Weighted by class period
employee years

or

more

years

of

data

Page 15
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page18

of

62

5 10 2013

CONFIDENTIAL

Figure

7 Large

Share

of

Lagged

Lagged

Coefficients are Positive

by

Coefficient

Titles

100

80

60

Share

40

20

0

ADOBE

GOOGLE

APPLE

INTEL

Negative

Source
Note

Distribution

Employee

Defendant

of estimated

lagged

PIXAR

INTUIT

Positive

Compensation

coefficient

Weighted by class

over

period

Data
titles

Regression

with

employee

seven

Analysis

or

more

years

of

data

years

Figure 8

Summary of Contemporaneous and Lagged Net

Positive

Employer

Sign

Not

Significant

Percent

Negative

Significant

Percent

Sign

Not

Significant

Effect

Percent

Total

Significant

Percent

Percent

ADOBE

22

75

0

3

100

APPLE

23

62

0

14

100

GOOGLE

12

69

2

17

100

INTEL

88

11

0

1

100

INTUIT
PIXAR

73

23

0

4

100

60

39

0

0

100

Source

Note

Distribution

of

Defendants
the

employee

compensation

sum of estimated contemporaneous
Weighted

data Regression Analysis
and lagged

coefficients

by class period employee

over

titles

with

six or

more years of data

years

Page 16
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page19

62

of

5 10 2013

CONFIDENTIAL

39

It

may be

important

relationships

Adobe

many

that

parameters

with

separately

40

The

regression

all

much

results for

1

Adobe

The

but instead

titles

relevant

here because

regression

coefficients

first

a

titles

t statistic

in

data

it

is

18

in excess of

excess of

2

has a

tvalue

above

this

would

yield a small

the

R

Among the

Column 15 The

has one negative

title

A

2 only 4 of 41
and no

information about
of

correlations

Column

9 which

compensation

is

is

together in Section

said to

4

is

produce

for a sharing

were connected

force

fit

sq in the

and probably insignificant

effect

of

titles

B

Column

market variable

variable

tstats The

significant

B then

unless

tstatistics more

and the revenue

results are

of observations diminishes

comes from the

title

reason

with eleven years of

titles

external

number

title

that

my

class overall

equally

A to

attempt to link

the variability in

The model that

I

across

all

the class

compensation

of the class

am

makes

estimating

it

effect

of the model associated
last

that

only to

For

than the class wide

positive

into the table as the

not more likely to find a sharing

The increment in the

comparing

7

by conventional standards

were largely determined by variability in compensation

7

not

of the estimated regression 7

part

in absolute value

Column 16

variable

in excess of

model looks

on the other hand

less likely

title

wide variables6

by the

are sorted

in average real
is

each

is far

regressions

are reported to their right in Section

are highlighted

effect

more mixed deeper

noted

2

corrective

Column 17

overall

descriptive

of the four variables are collected

significant estimate

contemporaneous

If

The

Adobe

the two sharing variables that jump out with high

often the

As I

which

connecting

Column

this correlation

require

tstatistics
statistically

titles

would

or more years of data are

two Sections give

tstatistics

3 and the corresponding

6

this

with the

with seven

titles

between the percent change

the correlation

Roughly

with other

For example with 101

titles

simpler structure

the log levels of average real compensation

41

by the conspiracy

more observations

the data and the two correlations These

The

a matrix of sharing

to estimate with only eleven years of data

of the other

reported in Exhibit

more

is

the estimation of a 101 by 101 matrix of connections

have estimated have a

I

there

in principle

directly affected

titles

in the class with six or

titles

potentially

too

connect

that

that

are tied together with these affected

that

titles

to understand

with the

column with the squared

last

three

explanatory

variables

can be found

by

of the correlation

Page 17
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page20

62

of

5 10 2013

CONFIDENTIAL

42

This confirms the
titles

The

summaryabove

almost always positive

to the lagged ratio of
indicates

that

the

if

title

V

Decile

43

equal

variable

compensation

in the

title

from its normal relationship

with

taken to either raise or lower compensation

in

Based Correlations and Multiple Regressions
titlebased

groups of

titles

regressions

form the

ten

deciles based

A Decile
The

To

on which

I split each

groups

these

to conduct

the

Defendants

I ranked

correlation

correlation

on the

study

analysis

I have

formed

and the multiple

To

Class titles into ten groups

Technical

basis of average

lifetime of the title

analysis of the ten groups

Class employees

ten groups that had positive
for the levels correlation

inflation adjusted

and then divided these up

shows both immediate

structures

yields strong evidence

for each

into

correlations

its

firms

and longrun

with and supports

of both short

subgroup of the Defendants

Figure 9 and Figure 10 indicate
with the Technical

and 10 out of 10

every group shares in

consistent

in this

that are

Correlation Analysis

and long run compensation

Thus

titles

titles

on employee years 8

Based

Technical

titles

excludes

necessity

include

compensation over the

total

by

study just described

populated

infrequently

is

departs

corrective

title

The

44

is

the

relative to title

compensation
action

on

coefficients

compensation

the class then corrective
the

providing direct evidence of sharing across

for the percent

the numbers of the
Class 10 out of 10

change

correlations

compensation structure Every group
correlation

my conclusion

structure
that

for every

group

This

the Defendants

compensation was semirigid

8

Since Lucasfilm

Although

I

did

not provide

title

data individuals

were ranked

attempted to break the firms up into 10 equal

sized

in

groups

a similar fashion
equal

for

Lucasfilm

based on employee

years some

groups end up being larger than others because of some big titles

Page 18
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page21

of

62

5 10 2013

CONFIDENTIAL

Figure

9 Large

Share

of

Change

Compensation Change

Correlations

Correlation

are Positive

by Deciles

100

80

60

Share

40

20

0

ADOBE

GOOGLE

APPLE

INTEL

INTUIT

Negative

Source
Note

Distribution

Defendant

of growth

in

Employee

LUCASFILM

PIXAR

Positive

Compensation

avg compensation

correlation

Data

Correlation

by class

weighted

Analysis

period

employee

years

Figure 10 Large Share of Level Correlations are Positive

Compensation

by Deciles

Correlation

100

80

60

Share

40

20

0

ADOBE

GOOGLE

APPLE

INTEL

INTUIT

Negative

Source
Note

Distribution

Defendant

of log

avg

Employee

compensation

LUCASFILM

PIXAR

Positive

Compensation
correlation

Data

weighted

Correlation

by

Analysis

class period

employee

years

Page 19
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page22

45

62

5 10 2013

CONFIDENTIAL

B

of

Decile

Based

Multiple Regression Results
have

Multiple regressions

summarized

in Figure

contemporaneous and

also

been estimated with these

11 and Figure

laggedare

12 below

the

positive

decile

data As

sharing

effects

both

rule

Figure 11 Large Share of Contemporaneous Coefficients are Positive

Contemporaneous

Coefficient

by Deciles

100

80

60

Share

40

20

0

ADOBE

APPLE

GOOGLE

INTEL

INTUIT

Negative

Source
Note

Distribution

Defendant

of estimated

Employee

LUCASFILM

PIXAR

Positive

Compensation

contemporaneous

coefficient

Data

Regression

weighted

by class

Analysis

period

employee

years

Page 20
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page23

of

62

5 10 2013

CONFIDENTIAL

Figure 12 Large Share of Lagged Coefficients are Positive

Lagged

by Deciles

Coefficient

100

80

60

Share

40

20

0

ADOBE

GOOGLE

APPLE

INTEL

Negative

Source
Note

46

The

Employee

Defendant

Distribution

of

almost always positive

indicate

that

the

if

title

Compensation

Data

Regression

by class

weighted

on the

coefficients

compensation of

Analysis

period

employee

years

corrective

a decile departs

with the class then corrective

relationship

PIXAR

Positive

lagged coefficient

estimated

LUCASFILM

INTUIT

action

is

variable

from

in Figure

12

normal

its

taken to either

or lower

raise

classwide
compensation
impact

in the decile

suppressing wages in

The cold
some

calling conspiracy

titles

would have some

effect

averages which in turn would suppress compensation

the

47

titles

are several

important

in either the

Second those

things to note

contemporaneous

group Many

of these are

of positive
statistically

support for internal relationships
impact

common to

of negative

First every group
relationship

are not statistically

that are negative

in the context of evidence

make

in all

direct

on the

or almost

all

of

in the class

Figure 11 and Figure 12 contain a few instances

effect

would have

that

sharing

across

all

has a positive

There
sharing

or the lagged relationship
significant

relationships

significant

estimates

In

Third these occur

for almost every

sum this

analysis provides

these groups that would tend to

each

Page 21
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page24

62

of

5 10 2013

CONFIDENTIAL

48

Here

I

want

to issue another

coefficients

It is important

the changing

composition

number of

adding a

compensation
workers
impact

To

into account

test this opinion

Adobe

dataset

as set

examined the

any of the

51

Adobe

The

have

forth

for each

closely

conclusion

I previously

statistical

which

the

reported

10

in their

headcounts

in

Class

and

outputs for the

correlation

my view

confirm

that the title
is

consistent

title

I

find nothing

have

similarly

in that

data to

that

the

movement

Class with the

compensation

of a

in the rest of the Technical

and the

class have

with sharing

Common Factors

characteristics

compare

movement

of the

Analysis

moves

in

a way

A
that

Class thus supporting the

coordinated compensation

that

compensation

individual

characteristics

title

title

title

levels

a

of gains and broad impact of the anti cold

on measured

within a

of real

Class overall but excluding the selected

means

but also depends

in

the Technical

confirmation that at least some individual

changes in the individual
the

correlation

demonstrated with the

on

examined the

below They

title in

compensation

similar to

any year depends

for titles with just a

find no compelling reason

from the Technical

of the other defendants

numerical correlations

high positive

9

of individual

large changes10

I

the

Adobe Correlations

compensation of the Technical

fact

average

company and

variability

of these data

titles

For instance

titles

be a bigger problem

to

9

Correlation Results

compensation

is

title

by

to the question of whether

Idiosyncratic

experience

the limitations

I

can be affected

this conclusion

contradict

1

data

going

title is

Additional Exploration of

50

within each

workers might bring down the

for titles that

this analysis to exclude

VI

of the workforce

of the unlawful agreements
within a

of negative

to realize that these coefficients

or vice versa for reasons unrelated

few employees and

Taking

about misinterpretation

share broadly in things such as the gains of the

characteristics

49

junior

warning

matter

can cause changes

at

characteristics

in

the individual
including

level in

age This

is

and this raises the possibility that
title

compensation

that

can mask

firmwide common component

Though

a stable

headcount

can come

from equal numbers of departures and new

arrivals

Page 22
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page25

of

62

5 10 2013

CONFIDENTIAL

whether

calling conspiracy

the Technical

52

directly affects

it

on 5

correlations

based

The

sorted

to

6 and

is

first

shaded

Technical

53

The
This

first

is

in

green

they are populated

title

Titles with the weakest

since

6 or more years

shaded

statistically

in

insignificant

Class overall

with the Technical
statistical

The

was populated from 11

with the Technical

correlation

statistical

are

in

of years the title

of the

Class at

correlation

Adobe

with the

yellow

in Exhibit

1 has the

first

year of data for each

title

the early years from 2001 to 2003 had a sharp decline

Class compensation for

It

under study or the rest of

observations are often

by the number

column of numbers

early years thus are

together

or fewer

Adobe

Class at

important

Technical

if

then by the correlation

Titles with the strongest
are

title

Class

Titles are included in the table

table

the

an important

would not be

Adobe

test

surprising

in Figure

as illustrated

bed for

identifying

which
weaker

to find statistically

in

13 and these

titles

moved

results if these

years are not included

Figure 13

Class Average Total Compensation

Technical

Compensation

Adobe

Total

Average

Source
Note

Inflation

Defendant

adjusted

average

Employee

Compensation

compensation

with

Data

2011 as base

year

Page 23
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page26

62

of

5 10 2013

CONFIDENTIAL

54

The second column
populated This

the

reports

number of

also important

is

since the

during which the

years

truncated this table

the number

at

For

that

6 or more

of years equal to

reason I have
since

with 5 or fewer years populated

55

are estimated with greater statistical

The

number

2
56

It

third

column measures

Headcount Matters
is

my view

that

title

structure

and personal

characteristics

consequently

some

changes

go

variability

titles

in average

may have

as

which masks

is

and

comes from

employees come and

unusual employee

substantially

a close

fully

tenure

on compensation

of employees

lose or gain a large fraction
that

company

at the title level

of employee characteristics

that

structure but not

age experience

compensation

in

compensation

of these characteristics

like

title

have an impact

have just a few employees

and

characteristics

Variables
to

error

Correlations

influenced by the

are likely

of the change

in the distribution

Titles that

is

the cases

of employee years

for Interpreting

compensation

by the

determined

the

was

accuracy of the estimate of

statistical

depends on the number of observations

correlation

title

connection

influenced by

may have

variability

with the Technical

Class overall

57

The

Technical

Figure

may

14

Class overall

are losing

gaining

may have

a rising

headcounts

employee

headcount

as illustrated

similar to the Technical
characteristics

while

in

Class

titles

that

workers much more rapidly than the Technical

average compensation histories different

Class not because

employees

in

similar movementsin

workers or

Class overall
Technical

movement

Titles with

experience

has experienced

there

in the title is changing

is

from the

no sharing but because the group of

enough to mask the sharing

Page 24
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page27

62

of

5 10 2013

CONFIDENTIAL

Figure 14

Adobe

Class Average Headcount

Technical

per

Title

40

30

Title

20

Employees

10

0

2001

2002

2003

2004

2005

Source

3
58

2006

2007

Employee

Defendant

Compensation

2009

2010

2011

Data

Correlations

As

described

determining
correlations

above
if

the

there are two types of correlations

movementsof

Section

compensation

2

the

in Exhibit

in the title

two

third

change in the logarithm of average real total

Technical

The

for

are relevant

The

first

column

1 compares the logarithm of average

The

Class

which

series are similar

and the logarithm of average

of the rest of the Technical

59

2008

real total

of

total real

compensation

column of Section 2 compares the

compensation

of the title with the

Class excluding the title

tstatistics

corresponding

for these correlations

are reported immediately

tstatistics
following each correlation
greater than

years in

which the

two

title is

and the

are

statistically

shaded

populated

The

significant correlations

table is sorted first

and second by the

by

correlation

the

with

number of

between the

log levels

60

The

statistically

from the

longest

most

significant

time series with

correlations
all

with the shaded

tstatistics come

eleven years of data populated That

is

a

Page 25
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page28

of

62

5 10 2013

CONFIDENTIAL

of any

feature

significant

statistically

61

There

are

no

negative

populated These
statistically

4
62

exercise

statistical

is

time series the more

the

are the findings

correlations

positive

significant

the longer

in

for the

correlations

41 titles

with

all

eleven years

larger than zero

are statistically

39 out of the 41 cases

Outliers

To

fully

these correlations and the significance

understand

anomalies it may be
16 have the average
employees

helpful
real

to look at

some

Class overall

Figure 15

eleven years of data that are most highly correlated
overall and Figure 16 has the least correlated
together

which

is

The

title

with the lowest correlation

different but not dramatically

of the

data displays Figure 15 and Figure

compensation for ten Adobe

in the Technical

not

or

titles
is

titles

and

illustrates

for the

the five

with the Technical
All these titles

Adobe
titles

with

Class

move

TECHNICALWRITER

2

so

Page 26
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page29

of

62

5 10 2013

CONFIDENTIAL

Figure 15 Selected Adobe Titles with a Full 11 years of Data

Most Correlated

Average Total Compensation

Compensation

Titles

Total

Average

Source
Note

Titles

Defendant
with

Inflation

highest

adjusted

Employee Compensation Data Correlation Analysis
log compensation
correlation
among fully populated
average total compensation
with 2011 as base year

titles

Figure 16

Compensation

Correlated Titles Average Total Compensation

Compensation

Least

Total

Total

Average

Average

Source
Note

Titles

Defendant
with highest

Inflation

adjusted

Employee
log

Compensation

compensation

average

total

Data

correlation

compensation

Correlation

among
with

fully

Analysis
populated

2011 as base

titles

year

Page 27
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page30

62

of

5 10 2013

CONFIDENTIAL

63

However

as noted

above when

may change

characteristics

with the lowest correlation

change

headcounts

too The

substantially
are illustrated
is

for the two

headcounts

in Figure

17 The

headcount

very volatile with a standard

the percent

change equal to 72 percent compared

benchmark

of 11 percent

away

compared

with the Technical

with an average annual percent
Class benchmark

17 Headcounts
Least

with the Technical

of

for

of

Class

basically

12 percent

of

increase

titles

deviation

title is

withering

Figure

employee

substantially

5 percent

Least Correlated Titles

Correlated

Titles

Headcount

12

10

8

Headcount

6

Title

4

2

0
2001

2002

2003

2004

Source
Note

64

The

variability

problem

It

Titles

2005

Defendant
with

lowest

2006

2008

Employee
Compensation
Data Correlation
Analysis
compensation
correlation
among fully populated

for these

substantially

two

to the

smooth

elevation

of
age of

compensation

titles

not just a hypothetical

Class overall

jump upward
is

disconnect

in the Technical

2011

titles

which

in Figure

18

are

In

of the median age of the class the median age
has a big

contribute to the apparent

titles is

2010

the median ages for these

contrasted with the median age of the Technical
contrast

2009

log

in the headcounts

has affected

2007

in

2006 and

highly volatile

These

between compensation

Class overall

And

in any

the median
facts surely

in these titles

event these

and

results

Page 28
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page31

of

62

5 10 2013

CONFIDENTIAL

offer

no reason

pay structure

my conclusion

to question

that

applied to

all

these titles I offer these

two

presence of a few outlier

titles

how

conclusions about

its

let

been harmed

examples simply to
in the analyses

rigid

in

the point that the

illustrate

our

does not challenge

basic

these companies pay their employees which are also

alone convincing

evidence

18 Median ages
Least

Correlated

I

have not seen any

any of these

that

by the anti competitive behavior

Figure

a somewhat

exhibits

employees including those

salaried

by economic theory and the evidentiary

supported

evidence

of

Adobe

that

I

titles

would not have

have studied

Least Correlated Titles

Titles

Median

Age

45

40

Age

Median

35

30

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

Technical

Source
Note

VII

Titles

Defendant
with

Employee
Compensation
Data Correlation
Analysis
compensation
correlation
among fully populated

lowest

log

2011

Class

titles

Internal Versus External Forces

65

The

regression

are generally

forces

I

analysis reported

more

detectable

expand on

above

indicates

generally

this finding

in this section

correlation

or the external

Class employees and the

within firms between

effects

market

with an examination of the

Class employees of each of the defendants

more

the internal sharing

than either revenue sharing

average real compensation for the Technical
Technical

that

I

show

non

here that there

is

these two groups than between

Page 29
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page32

62

of

5 10 2013

CONFIDENTIAL

firms for either group

Thus again

very evident while the external

66

Figure 19 below
for the Technical

compensation
average

that

For most defendants
closely

tracks

compensation

total

than between

RD and

is

firms In other

to detect

non Technical

for the

these two subgroups
should

It

much more

generally

forces are

the average total compensation

one another

words

more difficult

are

defendant

Class employees

the internal sharing

that

market forces

for each

illustrates

NRD

employees

I observe

also

Class

have total

be evident

similar within each

that

firm

the internal sharing forces dominate and

keep the compensation of the Technical

Class employees and the

non Technical

Class employees closely aligned

67

This visual observation

numerically by the computation of the

confirmed

is

over time of the change

correlations

in logarithms

of the average total

compensation between these fourteen groups of employees reported

The boxes down

Correlations in excess of 0.9 are shaded

RD and NRD

the within firm correlations

between

boxes refer to comparisons

between firms Four out of

boxes and

are in these

correlations
correlation

of 0.86

correlation

in

every

row and column

making
by very

contaminated

it

hard to estimate

large

bonuses

in

five of the

Google has an

the

percent

and

1

these

shaded

internal
is

the largest

Lucasfilm has a very

change

The

correlation

for producers

Table

Correlations outside

except for Lucasfilm

little variability

in

the diagonal contain

Furthermore the within firm correlation

short time series with very

compensation

in addition

real

in

Pixar

directors

data are

in

2002 and

2006

68

Table 2 has the levels correlations

that

capture the longer term

of the compensation series These confirm the importance
forces compared

with the external

forces

sense that the within firm correlation

column except
because

the

for Lucasfilm

Lucasfilm data

compensation

at

is

is

forces for

all

of the internal

but Lucasfilm

the largest correlation

Lucasfilm and Intel appear to
confined

co movements

in every

move

in the

row and

together only

to a brief period of stable growth of

both firms

Page 30
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page33

of

62

5 10 2013

CONFIDENTIAL

Figure 19 Defendant

RD vs NRD Average

Total

Compensation

Page 31
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page34

of

62

5 10 2013

CONFIDENTIAL

Table
Correlations of

Changes

in

1

Defendants Average

Total

Compensation

2001 2011
Adobe

Apple

Google

Intel

Intuit

Lucasfilm

Pixar

NRD RD

NRD RD

NRD RD

NRD RD

NRD RD

NRD RD

NRD RD

NRD

1.00

0.94

0.66

0.56

0.17

0.16

0.47

0.60

0.63

0.60

0.19

0.62

0.53

0.53

RD

0.94

1.00

0.64

0.65

0.13

0.24

0.34

0.45

0.53

0.51

0.12

0.67

0.51

0.37

NRD

0.66

0.64

1.00

0.93

0.48

0.17

0.02

0.16

0.85

0.73

0.08

0.87

0.56

0.16

RD

0.56

0.65

0.93

1.00

0.42

0.07

0.12

0.00

0.77

0.63

0.11

0.83

0.45

0.05

NRD

0.17

0.13

0.48

0.42

1.00

0.86

0.51

0.39

0.20

0.17

0.49

0.89

0.62

0.21

RD

0.16

0.24

0.17

0.07

0.86

1.00

0.53

0.50

0.09

0.06

0.68

0.83

0.50

0.19

NRD

0.47

0.34

0.02

0.12

0.51

0.53

1.00

0.97

0.31

0.30

0.01

0.92

0.00

0.89

RD

0.60

0.45

0.16

0.00

0.39

0.50

0.97

1.00

0.38

0.33

0.23

0.70

0.03

0.89

NRD

0.63

0.53

0.85

0.77

0.20

0.09

0.31

0.38

1.00

0.91

0.15

0.17

0.43

0.28

RD

0.60

0.51

0.73

0.63

0.17

0.06

0.30

0.33

0.91

1.00

0.51

0.55

0.63

0.34

NRD

0.19

0.12

0.08

0.11

0.49

0.68

0.01

0.23

0.15

0.51

1.00

0.24

0.03

0.38

RD

0.62

0.67

0.87

0.83

0.89

0.83

0.92

0.70

0.17

0.55

0.24

1.00

0.58

0.29

NRD

Adobe

0.53

0.51

0.56

0.45

0.62

0.50

0.00

0.03

0.43

0.63

0.03

0.58

1.00

0.29

RD

0.53

0.37

0.16

0.05

0.21

0.19

0.89

0.89

0.28

0.34

0.38

0.29

0.29

1.00

Apple

Google

Intel

Intuit

Lucasfilm

Pixar

Note

Values

above

0.9

shaded

Source Defendants employee compensation data

Page 32
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page35

of

62

5 10 2013

CONFIDENTIAL

Table
Correlations of

2

Defendants Average

Total

Compensation

2001 2011
Adobe

NRD RD

Apple

NRD RD

Google

Intel

NRD RD

NRD RD

Lucasfilm

Intuit

NRD RD

NRD RD

Pixar

NRD RD

NRD

1.00

0.88

0.17

0.17

0.43

0.73

0.18

0.58

0.50

0.41

0.15

0.04

0.33

0.38

RD

0.88

1.00

0.24

0.27

0.05

0.63

0.47

0.72

0.69

0.61

0.40

0.32

0.48

0.51

NRD

0.17

0.24

1.00

0.99

0.91

0.38

0.65

0.33

0.64

0.68

0.74

0.58

0.48

0.39

RD

0.17

0.27

0.99

1.00

0.90

0.33

0.69

0.37

0.64

0.66

0.83

0.72

0.46

0.40

NRD

0.43

0.05

0.91

0.90

1.00

0.67

0.53

0.13

0.36

0.44

0.81

0.59

0.46

0.28

RD

0.73

0.63

0.38

0.33

0.67

1.00

0.05

0.44

0.20

0.08

0.47

0.04

0.22

0.12

NRD

0.18

0.47

0.65

0.69

0.53

0.05

1.00

0.87

0.64

0.66

0.93

0.98

0.54

0.86

RD

0.58

0.72

0.33

0.37

0.13

0.44

0.87

1.00

0.65

0.62

0.91

0.96

0.48

0.90

NRD

0.50

0.69

0.64

0.64

0.36

0.20

0.64

0.65

1.00

0.94

0.63

0.54

0.55

0.54

RD

0.41

0.61

0.68

0.66

0.44

0.08

0.66

0.62

0.94

1.00

0.78

0.91

0.72

0.62

NRD

0.15

0.40

0.74

0.83

0.81

0.47

0.93

0.91

0.63

0.78

1.00

0.88

0.63

0.83

RD

0.04

0.32

0.58

0.72

0.59

0.04

0.98

0.96

0.54

0.91

0.88

1.00

0.62

0.86

NRD

Adobe

0.33

0.48

0.48

0.46

0.46

0.22

0.54

0.48

0.55

0.72

0.63

0.62

1.00

0.65

RD

0.38

0.51

0.39

0.40

0.28

0.12

0.86

0.90

0.54

0.62

0.83

0.86

0.65

1.00

Apple

Google

Intel

Intuit

Lucasfilm

Pixar

Note

Values

above

0.9

shaded

Source Defendants employee compensation data

Page 33
Supplemental Expert Report

of

Edward

E Leamer PhD

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page36

of

62

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Exhibit 1

Page37

of

62

Case5 11cv 02509LHK

Document424

Exhibit

2

Filed0517 13

Page38

62

of

1

Adobe

Section
First

Job

Title

Year

1

Years
of

1

Section

Total

Emp

Data

2

Level

Years

3

Avg

Emp
4

dlog

Avg

5

dlog

Std

Dev

6

Correlation

Coeff

7

T

Stat

8

2

Section

Change

Correlation

Regression

9

T Stat

Contemp

10

Coeff

11

Lagged

12

3

Section

Coefficients
Revenue

Regression
SJ

13

Emp
14

Contemp
15

Lagged

16

4

T

Section

Net

Stats

Revenue

SJ

17

Emp
18

C

L
19

5

Section

6

Effect

T Stat

Obs

r2

20

21

22

2001

11

170

15

0.27

0.34

0.90

6.07

0.89

5.55

1.18

1.04

0.12

0.02

5.15

6.71

1.77

0.07

2.22

8.15

10

0.98

2001

11

311

28

0.05

0.19

0.89

5.89

0.78

3.55

1.07

1.18

0.09

0.31

0.67

1.38

0.25

0.25

2.25

1.66

10

0.74

2001

11

371

34

0.11

0.16

0.89

5.73

0.79

3.59

0.67

1.33

0.12

0.34

0.66

1.95

0.45

0.36

2.01

1.99

10

0.81

2001

11

29

3

0.16

0.65

0.87

5.37

0.78

3.56

2.67

1.08

0.33

0.48

1.49

1.80

0.80

0.32

3.75

2.24

10

0.79

2001

11

82

7

0.10

0.25

0.85

4.87

0.72

2.97

0.89

1.09

0.46

0.58

0.65

1.99

1.23

0.39

1.97

1.39

10

0.77

2001

11

108

10

0.03

0.40

0.84

4.73

0.82

4.08

0.93

0.88

0.04

0.51

2.43

3.32

0.37

1.38

1.81

3.34

10

0.94

2001

11

96

9

0.12

0.37

0.84

4.65

0.85

4.56

0.80

0.59

0.05

0.84

1.93

2.68

0.45

1.89

1.38

2.66

10

0.95

2001

11

250

23

0.04

0.16

0.84

4.60

0.85

4.47

1.28

0.97

0.08

0.19

2.60

3.59

0.47

0.37

2.25

3.83

10

0.93

2001

11

559

51

0.11

0.20

0.83

4.53

0.88

5.31

0.94

0.80

0.21

0.04

2.27

2.28

1.45

0.08

1.74

3.24

10

0.92

2001

11

93

8

0.11

0.26

0.81

4.19

0.67

2.54

3.21

0.89

0.24

1.55

1.03

0.75

0.30

0.62

4.10

1.49

10

0.63

2001

11

14

1

0.00

0.45

0.80

3.97

0.63

2.29

2.50

0.06

0.51

0.17

0.50

0.04

0.40

0.04

2.57

0.56

10

0.57

2001

11

152

14

0.28

0.15

0.78

3.74

0.72

2.96

0.54

0.65

0.13

0.54

0.98

1.60

0.89

1.07

1.18

1.43

10

0.81

2001

11

202

18

0.06

0.25

0.78

3.74

0.70

2.78

0.68

1.24

0.21

0.34

1.30

4.27

1.40

0.67

1.91

3.24

10

0.92

2001

11

550

50

0.06

0.18

0.78

3.70

0.95

8.29

0.99

0.15

0.06

0.43

2.87

0.54

0.47

0.94

1.14

2.66

10

0.94

2001

11

234

21

0.07

0.22

0.78

3.68

0.73

2.98

0.97

1.14

0.12

0.29

1.56

2.19

0.43

0.48

2.11

2.22

10

0.82

2001

11

273

25

0.17

0.19

0.77

3.60

0.74

3.11

0.34

1.32

0.23

0.33

0.60

2.67

1.59

0.66

1.66

2.77

10

0.86

2001

11

327

30

0.11

0.14

0.74

3.34

0.82

4.00

0.66

0.40

0.11

0.19

1.39

1.12

0.74

0.38

1.06

1.67

10

0.78

2001

11

434

39

0.07

0.18

0.74

3.29

0.65

2.39

0.72

1.09

0.21

0.30

1.29

2.84

1.33

0.56

1.82

2.39

10

0.84

2001

11

196

18

0.13

0.24

0.74

3.27

0.82

4.06

1.23

0.57

0.09

0.02

1.48

1.38

0.29

0.02

1.80

1.87

10

0.78

2001

11

353

32

0.06

0.19

0.73

3.23

0.56

1.91

0.81

1.43

0.17

0.44

1.59

4.09

1.21

0.94

2.23

3.21

10

0.87

2001

11

309

28

0.08

0.23

0.71

3.03

0.61

2.20

0.96

1.13

0.06

0.24

1.27

2.23

0.24

0.34

2.09

1.95

10

0.73

2001

11

94

9

0.08

0.27

0.71

3.03

0.62

2.25

0.65

1.02

0.11

0.58

0.89

2.65

0.49

0.79

1.68

1.74

10

0.83

2001

11

2095

190

0.05

0.13

0.70

2.91

0.69

2.68

0.26

0.49

0.12

0.35

0.60

1.35

0.88

0.79

0.75

1.25

10

0.72

2001

11

514

47

0.08

0.22

0.70

2.90

0.63

2.27

0.71

0.97

0.08

0.45

0.91

2.30

0.29

0.57

1.68

1.66

10

0.77

2001

11

35

3

0.00

0.32

0.69

2.90

0.53

1.75

0.58

1.09

0.15

0.15

0.45

2.12

0.47

0.09

1.67

1.05

10

0.81

2001

11

215

20

0.07

0.53

0.69

2.88

0.46

1.48

0.35

1.26

0.07

0.47

0.51

3.49

0.39

0.69

1.61

1.88

10

0.82

2001

11

496

45

0.05

0.20

0.67

2.74

0.75

3.18

0.08

0.47

0.14

0.56

0.17

1.29

0.89

0.91

0.56

0.87

10

0.83

2001

11

466

42

0.06

0.11

0.67

2.74

0.69

2.71

0.27

0.62

0.10

0.27

0.49

1.62

0.59

0.48

0.89

1.33

10

0.71

2001

11

234

21

0.09

0.33

0.67

2.71

0.77

3.39

0.10

0.27

0.17

1.23

0.21

1.12

1.01

2.21

0.38

0.63

10

0.87

2001

11

1441

131

0.06

0.19

0.65

2.55

0.48

1.56

0.24

0.71

0.11

0.54

0.35

1.51

0.58

0.89

0.94

0.98

10

0.61

2001

11

302

27

0.00

0.21

0.64

2.49

0.91

6.03

0.62

0.10

0.17

0.94

2.20

0.67

1.72

2.57

0.72

2.18

10

0.95

2001

11

222

20

0.09

0.15

0.63

2.44

0.62

2.22

0.05

0.45

0.11

0.75

0.07

1.04

0.51

0.95

0.50

0.52

10

0.70

2001

11

975

89

0.12

0.23

0.63

2.42

0.48

1.55

0.24

0.49

0.00

0.40

0.39

1.05

0.01

0.71

0.73

0.86

10

0.42

2001

11

2041

186

0.05

0.20

0.61

2.33

0.57

1.94

0.07

0.43

0.14

0.55

0.14

1.04

0.80

1.04

0.50

0.67

10

0.62

2001

11

56

0.03

0.54

0.61

2.32

0.52

1.70

0.27

1.04

0.08

1.06

0.36

2.96

0.39

1.55

1.30

1.43

10

0.83

2001

11

2064

0.05

0.08

0.61

2.29

0.52

1.71

0.07

0.44

0.13

0.65

0.14

1.13

0.82

1.29

0.37

0.52

10

0.66

2001

11

100

9

0.09

0.31

0.60

2.27

0.61

2.20

1.92

0.91

0.00

3.12

1.44

1.96

0.00

2.95

2.83

2.36

10

0.86

2001

11

1008

92

0.06

0.27

0.59

2.17

0.56

1.91

0.36

0.56

0.26

0.29

0.57

1.18

1.41

0.48

0.91

1.09

10

0.62

2001

11

41

0.59

0.58

2.11

0.34

1.02

0.41

1.61

0.19

0.56

0.42

2.35

0.55

0.42

2.01

1.37

10

0.71

11

66

4
6

0.00

2001

0.06

0.72

0.51

1.77

0.37

1.13

1.62

0.86

0.57

1.57

4.28

3.06

4.84

5.82

2.48

3.98

10

0.91

5
188

Case5 11cv 02509LHK

Document424

Exhibit

2

Filed0517 13

Page39

62

of

1

Adobe

Section
First

Job

Title

Year

1

Years
of

1

Section

Total

Emp

Data

2

Level

Years

Avg

3

Emp
4

dlog

Avg

5

dlog

Std

Dev

6

Correlation

Coeff

7

T

Stat

8

2

Section

Change

Correlation

Regression

9

T Stat

Contemp

10

Coeff

11

Lagged

12

3

Section

Coefficients
Revenue

Regression
SJ

13

Emp
14

Contemp
15

Lagged

16

4

T

Section

Net

Stats

Revenue

SJ

17

Emp
18

C

L
19

5

Section

6

Effect

T Stat

Obs

r2

20

21

22

2005

7

22

3

0.18

0.41

0.76

2.64

0.15

0.31

0.14

0.93

0.38

0.36

0.11

1.48

0.68

0.21

1.07

0.60

6

0.91

2001

7

42

6

0.27

0.76

0.57

1.56

0.39

0.84

3.13

2.20

0.57

3.68

2.63

2.79

1.65

2.92

0.93

1.11

6

0.93

2001

7

88

13

0.41

0.33

0.53

1.38

0.38

0.82

3.36

5.49

1.61

7.47

4.12

6.77

4.51

5.53

2.13

6

1.00

2001

7

17

2

0.00

0.36

0.48

1.21

0.93

4.88

0.58

0.42

0.13

0.77

0.54

0.84

0.54

0.89

1.00

0.71

6

0.95

2005

7

93

13

0.00

0.27

0.40

0.98

0.97

7.56

1.30

0.10

0.07

0.02

2.06

0.28

0.24

0.03

1.40

1.76

6

0.94

2005

7

59

8

0.05

0.36

0.08

0.18

0.52

1.21

0.49

0.70

0.24

0.26

0.34

0.76

0.40

0.13

1.19

0.61

6

0.73

2001

6

46

8

0.14

0.21

0.98

0.90

3.49

2001

6

25

4

0.36

0.95

0.97

8.18

0.86

2.98

2001

6

19

3

0.06

0.45

0.96

7.28

0.93

4.41

2001

6

87

15

0.03

0.12

0.96

6.72

0.83

2.55

2001

6

13

2

0.28

1.05

0.94

5.50

0.94

4.92

2001

6

89

15

0.11

0.43

0.94

5.29

0.82

2.47

2001

6

108

18

0.01

0.23

0.93

5.23

0.74

1.90

2001

6

20

3

0.00

0.20

0.93

5.11

0.78

2.17

2001

6

16

3

0.06

0.70

0.92

4.77

0.58

1.23

2001

6

33

6

0.08

0.33

0.92

4.62

0.66

1.52

2001

6

22

4

0.03

0.74

0.89

3.99

0.94

4.80

2001

6

23

4

0.22

0.49

0.89

3.90

0.67

1.54

2001

6

35

6

0.09

0.26

0.89

3.87

0.91

3.90

2001

6

57

10

0.06

0.53

0.88

3.77

0.47

0.91

2001

6

10

2

0.22

0.32

0.88

3.74

0.50

1.00

2001

6

24

4

0.25

1.15

0.88

3.70

0.83

2.11

2001

6

21

4

0.36

0.59

0.88

3.66

0.49

0.97

2001

6

92

15

0.19

0.16

0.87

3.60

0.78

2.16

2001

6

68

11

0.00

0.21

0.86

3.44

0.66

1.51

2001

6

13

2

0.00

0.29

0.86

3.43

0.59

1.28

2001

6

27

5

0.42

0.63

0.86

3.38

0.74

1.92

2001

6

8

1

0.00

0.49

0.85

3.28

0.93

4.31

2001

6

15

3

0.08

0.34

0.85

3.18

0.27

0.49

2001

6

26

4

0.04

0.41

0.82

2.84

0.76

2.03

2006

6

7

1

0.14

0.31

0.81

2.81

0.85

2.85

2001

6

18

3

0.00

0.51

0.67

1.79

0.43

0.82

2001

6

105

18

0.04

0.36

0.66

1.74

0.68

1.59

2006

6

27

5

0.14

0.46

0.62

1.57

0.61

1.34

2006

6

19

3

0.08

0.52

0.61

1.55

0.54

1.11

2001

6

15

3

0.14

0.90

0.61

1.54

0.14

0.24

2001

6

12

2

0.22

0.32

0.57

1.39

0.76

2.05

2001

6

15

3

0.22

0.32

0.57

1.38

0.56

1.17

2006

6

19

0.53

0.34

0.72

0.21

0.38

6

6

3
1

0.28

2004

0.00

0.00

0.13

0.26

0.28

0.50

10.31

10.60

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Exhibit 2

Page40

of

62

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page41

62

of

Exhibit 2

Apple

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

Coeff

2

Section

Change

Correlation

Correlation

T

Coeff

Stat

3

Section

Regression

Stat

Contemp

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

Section

T Stats

Net

Revenue

SJ Emp

C

5

Section

T Stat

L

6

Effect
r2

11

294

0.98

13.53

0.74

3.11

0.80

0.04

0.34

0.06

1.64

0.05

0.81

0.13

0.84

0.76

0.71

11

501

0.98

13.42

0.87

4.91

2.46

1.09

0.70

0.67

5.33

1.71

1.82

1.18

3.56

4.85

0.92

11

229

0.98

13.33

0.65

2.41

1.15

0.97

0.09

0.08

2.58

1.52

0.26

0.19

2.12

2.15

0.73

11

169

0.97

12.72

0.70

2.79

1.29

1.49

0.57

0.28

2.17

1.67

1.00

0.46

2.78

2.20

0.72

11

352

0.95

9.16

0.71

2.82

0.92

0.22

0.76

0.16

1.56

0.39

1.55

0.26

0.71

0.72

0.78

11

189

0.93

7.38

0.84

4.39

1.68

0.36

0.20

0.87

1.81

0.38

0.26

0.87

2.04

1.39

0.82

11

428

0.91

6.72

0.65

2.45

0.51

4.63

2.48

1.62

0.53

2.82

2.28

1.57

5.14

2.93

0.82

11

156

0.88

5.54

0.39

1.21

0.71

0.25

0.28

0.62

0.95

0.38

0.40

0.67

0.96

0.77

0.29

11

118

0.68

2.82

0.36

1.09

0.58

0.17

0.11

0.23

0.86

0.31

0.16

0.24

0.75

0.70

0.16

11

686

0.49

1.69

0.43

1.33

0.66

0.47

0.15

0.49

0.68

0.60

0.18

0.40

1.13

0.73

0.52

11

58

0.50

1.71

0.07

0.20

0.03

0.11

0.21

0.27

0.05

0.28

0.49

0.47

0.09

0.11

0.10

10

82

0.67

2.52

0.03

0.08

0.38

0.08

0.18

0.01

0.39

0.10

0.22

0.01

0.30

0.19

0.34

10

184

0.81

3.84

0.25

0.68

0.17

0.08

0.18

0.91

0.20

0.11

0.24

0.81

0.09

0.07

0.40

10

110

0.81

3.93

0.71

2.64

0.69

0.07

0.04

0.53

2.98

0.36

0.18

1.86

0.76

2.06

0.75

10

66

0.89

5.57

0.04

0.11

0.14

0.06

0.06

0.20

1.03

0.53

0.47

1.12

0.20

0.92

0.36

0.85

4.33

0.55

1.59

0.43

0.03

0.14

0.95

1.37

0.14

0.54

1.36

0.39

0.79

0.83

11.69

0.59

1.27

1.84

3.27

2.40

1.69

9

116

8

44

0.98

8

35

0.97

9.97

0.78

2.48

0.30

0.21

1.02

0.21

1.13

0.37

3.49

0.93

0.50

0.73

0.99

8

19

0.76

2.89

0.62

1.78

0.16

0.16

0.02

0.78

0.78

0.97

0.13

1.91

0.00

0.01

0.86

8

52

0.82

3.57

0.02

0.05

0.14

0.08

0.13

0.07

0.50

0.28

0.51

0.36

0.22

0.40

0.57

8

13

0.96

7.90

0.24

0.55

0.09

0.05

0.03

0.22

0.84

0.50

0.27

0.69

0.14

0.78

0.51

7

71

0.99

22.21

0.95

5.95

0.54

0.46

0.07

0.06

1.39

0.22

0.15

0.04

0.08

0.03

0.94

7

193

0.99

20.45

0.95

6.20

1.49

1.49

0.41

0.82

12.36

3.86

2.99

1.89

2.98

6.80

1.00

7

626

0.99

16.77

0.94

5.77

1.41

1.40

0.29

0.07

30.92

4.57

3.71

0.27

2.82

8.34

1.00

7

184

0.99

16.70

0.96

6.91

1.16

1.48

0.31

0.23

3.69

0.99

0.69

0.27

2.64

1.81

0.97

0.99

14.96

0.92

4.55

0.88

0.60

0.16

0.65

10.23

3.85

1.64

3.23

1.48

7.27

0.99

0.99

13.76

0.81

2.81

0.24

0.38

0.08

0.22

0.48

0.29

0.14

0.12

0.14

0.08

0.80

7

7

2566

29

7

253

0.98

12.12

0.92

4.72

0.76

1.16

0.20

0.64

1.85

1.01

0.73

0.66

1.92

1.84

0.95

7

130

0.98

10.75

0.89

3.94

0.47

5.06

1.65

5.63

0.64

1.93

1.97

1.78

4.59

2.36

0.97

7

447

0.98

10.68

0.95

6.15

1.48

0.65

0.02

0.45

2.89

0.47

0.04

0.35

2.12

1.64

0.96

7

244

0.98

10.66

0.88

3.63

0.18

4.02

1.70

0.93

0.73

3.21

3.80

7.34

4.20

2.81

1.00

7

125

0.98

9.93

0.86

3.39

0.99

1.14

0.05

0.09

4.26

3.10

0.20

0.19

2.14

5.47

0.98

0.98

9.91

0.93

4.96

0.85

0.41

0.34

1.08

5.64

1.91

2.09

2.89

1.26

4.61

0.99

0.97

9.77

0.81

2.81

1.59

2.35

1.09

2.20

5.11

4.37

4.08

2.80

3.94

6.55

0.98

7

7

1364

54

7

236

0.97

9.58

0.97

7.42

0.99

0.57

0.28

0.18

2.55

1.16

0.76

0.24

1.56

3.63

0.97

7

475

0.97

9.33

0.84

3.04

0.55

0.80

0.42

1.16

2.01

1.71

1.67

1.34

1.35

2.55

0.95

0.97

9.17

0.81

2.81

0.66

0.37

0.03

0.87

9.39

3.50

0.50

5.68

1.03

6.50

0.99

4.22

3.00

79.73

1.00

7.66

1.52

14.05

1.00

7

1304

7

110

0.97

8.72

0.95

6.06

1.93

1.07

0.23

0.24

108.02

31.38

7

902

0.97

8.62

0.82

2.84

0.83

0.68

0.49

1.09

13.99

9.36

14.63
7.98

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page42

62

of

Exhibit 2

Apple

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data
7

17

T

Coeff

2

Section

Change

Correlation

Correlation

T

Coeff

Stat

3

Section

Regression

Stat

Contemp

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

Section

T Stats

Net

Revenue

0.95

7.08

0.71

2.01

1.88

6.66

3.36

7.09

7.10

9.22

8.26

SJ Emp
6.61

16.05

C

5

Section

T Stat

L

6

Effect
r2

10.88

1.00

0.75

7.07

1.00

8.54

7

127

0.95

6.94

0.52

1.21

0.56

0.19

0.28

1.66

15.61

2.26

7.87

7

142

0.95

6.80

0.83

2.99

0.30

3.49

0.40

0.56

0.28

2.08

0.91

0.41

3.19

2.94

0.95

1.09

2.55

0.84

2.00

2.49

4.60

2.54

2.18

3.64

5.17

0.98

2.37

0.57

0.11

0.28

3.89

0.73

0.43

0.38

1.80

3.15

0.98

7

63

0.95

6.73

0.69

1.92

7

45

0.95

6.73

0.99

12.42

7

98

0.95

6.52

0.84

3.11

0.42

0.03

0.15

0.86

2.29

0.16

0.60

1.47

0.39

1.18

0.93

7

70

0.94

6.46

0.88

3.72

1.03

3.36

0.26

1.34

1.02

0.65

0.33

0.29

4.39

1.02

0.95

0.94

6.42

0.96

7.04

1.85

0.66

0.02

0.43

20.57

4.80

0.28

1.92

2.51

19.20

1.00

0.94

6.33

0.60

1.52

0.75

0.73

0.18

0.36

3.05

2.18

0.96

0.70

1.48

2.83

0.92

7

7

182
2915

7

134

0.94

6.30

0.66

1.76

0.94

1.02

0.16

0.07

8.01

7.07

1.52

0.25

1.97

9.04

0.99

7

143

0.94

6.27

0.48

1.10

0.38

0.26

0.73

1.64

0.87

0.46

1.94

1.39

0.64

0.68

0.84

7

476

0.94

6.23

0.91

4.31

3.20

2.66

1.18

5.55

2.00

1.31

1.16

1.44

0.53

0.75

0.96

0.94

6.18

0.79

2.54

1.14

0.91

0.12

0.64

3.07

1.95

0.41

0.78

2.05

3.12

0.98

7

53

7

275

0.94

6.09

0.70

1.97

0.82

0.80

0.45

1.06

2.39

1.55

1.68

1.39

1.62

2.24

0.97

7

255

0.93

5.78

0.74

2.21

0.07

2.18

0.57

1.09

0.15

4.59

2.06

1.39

2.11

4.69

0.98

7

300

0.93

5.69

0.38

0.82

0.33

0.33

0.09

0.42

1.51

1.22

0.67

1.12

0.66

1.43

0.82

7

125

0.93

5.69

0.79

2.56

0.64

1.88

0.06

0.58

5.01

16.56

0.79

2.97

2.52

18.16

1.00

7

262

0.93

5.65

0.51

1.18

0.99

1.54

0.46

0.24

4.29

4.47

2.58

0.49

2.53

5.03

0.97

0.93

5.63

0.72

2.10

1.20

1.08

0.14

0.10

2.30

1.36

0.24

0.08

2.28

2.10

0.97

0.93

5.58

0.27

0.57

0.71

0.94

0.29

1.78

0.41

0.33

0.52

1.46

1.65

0.36

0.76

6.30

4.69

2.75

12.46

1.00

7

7

16
115

7

33

0.93

5.56

0.55

1.31

1.06

1.69

0.48

0.89

11.73

10.86

7

16

0.93

5.55

0.47

1.06

2.57

3.07

1.01

2.89

2.51

2.15

1.27

1.18

5.64

2.42

0.92

7

35

0.93

5.46

0.68

1.85

0.43

0.40

0.43

1.40

0.92

0.30

0.85

1.10

0.83

0.53

0.92

0.92

5.42

0.84

3.04

0.57

1.74

0.21

0.65

0.73

2.15

0.46

0.55

2.30

2.76

0.95

7

297

7

57

0.92

5.39

0.72

2.05

0.69

0.70

0.36

0.74

2.04

2.46

0.95

0.86

1.39

2.85

0.94

7

58

0.92

5.35

0.78

2.48

0.81

0.46

0.29

0.50

3.21

2.06

0.77

0.78

1.28

3.10

0.94

7

26

0.92

5.30

0.67

1.80

2.23

2.43

1.17

0.57

5.76

2.33

1.86

0.37

4.66

3.32

1.00

58.99

1.00

7

115

0.92

5.30

0.64

1.68

0.86

0.53

0.05

1.73

81.85

34.93

6.57

83.66

7

103

0.92

5.23

0.35

0.74

0.71

2.91

1.10

0.68

1.67

3.08

2.22

0.72

3.62

3.03

0.94

1.39

7

35

0.92

5.21

0.59

1.45

0.67

4.66

1.96

0.59

1.56

5.68

4.15

0.64

5.33

5.82

0.99

7

49

0.92

5.14

0.67

1.79

1.20

0.72

0.03

2.50

2.41

0.57

0.03

1.91

1.92

1.15

0.98

7

23

0.92

5.12

0.89

3.94

1.50

0.38

0.73

0.15

3.16

0.60

1.79

0.15

1.12

1.44

0.98

0.91

5.03

0.24

0.50

0.05

0.05

0.05

0.41

0.10

0.09

0.19

0.45

0.01

0.01

0.23

0.31

2.38

6.99

3.91

0.96

9.56

1.00

35.05

1.00

0.84

0.79

24.05

1.00

225.62

1.00

7

431

7

21

0.91

4.94

0.54

1.30

3.18

3.81

0.09

4.43

4.28

3.52

7

64

0.91

4.93

0.33

0.71

0.14

0.85

0.65

1.56

2.39

11.13

11.42

6.65

0.99

7

56

0.91

4.86

0.93

4.90

3.28

0.05

0.48

3.16

26.16

0.30

13.49

7.06

3.23

7

14

0.91

4.86

0.40

0.86

0.07

0.01

0.16

0.43

1.14

0.23

1.66

0.08

7

59

0.91

4.83

0.88

3.68

1.77

1.31

0.18

0.90

13.53

9.61

7

48

0.90

4.69

0.20

0.42

0.20

0.71

0.09

0.37

102.47

285.17

1.50

1.45

64.33

2.78

73.80

3.09

0.91

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page43

62

of

Exhibit 2

Apple

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

Coeff

2

Section

Change

Correlation

Correlation

T

Coeff

Stat

3

Section

Regression

Stat

Contemp

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

Section

T Stats

Net

Revenue

SJ Emp

C

5

Section

T Stat

L

6

Effect
r2

7

37

0.88

4.25

0.04

0.07

0.57

0.88

0.53

0.36

1.77

2.44

2.49

0.57

1.45

2.29

0.89

7

34

0.88

4.25

0.15

0.30

1.13

2.90

1.26

0.78

2.87

5.65

3.68

0.93

4.03

5.04

0.98

7

8

0.88

4.20

0.89

3.94

1.47

0.70

0.91

1.65

9.23

2.78

5.87

4.83

0.78

2.62

1.00

0.88

4.17

0.40

0.87

0.34

0.62

0.43

0.72

1.28

1.65

1.75

1.23

0.96

1.71

0.99

7

103

7

7

0.88

4.15

0.72

2.05

0.53

0.22

0.32

0.27

3.42

1.74

1.18

0.56

0.75

3.01

0.94

7

8

0.88

4.11

0.04

0.09

0.44

1.15

0.14

0.78

0.71

1.55

0.31

0.53

1.58

1.29

0.81

7

28

0.88

4.08

0.45

1.02

0.07

3.01

0.73

2.35

0.56

12.67

6.31

6.82

3.09

11.64

1.00

7

61

0.88

4.08

0.26

0.55

1.31

2.69

1.24

1.51

3.08

3.00

2.32

2.70

4.00

3.06

0.99

7

25

0.87

4.01

0.59

1.45

0.28

3.71

1.43

0.39

8.63

82.49

54.77

6.26

3.99

78.44

1.00

7

7

0.87

3.98

0.26

0.53

1.98

2.62

1.42

5.06

1.68

2.14

1.30

1.91

4.61

2.14

0.98

0.87

3.94

0.85

3.21

3.43

3.62

0.07

5.76

2.07

1.57

0.14

1.48

0.19

0.18

0.94

0.87

3.94

0.61

1.53

0.61

1.04

0.29

1.97

1.40

2.07

0.84

1.88

1.64

2.22

0.93

0.87

3.93

0.50

1.16

0.27

0.05

0.31

0.87

0.23

0.04

0.66

0.51

0.22

0.10

0.51

0.87

3.91

0.49

1.14

0.28

2.39

0.62

0.66

0.16

0.99

0.48

0.16

2.11

0.87

0.81

2.27

0.94

7

7

7

7

501

74
192

11

7

116

0.87

3.89

0.21

0.43

6.50

7.89

2.48

6.52

2.32

2.22

1.95

1.50

14.39

7

239

0.87

3.89

0.89

3.90

0.95

0.13

0.56

0.89

1.43

0.16

1.08

0.59

0.82

0.81

0.90

7

10

0.86

3.83

0.54

1.30

4.35

6.24

1.52

7.36

0.67

0.87

0.63

0.57

1.89

0.66

0.80

7

44

0.86

3.78

0.52

1.22

0.32

0.27

0.00

0.96

0.20

0.17

0.00

0.74

0.59

0.21

0.54

7

21

0.86

3.69

0.69

1.91

0.77

0.40

0.84

1.36

0.94

0.44

1.35

0.74

0.37

0.30

0.95

7

17

0.85

3.65

0.68

1.84

1.99

1.43

0.04

0.81

2.93

1.69

0.07

0.45

3.42

2.63

0.97

0.85

3.60

0.92

4.56

1.94

0.26

0.17

0.60

0.89

0.17

0.12

0.17

1.68

1.12

0.84

7

563

7

12

0.85

3.58

0.06

0.12

0.12

0.13

0.26

0.46

0.51

0.68

1.37

0.74

0.25

0.65

0.79

7

57

0.85

3.58

0.46

1.03

0.26

1.45

0.06

1.52

0.14

1.18

0.06

0.40

1.19

0.60

0.89

0.85

3.57

0.90

4.16

1.96

0.40

0.23

2.66

15.41

5.76

2.47

9.44

1.55

13.27

1.00

0.85

3.55

0.04

0.07

0.55

0.93

0.28

2.78

0.76

0.91

0.66

3.50

1.48

0.86

0.95

7

7

145

33

7

131

0.85

3.55

0.76

2.36

0.54

0.17

0.73

1.81

1.90

0.72

2.37

2.35

0.71

1.63

0.96

7

267

0.84

3.52

0.16

0.32

0.22

0.30

1.27

2.14

0.14

0.19

0.35

0.20

0.51

0.17

0.51

7

47

0.84

3.43

0.29

0.60

0.83

1.09

0.45

1.22

1.10

1.76

0.48

0.69

1.91

1.62

0.85

7

60

0.84

3.42

0.52

1.21

0.83

0.25

0.30

0.36

0.54

0.17

0.26

0.29

1.09

0.41

0.36

7

8

0.84

3.40

0.06

0.12

0.13

3.20

1.30

2.42

0.36

3.70

2.62

2.29

3.33

3.00

0.97

7

50

0.83

3.35

0.61

1.56

0.65

0.05

0.93

1.56

4.31

0.32

7.83

4.65

0.70

2.62

1.00

7

57

0.83

3.34

0.11

0.22

0.25

0.75

0.33

0.60

0.87

2.96

1.13

0.67

1.00

2.18

0.95

7

20

0.83

3.33

0.35

0.75

0.24

0.46

0.59

1.46

0.59

1.17

1.77

1.65

0.70

1.04

0.99

7

20

0.83

3.32

0.38

0.83

0.34

1.47

0.20

0.34

2.79

7.80

1.91

1.02

1.14

3.94

1.00

7

40

0.82

3.24

0.94

5.74

1.96

0.82

0.43

0.51

3.60

1.74

1.46

0.51

1.14

2.01

0.98

0.82

3.24

0.91

4.27

1.43

0.33

0.57

0.59

1.18

0.30

0.81

0.26

1.11

0.79

0.89

1.37

5.78

2.74

18.75

0.55

1.16

1.17

1.69

7.16

0.96

0.99

7

144

7

23

0.82

3.21

0.55

1.31

7

72

0.82

3.17

0.01

0.02

0.59

0.65

1.04

2.39

0.45

0.50

0.46

0.44

1.24

0.49

0.22

7

47

0.81

3.07

0.71

2.01

1.22

0.50

0.87

1.01

2.88

1.31

2.57

1.08

1.72

2.53

0.98

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page44

62

of

Exhibit 2

Apple

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

Coeff

2

Section

Change

Correlation

Correlation

T

Coeff

Stat

3

Section

Regression

Stat

Contemp

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

Section

T Stats

Net

Revenue

SJ Emp

7

26

0.70

2.22

0.23

0.48

0.23

0.43

0.86

0.85

15.24

28.90

49.37

7

25

0.70

2.20

0.68

1.86

0.94

0.69

0.33

3.50

0.89

0.45

0.39

1.53

7

38

0.70

2.20

0.79

2.56

9.17

2.35

7.19

19.15

1.03

2.58

0.73

0.53

7

18

0.66

1.95

0.11

0.22

2.32

2.16

0.46

7.71

1.39

1.67

0.35

1.99

7

58

0.66

1.95

0.07

0.15

0.76

0.61

1.32

2.62

1.44

1.90

2.40

7

26

0.65

1.90

0.43

0.95

1.80

1.36

0.95

0.78

1.91

1.85

7

13

0.65

1.90

0.51

1.18

1.56

2.39

0.40

6.21

0.64

7

51

0.64

1.88

0.23

0.47

1.80

1.79

0.28

0.82

7

14

0.64

1.87

0.38

0.82

0.56

0.52

0.89

7

57

0.64

1.86

0.03

0.05

0.09

0.08

7

11

0.63

1.82

0.45

1.01

1.68

7

24

0.63

1.80

0.57

1.40

0.62

1.79

0.04

7

127

12.29

C

5

Section

T Stat

L

6

Effect
r2

25.59

1.00

0.82

0.88

1.20

0.97

4.48

1.69

0.93

1.36

0.14

0.20

0.97

0.84

0.32

3.16

2.12

0.83

1.99

0.29

1.79

0.83

0.26

0.97

1.11

1.29

0.21

0.21

3.59

1.32

0.74

4.00

1.07

1.02

2.18

3.39

1.08

1.20

0.97

1.16

3.51

2.94

2.71

47.78

0.01

0.26

1.00

1.26

0.17

1.18

3.40

3.03

0.42

1.01

2.93

3.65

0.97

0.12

7.51

4.87

14.39

0.13

2.03

2.22

3.54

7.63

1.69

0.99

0.08

2.05

1.96

4.08

9.17

7.13

7.37

9.16

8.99

4.01

7.51

0.99

40.82

0.66

1.62

11.52

7

45

0.62

1.79

0.82

2.90

1.18

0.46

0.62

0.77

1.08

0.92

0.50

0.58

1.64

1.07

0.97

7

36

0.58

1.58

0.86

3.38

3.09

0.55

1.14

3.47

0.92

0.56

0.39

0.63

3.64

0.88

0.87

7

52

0.57

1.57

0.56

1.34

0.91

0.24

2.01

5.19

0.41

0.17

1.13

1.29

0.67

0.21

0.91

0.56

1.51

0.25

0.51

0.93

0.88

0.89

1.03

2.28

2.86

1.87

1.16

1.81

2.82

0.94

7

137

7

18

0.55

1.49

0.33

0.69

0.11

0.48

2.73

0.70

0.25

1.33

3.40

0.55

0.59

0.78

0.98

7

13

0.55

1.48

0.52

1.23

0.42

1.07

2.09

2.76

0.47

1.46

3.04

1.39

0.65

0.46

0.97

7

59

0.55

1.46

0.06

0.12

0.37

0.17

0.75

5.12

0.25

0.10

0.87

2.95

0.54

0.18

0.93

7

16

0.54

1.45

0.47

1.07

3.59

2.10

0.38

3.17

2.25

1.44

0.44

0.55

5.69

1.95

0.93

7

34

0.54

1.42

0.41

0.90

0.50

0.48

1.73

2.69

1.25

1.52

5.22

2.97

0.01

0.02

0.98

7

35

0.53

1.39

0.50

1.17

0.35

0.64

1.85

0.64

0.37

0.92

2.26

0.30

0.30

0.20

0.94

7

41

0.53

1.38

0.52

1.21

0.82

0.14

0.66

1.97

1.45

0.32

0.93

1.55

0.96

1.08

0.86

7

46

0.52

1.36

0.33

0.69

1.08

1.05

0.12

0.61

6.60

8.54

0.68

1.70

2.13

8.30

1.00

7

15

0.52

1.35

0.73

2.16

0.40

0.56

0.89

2.38

0.20

0.51

0.62

0.68

0.96

0.38

0.84

0.52

1.35

0.00

0.00

0.17

0.16

0.08

0.05

2.19

2.67

0.66

0.17

0.33

2.68

0.95

7

646

7

14

0.51

1.33

0.20

0.41

0.55

0.31

1.05

0.73

0.34

0.25

0.34

0.15

0.86

0.37

0.55

7

47

0.51

1.31

0.96

6.64

1.90

0.37

0.00

0.63

3.25

1.24

0.00

0.50

1.53

2.14

0.97

7

27

0.50

1.30

0.11

0.23

1.00

1.69

0.29

2.35

5.14

10.67

1.89

4.56

0.69

2.25

1.00

7

17

0.49

1.25

0.19

0.38

0.61

0.30

1.50

1.86

1.24

0.79

3.44

1.64

0.91

1.17

0.98

7

13

0.49

1.24

0.72

2.07

2.54

2.26

1.08

0.56

0.86

0.91

0.51

0.10

0.28

0.06

0.95

7

63

0.47

1.20

0.14

0.29

0.42

0.43

1.07

1.20

0.49

0.68

1.03

0.38

0.01

0.01

0.91

7

85

0.47

1.18

0.43

0.96

0.01

0.28

1.09

0.67

0.01

0.41

0.72

0.15

0.27

0.14

0.90

7

60

0.45

1.11

0.74

2.18

0.54

0.10

0.20

0.50

1.72

0.44

0.46

0.35

0.64

1.31

0.88

7

19

0.44

1.10

0.46

1.03

0.96

1.03

1.13

5.03

2.16

3.27

4.56

5.89

1.98

2.79

1.00

7

10

0.44

1.08

0.78

2.50

5.10

0.41

0.38

7.95

0.71

0.11

0.12

0.81

4.69

0.46

0.79

7

69

0.42

1.04

0.10

0.20

0.58

2.48

1.20

12.14

0.30

0.89

1.11

0.64

3.06

1.33

0.87

7

36

0.42

1.03

0.34

0.73

0.26

0.23

0.76

0.15

1.39

1.93

2.82

0.32

0.03

0.10

0.94

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page45

62

of

Exhibit 2

Apple

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data
7

7

T

Coeff

29
117

2

Section

Change

Correlation

Correlation

T

Coeff

Stat

3

Section

Regression

Stat

Contemp

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

Section

T Stats

Net

Revenue

SJ Emp

C

5

Section

T Stat

L

6

Effect
r2

0.08

0.18

0.33

0.70

0.04

0.09

1.00

2.10

0.04

0.12

0.78

0.64

0.13

0.08

0.73

0.04

0.08

0.26

0.55

0.56

1.36

6.15

1.05

1.15

3.45

3.63

0.97

0.80

1.44

0.96

7

26

0.04

0.08

0.21

0.43

0.76

0.43

1.14

2.09

0.93

0.73

1.69

1.20

0.34

0.26

0.99

7

22

0.04

0.10

0.17

0.34

4.02

1.91

5.35

23.44

2.41

1.95

2.23

2.84

5.94

2.26

0.97

7

31

0.07

0.16

0.29

0.62

0.47

1.28

2.00

1.97

0.26

0.93

1.01

0.32

1.74

0.61

0.64

7

11

0.27

0.63

0.23

0.48

0.75

0.14

0.01

0.87

0.25

0.05

0.00

0.37

0.89

0.17

0.21

7

46

0.28

0.66

0.02

0.03

2.17

1.69

6.68

6.27

1.26

1.10

1.77

1.60

0.48

0.21

0.82

7

52

0.36

0.87

0.37

0.79

1.19

0.84

0.81

2.05

2.75

2.07

1.17

2.10

2.04

2.78

0.95

7

50

0.43

1.06

0.96

6.86

0.30

0.06

0.07

0.09

6.12

1.64

0.89

0.54

0.24

3.24

0.99

7

49

0.48

1.23

0.27

0.57

0.03

0.11

0.46

1.13

0.06

0.26

0.43

0.48

0.14

0.18

0.55

0.49

1.25

0.44

0.97

0.12

0.34

0.76

0.70

1.22

3.94

3.94

2.00

0.22

1.43

0.96

7

166

7

36

0.50

1.29

0.05

0.10

1.28

3.22

5.96

8.31

0.99

1.06

0.97

1.00

4.50

1.10

0.61

7

21

0.54

1.42

0.80

2.66

1.42

0.36

0.68

1.28

6.57

1.85

1.97

2.37

1.77

4.71

0.99

7

59

0.62

1.79

0.31

0.65

0.43

0.52

0.51

0.18

0.46

0.58

0.70

0.24

0.94

0.59

0.48

7

40

0.65

1.92

0.35

0.74

0.75

0.85

0.63

0.30

0.43

0.46

0.41

0.22

1.61

0.50

0.32

6

16

0.98

9.32

0.93

4.31

6

19

0.96

7.34

0.85

2.85

6

54

0.96

7.16

0.89

3.46

6

48

0.93

4.91

0.94

4.62

6

44

0.87

3.58

0.64

1.18

6

20

0.87

3.48

0.45

0.72

6

73

0.85

3.24

0.41

0.78

6

19

0.77

2.41

0.51

1.03

6

6

0.76

2.35

0.46

0.91

6

15

0.76

2.31

0.90

3.49

6

24

0.75

2.27

0.08

0.12

6

6

0.75

2.26

0.53

1.07

6

57

0.73

2.13

0.47

0.92

6

8

0.72

2.05

0.36

0.55

6

10

0.71

2.04

0.55

1.14

6

6

0.67

1.81

0.59

1.26

6

6

0.63

1.61

0.81

1.95

6

8

0.63

1.61

0.82

2.00

6

11

0.60

1.49

0.83

2.59

6

19

0.59

1.45

0.05

0.08

6

12

0.48

1.08

0.06

0.09

6

19

0.47

1.07

0.04

0.07

18

0.42

0.93

0.61

1.09

0.42

0.92

0.55

1.14

6

6

166

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page46

62

of

Exhibit 2

Apple

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

Coeff

2

Section

Change

Correlation

Correlation

T

Coeff

Stat

Stat

6

17

0.16

0.32

0.07

0.13

6

16

0.29

0.60

0.78

2.16

6

6

0.30

0.62

0.55

1.13

6

40

0.31

0.65

0.11

0.19

6

6

0.45

1.02

0.84

2.67

0.65

1.70

0.32

0.59

6

1398

3

Regression

6

15

0.76

2.36

0.93

4.48

6

19

0.85

3.22

0.43

0.83

Contemp

Lagged

Revenue

Section

Coefficients

Regression
SJ

Emp

Contemp

Lagged

4

Section

T Stats
Revenue

Net
SJ Emp

C

L

5

Section

Effect

T Stat

r2

6

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page47

of

62

Exhibit 2
Google

Section
Years
Job

Title

of

Data

1

Section

Total

Emp

Years

Level

Correlation

T

Coeff

2
Change

Section

T

Coeff

Stat

Regression

Correlation

Contemp

Stat

Lagged

3

Section
Regression

Coefficients

Revenue

SJ

Emp

Contemp

Lagged

4

T

Section

5

Section

6

Net Effect

Stats

Revenue

SJ

Emp

C

T

L

Stat

r2

0.94

8.15

0.89

5.63

0.08

0.07

1.36

2.10

0.45

0.26

3.49

3.85

0.15

0.37

0.96

0.91

6.58

0.88

5.21

0.26

0.10

0.73

0.87

1.01

0.27

1.53

1.28

0.36

0.62

0.91

0.91

6.51

0.83

4.27

0.80

0.26

0.48

1.30

0.87

0.13

0.35

0.64

1.06

0.37

0.88

0.86

5.00

0.76

3.30

0.16

0.08

0.70

1.49

0.40

0.14

0.89

1.13

0.24

0.26

0.75

0.82

4.29

0.82

4.05

0.08

1.78

2.60

0.26

0.11

1.70

2.30

0.15

1.86

1.10

0.89

0.79

3.89

0.78

3.55

0.21

1.42

2.46

2.14

0.56

2.52

4.01

2.41

1.63

1.80

0.94

0.79

3.86

0.75

3.22

0.45

0.57

0.45

2.87

0.99

0.55

0.79

1.95

1.02

0.69

0.77

0.79

3.83

0.61

2.21

0.27

0.71

2.24

3.07

0.83

1.34

4.09

3.87

0.98

1.19

0.95

0.79

3.82

0.84

4.31

0.61

0.50

0.12

1.31

1.49

0.56

0.20

1.16

1.11

0.87

0.79

0.78

3.75

0.82

4.01

0.38

0.24

0.53

2.31

1.00

0.27

0.99

1.54

0.62

0.50

0.80

0.74

3.33

0.75

3.24

0.64

0.88

0.45

0.85

2.62

1.79

1.17

0.82

1.52

2.14

0.74

0.71

3.05

0.72

2.91

0.30

2.66

3.51

1.03

0.32

1.73

2.31

0.42

2.97

1.23

0.86

0.71

3.01

0.83

4.25

0.68

0.53

0.03

1.25

1.35

0.47

0.04

0.83

1.21

0.75

0.75

0.70

2.90

0.70

2.78

0.29

1.04

1.65

1.88

0.93

2.14

2.97

1.92

1.33

1.73

0.84

0.67

2.68

0.50

1.64

0.72

1.63

2.36

3.79

2.59

3.56

4.96

5.62

2.35

3.28

0.91

0.62

2.39

0.47

1.52

0.27

0.41

0.37

1.40

0.48

0.50

0.37

0.72

0.68

0.51

0.59

0.59

2.20

0.55

1.84

1.63

4.50

5.16

4.24

1.47

2.51

2.86

1.61

6.13

2.16

0.82

0.56

2.05

0.53

1.77

2.49

7.13

7.79

5.04

2.28

3.94

4.41

1.94

9.62

3.40

0.91

0.51

1.78

0.23

0.66

1.01

1.63

2.56

2.55

1.52

1.63

2.14

1.56

2.64

1.62

0.68

0.48

1.63

0.39

1.21

0.98

2.45

3.07

5.23

0.85

1.26

1.94

2.93

3.43

1.12

0.83

0.27

0.84

0.02

0.05

0.15

0.67

0.31

4.53

0.32

0.91

0.40

3.20

0.82

0.70

0.75

0.81

3.90

0.77

3.21

0.35

0.43

0.23

2.19

1.13

0.64

0.53

1.75

0.78

0.81

0.77

0.80

3.75

0.72

2.51

0.11

0.45

1.71

3.16

0.14

0.24

1.71

2.76

0.56

0.21

0.90

0.75

3.16

0.85

4.29

1.58

2.53

1.92

2.75

3.14

2.44

2.19

1.43

4.11

2.77

0.92

0.71

2.82

0.47

1.42

1.78

3.60

2.30

0.40

2.18

2.42

1.61

0.12

5.38

2.41

0.86

0.66

2.47

0.50

1.53

1.25

1.78

1.19

1.94

3.31

3.15

1.67

1.23

3.03

3.39

0.89

0.52

1.74

0.62

2.09

0.46

0.10

0.22

1.96

0.71

0.09

0.15

1.13

0.56

0.33

0.63

0.32

0.95

0.68

2.45

1.20

1.43

0.38

3.13

1.21

0.71

0.24

1.47

2.62

0.89

0.77

0.84

4.08

0.82

3.45

1.37

2.09

0.38

0.78

4.96

3.34

0.84

0.51

3.46

4.07

0.97

0.78

3.27

0.77

2.94

0.96

1.43

0.46

1.25

5.78

3.93

1.70

1.37

2.40

4.80

0.96

0.73

2.80

0.80

3.23

1.06

1.36

0.75

0.45

2.63

1.44

1.12

0.23

2.42

1.86

0.82

0.71

2.63

0.70

2.43

1.73

2.75

2.01

1.05

7.82

6.48

5.33

0.90

4.48

7.35

0.97

0.67

2.38

0.71

2.45

0.80

0.83

0.13

0.74

2.41

1.03

0.21

0.54

1.62

1.47

0.93

0.64

2.18

0.60

1.84

0.28

0.10

0.34

0.24

0.63

0.10

0.55

0.18

0.38

0.27

0.80

0.56

1.79

0.83

3.70

0.12

0.02

1.64

0.59

0.18

0.03

1.22

0.27

0.14

0.11

0.92

0.44

1.28

0.63

2.00

2.00

0.63

0.47

0.85

0.89

0.16

0.13

0.07

2.63

0.45

0.77

0.34

0.95

0.18

0.46

1.05

1.92

0.72

0.01

1.31

1.32

0.55

0.00

2.97

1.39

0.63

0.31

0.86

0.54

1.58

0.17

0.39

2.01

1.80

0.23

0.39

1.39

0.70

0.56

0.34

0.85

0.26

0.72

0.45

1.12

0.44

0.25

0.04

1.69

0.59

0.24

0.03

0.85

0.69

0.39

0.60

0.22

0.59

0.30

0.77

0.23

1.16

2.30

0.22

0.78

2.06

4.60

0.12

1.39

1.72

0.97

0.09

0.23

0.11

0.27

0.35

0.55

0.79

2.64

1.22

1.12

0.93

1.48

0.91

1.23

0.74

0.06

0.17

0.01

0.02

0.56

1.41

0.72

1.11

1.04

1.55

0.68

0.37

1.96

1.43

0.74

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page48

of

62

Exhibit 2
Google

Section
Years
Job

Title

of

Data

1

Section

Total

Emp

Years

Level

Correlation

T

Coeff

2

Section

Change

T

Coeff

Stat

Regression

Correlation

Contemp

Stat

Lagged

3

Section
Regression

Coefficients

Revenue

SJ

Emp

Contemp

Lagged

4

T

Section

5

Section

6

Net Effect

Stats

Revenue

SJ

Emp

C

T

L

Stat

r2

0.11

0.26

0.05

0.12

1.78

4.82

3.95

8.75

0.53

0.77

0.59

0.38

6.61

0.72

0.69

0.10

0.25

0.40

0.98

0.64

1.19

2.95

1.74

0.96

1.03

2.17

0.98

1.83

1.01

0.98

0.09

0.22

0.47

1.20

0.22

0.67

2.13

1.85

0.22

0.39

1.04

0.85

0.89

0.33

0.96

0.08

0.19

0.61

1.74

0.11

0.73

1.64

0.18

0.35

1.26

2.37

0.16

0.84

0.97

0.92

0.00

0.00

0.54

1.44

0.19

1.04

2.39

4.19

0.27

0.75

1.17

1.85

1.24

0.59

0.95

0.19

0.47

0.36

0.87

0.44

1.21

2.37

2.43

0.85

1.21

2.05

1.36

1.66

1.10

0.94

0.94

6.31

0.98

0.92

0.44

0.15

1.14

1.60

0.34

0.13

0.94

1.36

0.76

0.99

0.88

4.22

0.98

9.66

1.71

1.08

1.17

1.74

2.76

0.95

0.95

1.42

2.78

1.63

0.99

0.81

3.05

0.93

5.04

2.09

1.73

1.40

4.09

5.52

4.20

3.82

7.88

1.00

0.80

2.97

0.89

3.87

1.89

2.59

2.38

0.19

1.24

0.96

0.73

0.07

4.48

1.07

0.91

0.78

2.79

0.92

4.85

0.04

1.56

2.30

0.05

0.07

1.45

2.12

0.04

1.60

0.99

0.99

0.77

2.68

0.87

3.50

0.01

0.93

1.40

1.72

0.03

1.46

2.31

2.49

0.94

1.01

0.99

0.76

2.60

0.79

2.55

2.08

3.14

6.08

2.19

1.36

1.38

1.95

0.97

5.22

1.38

0.98

0.73

2.36

0.77

2.38

0.48

1.11

2.62

0.84

6.23

8.70

7.53

1.59

7.81

1.00

0.72

2.31

0.73

2.15

2.48

6.19

6.26

2.27

3.18

3.57

2.61

8.67

3.46

1.00

0.70

2.22

0.77

2.40

0.78

1.84

3.07

1.89

9.88

0.69

2.14

0.75

2.28

0.69

2.40

3.41

7.95

0.25

0.42

0.61

1.33

3.09

0.37

0.93

0.67

2.00

0.86

3.38

1.48

1.36

0.94

2.69

0.97

0.51

0.33

0.73

2.85

0.69

0.94

0.64

1.87

0.87

3.48

0.04

0.79

1.30

0.83

0.15

1.63

2.67

1.56

0.83

1.15

0.99

0.63

1.80

0.55

1.14

0.39

0.10

2.24

0.62

1.76

0.63

1.61

0.92

2.25

3.15

0.31

4.54

5.33

8.35

0.79

3.17

5.10

1.00

0.61

1.74

0.68

1.83

0.01

0.21

1.26

0.28

0.02

0.15

0.74

0.18

0.20

0.09

0.89

0.60

1.68

0.64

1.66

0.89

1.99

3.14

0.82

5.88

6.81

2.59

2.88

6.54

1.00

0.60

1.67

0.75

2.29

0.41

0.22

0.58

1.15

0.85

0.25

0.60

1.23

0.64

0.47

0.99

0.57

1.56

0.90

4.02

0.15

0.71

1.44

1.90

0.22

0.49

1.27

1.40

0.56

0.26

0.97

0.56

1.52

0.76

2.33

0.78

0.82

0.11

0.71

1.67

0.94

0.12

0.79

1.60

1.20

0.99

0.50

1.29

0.39

0.83

4.23

8.54

8.63

7.90

1.16

1.18

1.07

1.13

0.49

1.26

0.67

1.78

1.37

4.14

4.70

0.11

0.20

0.22

0.91

0.47

1.20

0.38

0.82

0.80

1.63

2.83

2.19

3.13

3.15

6.13

0.44

1.11

0.37

0.81

1.66

2.94

4.48

6.60

0.97

0.89

1.31

0.44

1.09

0.42

0.92

0.82

1.60

2.92

2.97

0.73

0.68

0.43

1.06

0.45

0.99

0.65

1.18

2.15

1.97

0.59

0.57

0.41

1.02

0.49

0.79

1.37

2.80

2.02

0.00

0.40

0.97

0.54

1.30

5.72

5.70

1.24

0.23

0.53

0.45

1.01

0.28

0.43

0.82

0.22

0.38

0.22

0.51

0.16

0.22

2.68

4.65

1.97

0.00

0.21

0.49

0.41

0.90

0.83

3.92

4.02

7.39

2.91

0.18

0.41

0.31

0.66

0.20

0.67

2.19

2.29

0.92

0.13

0.29

0.00

0.01

0.36

0.84

1.88

1.39

0.30

0.69

0.11

0.22

3.76

6.86

6.03

0.30

0.69

0.60

1.51

1.75

2.91

2.70

0.94

5.52

0.96

5.86

10.15

13.34

10.00

11.51

12.40

18.00

4.53

19.74

10.69

11.61

2.62

11.63

1.00

12.58

10.24

1.17

0.85

2.77

0.08

0.84

4.50

2.43

3.16

0.99

1.73

4.59

0.92

0.93

1.34

1.06

2.42

0.70

0.88

0.99

0.91

1.83

0.58

0.98

1.29

1.52

1.11

1.27

0.94

0.26

0.38

0.10

0.71

0.30

1.00

5.34

6.73

2.49

4.76

4.89

0.99

1.76

4.10

1.40

0.87

1.55

0.98

1.32

1.38

3.83

0.58

1.20

1.43

0.99

2.52

6.36

5.97

5.30

2.11

1.26

2.35

2.34

2.92

1.03

24.13

12.77

19.06

10.62

4.65

6.14

1.00

2.36

0.94

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page49

of

62

Exhibit 2
Google

Section
Years
Job

Title

of

Data

1

Section

Total

Emp

Years

Level

Correlation

T

Coeff

2

Section

Change

T

Coeff

Stat

Regression

Correlation

Contemp

Stat

0.44

0.99

0.60

1.32

0.42

0.93

0.50

0.99

0.38

0.83

0.42

0.81

0.35

0.74

0.27

0.49

0.34

0.72

0.64

1.45

0.30

0.63

0.95

3.20

0.30

0.63

0.18

0.32

0.29

0.61

0.17

0.30

0.25

0.51

0.18

0.32

0.22

0.45

0.08

0.14

0.19

0.39

0.55

1.13

0.15

0.31

0.30

0.45

0.14

0.29

0.37

0.69

0.12

0.23

0.15

0.27

0.10

0.20

0.58

1.24

0.09

0.18

0.01

0.01

0.07

0.13

0.07

0.12

0.04

0.09

0.37

0.69

0.05

0.11

0.28

0.51

0.24

0.48

0.60

1.31

Lagged

3

Section
Regression

Coefficients

Revenue

SJ

Emp

Contemp

Lagged

4

T

Section

Revenue

5

Section

Net Effect

Stats

SJ

Emp

C

L

T

Stat

r2

6

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page50

of

62

Exhibit 2
Intel

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

2

Section

Change

Correlation

Coeff

Stat

Correlation

T

Coeff

3

Section

Regression

Contemp

Stat

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

T

Section

Net

Stats

Revenue

SJ

Emp

C

L

5

Section

6

Effect

T

Stat

r2

11

432

0.96

10.82

0.95

8.41

2.03

0.51

0.64

0.34

6.11

0.78

1.25

0.76

1.52

1.78

0.95

11

1501

0.96

9.78

0.94

7.56

1.56

0.30

0.32

0.54

6.76

0.36

0.73

1.63

1.86

2.07

0.96

11

233

0.94

8.46

0.91

6.14

1.47

1.33

0.23

0.09

4.71

0.74

0.25

0.15

2.80

1.46

0.92

11

3042

0.94

8.03

0.89

5.67

0.61

0.39

0.20

0.31

7.76

2.09

1.33

1.93

1.00

4.39

0.95

11

5042

0.92

7.30

0.91

6.21

0.81

2.22

0.06

0.63

3.59

2.93

0.23

2.53

3.03

4.40

0.96

11

293

0.91

6.73

0.89

5.46

2.30

0.95

0.19

0.45

4.05

0.63

0.18

0.54

3.25

1.88

0.88

11

724

0.88

5.65

0.94

8.07

1.43

0.58

0.19

0.55

1.48

0.38

0.39

1.04

2.00

2.26

0.91

11

59

0.88

5.56

0.72

2.91

1.12

0.73

0.22

0.33

2.35

0.84

0.37

0.49

1.85

1.54

0.81

11

394

0.88

5.52

0.88

5.34

0.63

0.35

0.13

0.06

4.97

1.77

0.54

0.30

0.98

3.77

0.87

11

3991

0.88

5.51

0.96

9.32

1.21

0.07

0.45

0.45

5.45

0.12

2.00

1.73

1.28

2.52

0.97

11

715

0.86

4.96

0.96

9.29

1.41

0.28

0.49

0.32

4.26

0.51

1.60

0.87

1.13

2.18

0.95

11

437

0.85

4.85

0.84

4.41

0.76

0.75

0.30

0.49

4.90

1.85

1.46

2.05

1.51

3.13

0.95

11

6082

0.85

4.85

0.94

7.51

0.81

0.45

0.34

0.48

6.95

1.58

2.34

2.61

1.27

4.17

0.97

11

912

0.85

4.76

0.94

7.60

0.95

0.69

0.20

0.59

3.95

1.52

0.76

1.49

1.64

3.53

0.94

11

31

0.84

4.74

0.82

4.00

0.59

0.35

0.44

0.13

3.17

0.95

1.78

0.52

0.94

2.06

0.91

11

216

0.83

4.50

0.83

4.23

0.66

0.62

0.09

0.03

4.10

2.02

0.34

0.08

1.28

3.57

0.93

11

1681

0.83

4.45

0.92

6.69

0.78

0.39

0.30

0.37

5.05

1.16

1.60

1.35

1.17

3.20

0.96

11

103

0.81

4.17

0.87

4.91

0.76

0.70

0.09

0.30

4.60

2.74

0.41

1.11

1.46

4.40

0.93

11

2903

0.81

4.12

0.95

8.50

0.92

0.20

0.32

0.30

8.74

0.80

2.51

1.67

1.12

4.24

0.98

11

413

0.81

4.11

0.95

8.85

0.88

0.38

0.07

0.09

5.34

1.23

0.34

0.31

1.26

3.91

0.95

11

1438

0.81

4.08

0.93

7.04

0.96

0.63

0.02

0.19

3.97

1.40

0.08

0.43

1.58

3.38

0.92

11

2235

0.80

4.01

0.89

5.55

0.73

0.22

0.42

0.36

7.48

1.12

3.34

2.29

0.95

4.04

0.98

11

4821

0.80

4.00

0.96

9.45

0.80

0.19

0.27

0.26

1.28

3.21

2.23

1.00

5.90

0.99

11

638

0.80

3.98

0.91

6.09

0.77

0.53

0.13

0.22

4.39

1.74

0.59

0.74

1.31

3.66

0.94

11

760

0.80

3.97

0.93

7.45

0.94

0.34

0.23

0.29

5.66

1.03

1.16

1.11

1.28

3.47

0.96

11

501

0.79

3.91

0.88

5.24

0.75

0.24

0.46

0.50

4.67

0.68

2.22

1.90

0.99

2.42

0.96

11

1538

0.79

3.90

0.91

6.15

0.78

0.20

0.22

0.05

3.77

0.59

0.79

0.17

0.98

2.32

0.90

11

292

0.79

3.89

0.82

4.10

0.70

0.83

0.05

0.23

3.30

2.23

0.16

0.52

1.53

3.43

0.85

11

528

0.79

3.81

0.75

3.23

0.84

1.07

0.36

0.95

4.51

2.41

1.37

3.86

1.91

3.58

0.96

11

75

0.78

3.80

0.81

3.88

2.04

0.36

0.21

0.24

3.00

0.25

0.19

0.23

2.40

1.22

0.83

11

244

0.78

3.78

0.90

5.76

0.68

0.61

0.06

0.23

9.04

4.38

0.55

1.62

1.29

7.24

0.97

11

5735

0.78

3.75

0.91

6.32

0.76

0.29

0.30

0.31

6.40

1.23

2.00

1.53

1.06

3.83

0.97

11

2120

0.78

3.72

0.95

9.08

0.74

0.29

0.11

0.08

2.62

1.25

0.67

1.03

7.72

0.99

11

328

0.77

3.66

0.77

3.41

0.75

0.71

0.38

0.88

4.32

2.20

1.67

3.46

1.46

3.53

0.93

11

1011

0.77

3.64

0.91

6.37

0.74

0.36

0.06

0.16

6.31

1.72

0.35

0.66

1.09

4.25

0.95

11

811

0.77

3.62

0.84

4.31

0.67

0.44

0.10

0.20

3.33

1.31

0.35

0.63

1.11

2.49

0.81

11

262

0.77

3.61

0.91

6.02

0.75

0.54

0.02

0.17

4.38

2.21

0.07

0.64

1.28

4.18

0.92

11

1332

0.77

3.61

0.92

6.65

0.79

0.51

0.18

0.35

4.64

1.60

0.85

1.17

1.30

3.57

0.94

11

104

0.77

3.57

0.84

4.35

0.53

0.19

0.54

0.50

4.55

0.98

3.37

2.61

0.72

2.80

0.96

11

91

0.76

3.52

0.89

5.55

1.09

0.23

0.37

0.29

3.84

0.37

0.82

0.50

1.32

2.15

0.83

11

127

0.75

3.44

0.90

6.00

0.35

0.00

0.00

0.08

3.84

0.02

0.00

0.63

0.35

1.75

0.86

12.44

11.59

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page51

of

62

Exhibit 2
Intel

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

2

Section

Change

Correlation

Coeff

Stat

Correlation

T

Coeff

3

Section

Regression

Contemp

Stat

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

T

Section

Net

Stats

Revenue

SJ

Emp

C

L

5

Section

6

Effect

T

Stat

r2

11

1906

0.71

3.04

0.97

0.85

0.22

0.13

0.14

9.03

1.52

1.05

0.83

1.07

6.07

0.98

11

3531

0.71

3.03

0.89

5.61

0.72

0.21

0.33

0.26

7.95

1.32

2.79

1.66

0.93

4.71

0.98

11

934

0.71

3.03

0.92

6.73

0.72

0.36

0.04

0.02

7.71

2.74

0.33

0.10

1.08

6.22

0.98

11

1873

0.71

3.02

0.96

9.25

0.85

0.36

0.21

0.43

9.91

2.60

2.01

2.80

1.21

7.24

0.99

11

130

0.71

2.99

0.90

5.77

0.86

0.03

0.42

0.38

3.38

0.07

1.28

0.96

0.89

1.59

0.89

11

2037

0.70

2.98

0.92

6.42

0.63

0.23

0.18

0.11

8.43

1.93

1.78

0.85

0.86

5.50

0.98

11

88

0.70

2.98

0.91

6.08

0.69

0.06

0.27

0.13

3.97

0.21

1.20

0.54

0.75

2.06

0.91

11

366

0.70

2.95

0.95

8.65

0.67

0.18

0.16

0.11

2.16

2.20

1.16

0.85

8.03

0.99

11

137

0.70

2.94

0.67

2.53

0.71

0.37

0.76

0.67

3.37

1.01

2.72

2.35

1.08

2.16

0.96

11

828

0.70

2.92

0.93

7.12

0.63

0.27

0.10

0.06

5.89

1.78

0.60

0.37

0.89

4.25

0.92

11

969

0.70

2.91

0.91

6.08

0.66

0.35

0.18

0.26

6.39

2.57

1.14

1.47

1.01

5.16

0.94

11

87

0.69

2.89

0.75

3.25

0.92

1.57

0.16

0.84

4.41

3.84

0.58

2.69

2.49

5.00

0.93

11

179

0.69

2.87

0.87

5.06

0.64

0.05

0.57

0.77

5.02

0.28

3.39

4.03

0.69

2.74

0.96

11

8983

0.69

2.87

0.96

9.77

0.78

0.25

0.03

0.09

12.24

2.56

0.35

0.73

1.03

8.27

0.99

11

934

0.69

2.86

0.96

0.83

0.15

0.12

0.04

12.79

1.57

1.42

0.31

0.98

8.03

0.99

11

1049

0.69

2.85

0.89

5.67

0.68

0.28

0.40

0.60

4.91

1.15

2.29

2.62

0.96

3.24

0.96

11

146

0.69

2.84

0.65

2.41

0.39

0.43

0.29

0.16

1.82

1.34

1.00

0.52

0.82

1.81

0.84

11

509

0.69

2.84

0.89

5.51

0.70

0.18

0.30

0.17

4.88

0.78

1.58

0.74

0.88

2.97

0.95

11

1402

0.69

2.83

0.94

7.53

0.77

0.19

0.26

0.34

4.41

0.81

1.16

1.27

0.96

3.11

0.94

11

2097

0.68

2.81

0.97

0.78

0.15

0.07

0.02

1.65

0.91

0.15

0.93

8.04

0.99

11

268

0.68

2.77

0.95

8.82

0.83

0.00

0.24

0.10

0.01

1.66

0.55

0.83

3.85

0.97

11

546

0.68

2.76

0.94

7.55

0.72

0.29

0.07

0.04

10.66

2.99

0.76

0.29

1.01

7.77

0.99

11

12004

0.68

2.75

0.95

8.95

0.76

0.28

0.02

0.07

16.18

4.10

0.24

0.81

1.04

11

577

0.67

2.74

0.96

9.51

0.82

0.18

0.02

0.02

6.42

1.06

0.12

0.08

1.00

4.25

0.95

11

50

0.67

2.72

0.45

1.42

1.17

0.66

0.28

0.65

1.34

0.35

0.19

0.47

1.83

0.73

0.66

11

358

0.67

2.72

0.85

4.50

0.58

0.40

0.23

0.30

4.89

2.47

1.27

1.51

0.98

4.17

0.90

11

753

0.67

2.70

0.97

0.91

0.20

0.12

0.25

2.81

1.70

2.54

1.11

11

517

0.67

2.69

0.84

4.39

0.49

0.28

0.06

0.02

3.39

1.48

0.26

0.07

0.77

2.76

0.87

11

547

0.67

2.68

0.95

9.06

0.78

0.29

0.08

0.16

9.05

2.41

0.68

1.08

1.07

6.51

0.98

11

834

0.66

2.67

0.94

7.57

0.81

0.02

0.36

0.27

8.80

0.16

3.17

1.75

0.83

4.89

0.99

11

556

0.66

2.66

0.89

5.49

0.73

0.28

0.05

0.02

3.34

1.08

0.15

0.07

1.00

2.64

0.84

11

361

0.66

2.65

0.55

1.88

1.08

1.40

0.82

0.92

4.01

3.26

2.79

2.83

2.48

3.84

0.79

11

955

0.66

2.65

0.95

8.72

0.67

0.22

0.09

0.12

6.90

1.62

0.63

0.72

0.89

4.72

0.95

11

188

0.66

2.64

0.88

5.23

0.67

0.43

0.06

0.11

7.43

3.15

0.46

0.60

1.09

5.97

0.97

11

169

0.66

2.63

0.92

6.63

0.78

0.01

0.43

0.34

4.71

0.04

2.09

1.30

0.77

2.49

0.96

11

91

0.66

2.62

0.84

4.34

1.85

0.51

0.49

0.64

3.26

0.37

0.49

0.66

2.36

1.55

0.91

11

94

0.66

2.60

0.84

4.32

0.61

0.00

0.79

0.89

1.50

0.00

1.68

1.45

0.61

0.89

0.87

11

59

0.65

2.59

0.81

3.93

0.97

0.78

0.29

0.10

2.52

1.75

0.57

0.15

1.75

2.67

0.82

11

537

0.65

2.59

0.97

0.81

0.20

0.12

0.17

2.35

1.33

1.54

1.01

8.64

0.99

11

249

0.65

2.59

0.78

3.47

0.69

0.54

0.26

0.23

3.24

1.77

0.82

0.62

1.23

2.81

0.77

11

557

0.65

2.58

0.90

5.76

0.61

0.06

0.35

0.26

4.22

0.27

1.84

1.16

0.67

2.34

0.94

10.58

10.05

11.50

11.28

11.61

13.19

13.52
7.42

18.00

12.42

11.58

11.93

0.99

0.99

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page52

of

62

Exhibit 2
Intel

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

2

Section

Change

Correlation

Coeff

Stat

Correlation

T

Coeff

3

Section

Regression

Contemp

Stat

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

T

Section

Net

Stats

Revenue

SJ

Emp

C

L

5

Section

6

Effect

T

Stat

r2

11

283

0.62

2.37

0.94

7.52

0.65

0.05

0.30

0.25

7.48

0.42

2.54

1.81

0.71

4.13

0.98

11

142

0.62

2.37

0.83

4.28

0.72

0.31

0.18

0.06

8.01

2.32

1.48

0.33

1.03

5.73

0.99

11

2959

0.62

2.36

0.92

6.75

0.72

0.20

0.13

0.05

8.49

1.69

1.18

0.35

0.92

5.64

0.98

11

880

0.62

2.36

0.93

7.34

0.70

0.32

0.11

0.18

5.74

1.75

2.20

1.03

11

202

0.61

2.34

0.85

4.49

0.77

0.28

0.22

0.61

5.44

1.53

1.16

2.85

1.05

3.95

0.94

11

1662

0.61

2.32

0.91

6.38

0.61

0.23

0.05

0.04

6.83

1.91

0.38

0.23

0.85

4.89

0.97

11

731

0.61

2.29

0.94

7.46

0.95

0.31

0.11

0.18

5.47

1.49

0.46

0.60

1.26

4.15

0.94

11

2205

0.61

2.29

0.68

2.62

0.68

0.79

0.41

0.54

4.01

2.68

2.34

2.66

1.47

3.36

0.81

11

2086

0.61

2.29

0.95

8.39

0.76

0.23

0.10

0.10

2.61

1.13

0.81

0.99

8.19

0.99

11

1156

0.61

2.29

0.71

2.86

0.74

0.86

0.51

0.62

6.12

4.08

4.06

4.03

1.60

5.14

0.91

11

91

0.61

2.29

0.64

2.33

0.90

1.18

0.26

1.23

2.02

1.38

0.41

2.07

2.07

1.92

0.87

11

1393

0.60

2.26

0.84

4.44

0.72

0.19

0.02

0.05

2.62

0.50

0.06

0.13

0.90

1.70

0.74

11

96

0.60

2.26

0.84

4.38

0.60

0.15

0.30

0.21

2.16

0.44

0.84

0.51

0.75

1.50

0.84

11

281

0.60

2.25

0.80

3.74

0.73

0.65

0.11

0.15

3.57

2.18

0.38

0.43

1.37

3.32

0.84

11

128

0.60

2.24

0.94

7.89

0.75

0.26

0.11

0.04

6.10

1.57

0.60

0.20

1.01

4.27

0.93

11

601

0.60

2.23

0.91

6.20

0.57

0.17

0.08

0.06

7.66

1.70

0.72

0.46

0.74

5.09

0.97

11

303

0.60

2.23

0.55

1.87

0.48

0.48

0.59

0.82

1.84

1.14

1.74

2.12

0.96

1.62

0.90

11

147

0.59

2.21

0.85

4.64

0.47

0.12

0.26

0.16

5.01

0.93

2.02

1.06

0.59

3.12

0.96

11

261

0.59

2.20

0.68

2.60

0.63

0.93

0.49

1.91

1.52

1.85

1.07

3.78

1.56

2.22

0.93

11

282

0.59

2.19

0.68

2.59

0.54

0.63

0.41

0.42

4.24

3.97

1.95

1.85

1.18

4.79

0.89

11

223

0.59

2.18

0.88

5.13

0.59

0.14

0.15

0.09

2.98

0.53

0.52

0.31

0.72

1.93

0.83

11

5107

0.59

2.18

0.95

8.21

0.84

0.24

0.15

0.28

5.52

1.16

0.75

1.03

1.08

3.87

0.95

11

213

0.59

2.18

0.82

4.07

0.45

0.03

0.45

0.36

2.95

0.14

2.19

1.50

0.48

1.57

0.92

11

347

0.58

2.15

0.93

6.90

0.76

0.02

0.21

0.24

3.83

0.08

0.78

0.66

0.79

1.98

0.88

11

135

0.58

2.15

0.76

3.34

0.38

0.13

0.15

0.03

2.29

0.56

0.61

0.13

0.50

1.49

0.80

11

1471

0.58

2.13

0.93

7.06

0.65

0.32

0.21

0.23

8.65

3.35

1.84

1.76

0.97

6.81

0.96

11

2090

0.58

2.13

0.95

9.03

0.60

0.18

0.03

0.01

8.79

1.97

0.27

0.07

0.79

5.91

0.97

11

197

0.58

2.13

0.91

6.18

0.77

0.16

0.05

0.05

3.57

0.62

0.15

0.13

0.94

2.36

0.86

11

35

0.58

2.12

0.76

3.32

0.76

0.57

0.11

0.36

1.09

0.86

0.14

0.34

1.33

1.21

0.72

11

159

0.57

2.11

0.85

4.48

0.98

0.74

0.47

0.30

4.51

2.97

1.57

0.79

1.72

4.57

0.90

11

126

0.57

2.10

0.69

2.71

1.14

1.07

0.90

0.54

4.38

3.53

2.14

1.07

2.21

4.76

0.86

11

223

0.57

2.09

0.95

8.28

0.68

0.18

0.13

0.18

6.80

1.40

0.96

1.01

0.86

4.66

0.97

11

934

0.57

2.08

0.91

6.20

0.82

0.33

0.01

0.06

6.86

2.24

0.05

0.28

1.15

5.30

0.97

11

403

0.57

2.07

0.87

4.91

0.55

0.13

0.29

0.30

3.25

0.58

1.23

1.05

0.68

2.07

0.89

11

1801

0.57

2.06

0.96

9.45

0.70

0.22

0.06

0.09

3.09

0.78

0.93

0.91

9.09

0.99

11

400

0.57

2.06

0.85

4.49

0.67

0.45

0.41

0.40

5.67

2.89

2.06

1.68

1.11

4.91

0.90

11

390

0.57

2.06

0.88

5.26

0.57

0.16

0.17

0.11

3.80

0.78

0.81

0.43

0.73

2.48

0.91

11

115

0.56

2.04

0.57

1.97

0.29

0.20

0.31

0.21

1.24

0.60

0.89

0.54

0.49

1.00

0.64

11

556

0.56

2.03

0.95

8.49

0.65

0.18

0.00

0.02

6.90

1.45

0.01

0.13

0.84

4.56

0.95

11

120

0.56

2.03

0.62

2.25

0.48

0.36

0.00

0.07

1.78

0.90

0.00

0.15

0.83

1.46

0.50

11

5274

0.56

2.02

0.92

6.52

0.60

0.23

0.29

0.32

6.74

1.85

2.08

2.13

0.83

4.58

0.93

16.35

11.64

13.13

12.77

0.99

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page53

of

62

Exhibit 2
Intel

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

2

Section

Change

Correlation

Coeff

Stat

Correlation

T

Coeff

3

Section

Regression

Contemp

Stat

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

T

Section

Net

Stats

Revenue

SJ

Emp

C

L

5

Section

6

Effect

T

Stat

r2

11

147

0.51

1.78

0.54

1.81

1.24

1.41

0.21

0.81

1.90

1.26

0.23

0.75

2.66

1.76

0.77

11

86

0.51

1.77

0.79

3.65

1.01

0.67

0.58

0.49

3.20

1.85

1.27

1.00

1.68

2.97

0.78

11

102

0.50

1.75

0.81

3.91

0.54

0.33

0.22

0.52

3.49

1.69

0.97

1.88

0.87

2.97

0.90

11

4667

0.50

1.75

0.98

0.61

0.16

0.13

0.18

23.02

4.67

3.16

3.88

0.77

11

1283

0.50

1.74

0.96

9.47

0.92

0.32

0.18

0.20

11.04

3.34

1.57

1.40

1.24

8.41

0.98

11

54

0.50

1.74

0.57

1.94

0.57

0.03

0.42

0.12

0.99

0.03

0.60

0.17

0.54

0.38

0.54

11

222

0.49

1.67

0.70

2.76

0.62

0.56

0.36

0.29

2.67

1.82

0.99

0.69

1.18

2.61

0.70

11

43

0.48

1.66

0.60

2.11

0.79

1.05

0.64

0.46

2.16

2.61

1.17

0.66

1.84

2.77

0.79

11

56

0.47

1.62

0.76

3.30

0.53

0.16

0.41

0.70

1.48

0.37

0.91

1.23

0.70

1.05

0.81

11

536

0.46

1.56

0.88

5.16

0.70

0.04

0.16

0.24

3.19

0.13

0.51

0.82

0.66

1.60

0.81

11

7841

0.46

1.55

0.94

7.67

0.82

0.32

0.37

0.32

9.49

2.99

2.82

2.15

1.14

7.10

0.96

11

325

0.46

1.55

0.68

2.65

0.21

0.18

0.74

0.69

1.37

0.86

3.29

2.77

0.04

0.12

0.89

11

249

0.46

1.54

0.53

1.79

1.23

1.07

0.31

0.93

1.94

0.98

0.36

0.84

2.29

1.50

0.62

11

666

0.46

1.54

0.96

9.70

0.68

0.13

0.01

0.03

6.56

1.02

0.06

0.15

0.81

4.14

0.94

11

150

0.46

1.54

0.91

6.38

0.52

0.03

0.28

0.35

6.29

0.24

2.44

2.80

0.55

3.42

0.96

11

106

0.44

1.49

0.78

3.50

0.66

0.53

0.14

0.07

2.86

2.01

0.44

0.16

1.19

2.82

0.87

11

101

0.44

1.46

0.72

2.94

0.57

0.04

0.50

0.56

1.39

0.07

0.93

0.89

0.62

0.76

0.76

11

1976

0.44

1.46

0.83

4.16

0.68

0.48

0.47

0.38

6.73

3.82

2.95

2.20

1.16

5.99

0.92

11

353

0.43

1.43

0.82

4.00

0.71

0.28

0.25

0.20

2.97

0.92

0.68

0.53

0.99

2.16

0.72

11

56

0.43

1.42

0.49

1.57

1.04

1.39

0.40

0.48

1.87

1.86

0.52

0.57

2.43

2.24

0.67

11

137

0.43

1.42

0.87

4.89

0.81

0.36

0.30

0.35

3.47

1.33

0.87

0.85

1.18

2.78

0.83

11

105

0.42

1.38

0.86

4.75

0.84

0.39

0.31

0.05

6.05

2.44

1.50

0.24

1.23

4.84

0.92

11

125

0.41

1.34

0.58

2.03

0.57

0.70

0.34

0.12

2.36

2.39

0.99

0.34

1.27

2.77

0.77

11

117

0.41

1.33

0.58

2.03

0.53

0.23

0.87

1.07

0.83

0.25

1.07

1.28

0.30

0.24

0.67

11

65

0.40

1.32

0.02

0.07

0.48

1.30

0.35

0.08

1.01

2.07

0.47

0.10

1.78

1.85

0.59

11

156

0.38

1.22

0.74

3.13

0.60

0.32

0.49

0.61

3.02

1.23

1.54

1.64

0.92

2.34

0.73

11

35

0.35

1.14

0.59

2.08

0.13

0.31

0.80

0.34

0.31

0.61

1.51

0.55

0.18

0.23

0.82

11

98

0.35

1.12

0.57

1.97

0.63

0.55

0.53

0.51

1.92

1.28

1.03

0.93

1.18

1.83

0.50

11

225

0.34

1.10

0.71

2.82

0.58

0.08

0.58

0.82

1.30

0.14

0.92

1.07

0.50

0.59

0.67

11

171

0.34

1.08

0.80

3.76

0.70

0.12

0.43

0.34

3.96

0.49

1.54

1.13

0.82

2.35

0.78

11

45

0.34

1.08

0.50

1.62

0.09

0.43

1.15

1.06

0.44

1.56

3.87

3.50

0.34

0.82

0.87

11

533

0.34

1.07

0.41

1.28

1.15

1.12

0.12

1.23

1.70

1.00

0.13

1.01

2.27

1.42

0.66

11

243

0.33

1.05

0.86

4.84

0.61

0.24

0.31

0.42

4.09

1.26

1.28

1.53

0.85

2.92

0.85

11

774

0.33

1.04

0.83

4.27

0.45

0.16

0.02

0.16

3.29

0.89

0.08

0.75

0.60

2.26

0.86

11

47

0.29

0.92

0.73

3.05

0.47

0.13

0.47

0.46

1.38

0.30

1.06

0.98

0.34

0.53

0.69

11

199

0.27

0.84

0.60

2.10

0.44

0.37

0.19

0.36

1.43

0.96

0.38

0.55

0.81

1.32

0.68

11

111

0.25

0.76

0.48

1.56

0.31

0.18

0.21

0.29

1.00

0.46

0.51

0.68

0.49

0.81

0.53

11

30

0.21

0.64

0.09

0.25

0.14

0.54

0.12

0.12

0.33

0.99

0.19

0.19

0.68

0.80

0.43

11

31

0.17

0.52

0.66

2.46

0.23

0.65

0.88

0.73

0.98

1.97

2.75

2.07

0.42

0.85

0.79

11

361

0.12

0.38

0.79

3.70

0.59

0.11

0.24

0.14

3.26

0.46

0.90

0.50

0.70

1.96

0.71

11

734

0.03

0.08

0.47

1.51

0.65

0.02

0.22

0.63

3.20

0.07

0.77

2.03

0.63

1.49

0.84

12.47

14.83

0.99

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page54

of

62

Exhibit 2
Intel

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

2

Section

Change

Correlation

Coeff

Stat

Correlation

T

Coeff

3

Section

Regression

Contemp

Stat

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

T

Section

Net

Stats

Revenue

SJ

Emp

C

L

5

Section

6

Effect

T

Stat

r2

10

149

0.69

2.70

0.84

3.75

0.24

1.11

0.15

0.83

0.32

2.13

0.30

1.34

1.35

1.70

0.92

10

1401

0.68

2.61

0.96

9.53

0.72

0.27

0.06

0.09

9.38

2.52

0.50

0.62

0.99

6.88

0.98

10

81

0.68

2.61

0.75

2.96

1.20

1.12

0.39

0.27

2.19

1.58

0.57

0.35

2.31

2.29

0.76

10

1872

0.63

2.29

0.95

8.08

0.69

0.29

0.05

0.06

8.10

2.53

0.42

0.42

0.98

6.08

0.97

10

53

0.62

2.26

0.46

1.25

0.68

0.75

0.23

0.32

5.31

4.85

1.71

1.26

1.42

5.73

0.97

10

31

0.61

2.20

0.94

7.06

1.28

0.42

0.82

1.13

5.65

1.31

2.71

3.26

0.86

2.03

0.98

10

40

0.60

2.09

0.89

5.10

1.03

0.47

0.29

0.81

2.39

0.97

0.62

1.30

1.50

2.17

0.90

10

951

0.59

2.06

0.93

6.70

0.62

0.30

0.24

0.25

3.72

2.57

2.31

0.92

7.78

0.97

10

20

0.58

2.04

0.56

1.66

0.30

0.27

0.47

0.38

1.05

0.68

1.35

0.78

0.03

0.04

0.87

10

37

0.58

2.04

0.89

4.84

1.29

0.23

0.09

0.05

2.39

0.41

0.19

0.08

1.51

2.05

0.90

10

113

0.57

1.98

0.73

2.61

0.21

0.27

0.20

0.09

0.34

0.52

0.37

0.15

0.48

0.52

0.81

10

464

0.57

1.97

0.82

3.86

0.84

0.93

0.34

0.18

1.89

1.44

0.72

0.27

1.77

2.36

0.95

10

86

0.55

1.88

0.56

1.64

1.30

2.76

0.18

0.29

2.28

1.89

0.22

0.34

4.05

2.12

0.73

10

29

0.48

1.55

0.90

5.35

0.63

0.27

0.06

0.22

4.22

1.40

0.28

0.76

0.90

3.16

0.90

10

107

0.48

1.54

0.78

3.31

0.67

0.81

0.22

0.33

3.00

3.39

0.89

0.77

1.48

4.16

0.98

10

878

0.47

1.52

0.92

6.26

0.96

0.40

0.12

0.15

4.86

1.53

0.43

0.32

1.37

3.79

0.93

10

42

0.46

1.45

0.87

4.28

0.72

0.53

0.35

0.76

0.50

0.79

0.45

1.06

1.24

0.73

0.95

10

281

0.45

1.42

0.66

2.34

0.30

0.20

0.23

0.09

1.79

0.88

0.85

0.28

0.50

1.48

0.78

10

49

0.37

1.13

0.94

7.27

0.64

0.15

0.13

0.28

5.60

0.83

0.83

1.38

0.49

2.03

0.93

10

340

0.34

1.02

0.92

6.08

0.52

0.16

0.11

0.21

6.64

1.52

0.96

1.61

0.68

4.43

0.96

10

44

0.26

0.78

0.91

5.82

1.04

0.03

0.32

0.06

3.33

0.08

0.84

0.10

1.01

1.83

0.91

10

42

0.26

0.76

0.79

3.13

3.52

1.68

0.54

1.64

6.75

2.12

0.79

1.94

5.21

4.68

0.97

10

157

0.23

0.68

0.40

1.17

0.28

0.30

0.16

0.07

0.52

0.43

0.22

0.09

0.58

0.54

0.43

10

20

0.28

0.83

0.32

0.88

0.07

0.37

0.13

1.18

0.16

0.68

0.21

1.75

0.44

0.52

0.68

10

40

0.34

1.02

0.48

1.45

0.16

0.33

0.16

1.00

0.30

0.52

0.25

1.38

0.17

0.16

0.68

9

72

0.84

4.12

0.73

2.59

2.09

0.76

0.09

1.59

1.57

0.30

0.04

0.56

2.86

0.82

0.81

9

46

0.78

3.34

0.77

2.94

1.06

0.67

0.54

0.76

1.37

0.24

0.30

0.43

1.73

0.56

0.81

105

0.78

3.31

0.79

3.13

1.15

0.86

0.01

0.49

9.29

0.16

3.20

2.02

9

18

0.77

3.16

0.75

2.57

0.57

0.15

0.76

0.64

0.99

0.19

1.43

0.64

0.72

0.80

0.89

9

50

0.75

3.01

0.85

3.89

0.77

0.92

0.37

1.82

0.50

0.82

0.35

0.91

1.69

0.90

0.87

9

64

0.75

2.98

0.92

4.79

3.72

0.33

1.05

1.80

1.60

0.23

0.69

0.79

4.05

1.75

0.92

172

0.72

2.73

0.85

3.92

0.82

0.28

0.19

0.33

1.36

0.33

0.19

0.26

1.10

0.91

0.75

9

50

0.61

2.03

0.70

2.19

0.92

0.94

0.21

0.16

3.01

1.38

0.49

0.26

1.86

2.00

0.97

9

67

0.43

1.26

0.21

0.49

0.05

0.30

0.88

0.96

0.13

0.54

1.61

1.59

0.26

0.31

0.71

9

17

0.36

1.01

0.55

1.31

5.91

3.81

2.42

0.48

2.49

2.36

2.09

0.41

9.72

3.51

0.96

9

13

0.17

0.46

0.58

1.41

0.10

0.15

0.52

0.29

0.10

0.12

0.49

0.29

0.05

0.02

0.79

9

52

0.08

0.22

0.60

1.81

1.09

0.34

0.38

0.65

3.50

1.05

1.09

0.99

1.43

2.58

0.95

8

283

0.99

17.90

0.97

9.74

0.86

0.01

0.14

0.01

6.72

0.02

1.05

0.05

0.85

1.38

0.97

8

864

0.98

12.28

0.98

9.96

0.75

0.36

0.18

0.24

1.90

2.63

1.88

1.12

5.69

0.99

8

1526

0.98

11.20

0.96

7.28

0.74

0.02

0.19

0.29

4.74

0.04

1.16

0.93

0.72

1.51

0.95

8

50

0.97

10.69

0.96

7.81

0.91

0.17

0.09

0.12

4.85

0.29

0.41

0.31

1.08

1.69

0.94

9

9

10.29

16.00

12.01

14.45

0.99

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page55

of

62

Exhibit 2
Intel

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

2

Section

Change

Correlation

Coeff

Stat

Correlation

T

Coeff

3

Section

Regression

Contemp

Stat

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

T

Section

Net

Stats

Revenue

SJ

Emp

C

L

5

Section

6

Effect

T

Stat

r2

8

359

0.91

5.42

0.84

3.53

0.59

0.10

0.50

0.47

4.46

0.38

3.44

1.76

0.49

1.50

0.97

8

152

0.91

5.32

0.67

2.01

0.43

0.75

0.12

0.29

3.08

3.03

0.66

1.03

1.18

3.69

0.92

98

0.90

5.00

0.94

6.09

1.44

0.21

0.34

0.88

2.61

0.25

0.71

0.98

1.65

2.00

0.94

8

374

0.90

4.94

0.70

2.19

0.58

0.04

0.53

0.61

1.66

0.04

1.49

1.07

0.62

0.51

0.92

8

689

0.90

4.92

0.94

6.39

0.87

1.50

0.11

0.83

1.67

1.34

0.30

1.87

2.37

3.38

0.99

8

203

0.88

4.53

0.95

6.80

2.06

0.25

0.42

0.73

2.43

7.04

9.31

2.32

8

291

0.87

4.41

0.94

6.05

0.64

0.46

0.13

0.17

7.62

3.81

1.42

1.07

1.10

7.90

0.99

65

0.86

4.07

0.65

1.90

1.50

0.76

1.51

1.81

3.06

1.28

1.46

3.42

2.26

5.65

0.95

318

0.84

3.77

0.91

4.83

0.65

0.35

0.12

0.06

4.53

1.09

0.46

0.15

1.00

2.51

0.98

24

0.83

3.68

0.74

2.49

0.68

2.29

0.04

3.04

1.06

2.31

0.04

1.88

2.97

2.25

0.94

8

217

0.82

3.50

0.96

7.70

0.63

0.17

0.21

0.27

6.94

8.81

6.75

0.80

8

201

0.82

3.49

0.84

3.43

0.54

0.30

0.04

0.29

3.36

1.40

0.22

0.89

0.84

2.75

0.88

8

214

0.81

3.36

0.94

6.24

0.62

0.24

0.01

0.06

5.00

1.47

0.07

0.25

0.86

3.82

0.96

8

304

0.81

3.36

0.52

1.37

0.23

0.07

0.37

0.43

0.84

0.18

1.44

0.91

0.16

0.25

0.72

8

266

0.80

3.32

0.91

4.94

0.48

0.43

0.41

0.06

4.05

1.36

2.10

0.27

0.05

0.12

0.98

8

116

0.80

3.30

0.91

4.95

0.62

0.34

0.34

0.76

6.72

2.40

3.96

4.87

0.95

6.30

0.99

8

180

0.78

3.03

0.88

4.13

0.40

0.16

0.13

0.04

2.75

0.51

0.48

0.12

0.56

1.37

0.94

8

1077

0.77

2.97

0.92

5.43

0.57

0.26

0.09

0.20

5.81

2.02

0.83

1.08

0.83

4.79

0.98

8

155

0.77

2.92

0.95

6.98

0.93

0.43

0.15

0.20

9.92

4.50

1.94

1.44

1.36

8

57

0.76

2.91

0.54

1.45

1.12

1.05

0.14

1.02

2.78

1.86

0.43

1.65

2.17

2.40

0.80

8

48

0.76

2.90

0.40

0.99

0.70

0.73

0.12

0.95

4.75

3.51

0.65

3.27

1.44

4.66

0.95

8

64

0.76

2.90

0.56

1.52

0.81

1.22

0.14

0.05

1.51

1.36

0.12

0.04

2.02

1.58

0.70

8

246

0.76

2.87

0.93

5.66

0.99

0.13

0.37

0.25

6.76

0.56

2.17

0.84

0.86

3.16

0.97

8

157

0.75

2.81

0.88

4.13

0.60

0.45

0.59

0.51

1.01

0.65

1.45

0.67

1.06

1.60

0.94

8

33

0.75

2.81

0.83

3.26

2.33

0.41

1.63

0.25

2.80

0.74

2.32

0.76

2.74

6.78

0.98

8

41

0.75

2.81

0.39

0.95

0.84

0.76

0.40

0.84

2.52

1.52

1.13

1.30

1.60

2.23

0.90

8

87

0.75

2.79

0.11

0.24

0.77

1.02

0.04

0.44

1.60

1.67

0.10

0.68

1.79

1.75

0.62

8

62

0.75

2.77

0.94

6.08

0.71

0.47

0.37

0.27

1.17

129.23

1.00

8

72

0.75

2.75

0.37

0.89

0.42

0.15

0.43

0.35

0.63

0.13

0.81

0.32

0.57

0.33

0.67

8

69

0.72

2.56

0.20

0.46

0.04

0.06

0.24

0.32

0.34

0.35

1.65

1.25

0.10

0.41

0.76

8

10

0.72

2.51

0.15

0.34

0.79

1.77

0.84

0.75

0.58

1.22

0.48

0.26

2.56

1.06

0.64

460

0.71

2.50

0.91

5.04

0.53

0.31

0.03

0.20

8.28

3.96

0.37

1.66

0.84

7.81

0.99

8

29

0.71

2.44

0.29

0.67

0.18

0.87

0.37

1.00

0.59

1.92

1.14

1.66

1.05

1.65

0.93

8

53

0.70

2.38

0.41

1.00

0.47

0.78

0.27

0.65

0.89

1.19

0.45

0.62

1.25

1.31

0.74

102

0.69

2.34

0.66

1.76

1.06

2.22

0.21

0.04

3.56

2.14

0.18

0.03

3.28

2.68

0.95

33

0.69

2.32

0.74

2.47

1.60

0.84

1.27

1.94

8.75

5.92

5.09

5.55

2.44

324

0.67

2.23

0.58

1.60

0.29

0.18

0.23

0.05

2.30

1.15

1.57

0.20

0.47

2.02

0.90

8

8

8

8

8

8

8

8

43.99

32.69

85.90

58.69

59.28

22.97

23.84

22.98

12.17

10.41

1.00

1.00

1.00

0.99

14

0.67

2.20

0.55

1.48

1.25

0.61

0.13

1.89

6.47

2.56

0.58

4.51

1.86

5.42

0.98

132

0.65

2.11

0.94

5.98

0.89

0.41

0.13

0.37

5.32

2.31

0.80

1.43

1.30

5.86

0.98

8

34

0.65

2.10

0.52

1.38

0.59

0.38

0.09

0.89

5.92

1.33

7.48

0.96

8

79

0.65

2.08

0.85

3.63

0.62

0.05

0.59

1.07

0.11

1.98

1.94

0.57

8

8

10.55
1.95

10.05

1.11

0.99

0.95

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page56

of

62

Exhibit 2
Intel

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

2

Section

Change

Correlation

Coeff

Stat

Correlation

T

Coeff

3

Section

Regression

Contemp

Stat

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

T

Section

Net

Stats

Revenue

SJ

Emp

C

L

5

Section

6

Effect

T

Stat

r2

17

0.09

0.23

0.04

0.10

0.56

0.38

1.10

1.61

0.55

0.23

0.59

0.80

0.94

0.37

0.70

7

104

0.99

14.44

0.82

2.85

1.14

1.18

0.09

0.01

1.48

1.03

0.25

0.01

2.32

1.59

0.85

7

163

0.99

13.23

0.85

3.20

0.57

0.15

0.26

0.17

0.71

117.74

1.00

7

283

0.98

10.30

0.90

4.19

0.89

0.35

0.06

0.14

2.34

1.14

0.36

0.51

1.24

1.91

0.97

7

245

0.97

9.67

0.79

2.57

1.32

0.61

0.15

0.49

3.19

2.10

0.84

1.67

1.94

2.94

0.95

7

236

0.97

8.77

0.68

1.87

1.14

0.88

0.17

0.10

6.89

2.33

2.02

7

18

0.96

8.21

0.38

0.82

0.14

0.87

0.01

0.31

0.12

1.14

0.02

0.36

0.73

0.49

0.87

7

43

0.95

7.10

0.23

0.47

0.31

0.93

0.23

0.25

0.21

0.95

0.33

0.22

1.24

0.59

0.60

7

26

0.95

6.90

0.26

0.53

0.70

0.15

0.73

1.01

0.49

0.12

1.14

1.09

0.55

0.21

0.84

116

0.95

6.82

0.67

1.83

0.38

0.04

0.24

0.10

2.45

0.44

4.00

1.02

0.42

1.74

0.99

38

0.95

6.61

0.71

2.03

0.38

0.23

0.56

0.87

0.16

0.16

0.55

0.48

0.61

0.19

0.77

7

118

0.94

6.35

0.25

0.52

0.97

1.19

0.03

0.07

3.96

6.27

0.33

0.46

2.16

5.16

0.99

7

331

0.94

6.31

0.74

2.23

0.48

1.24

0.22

0.07

0.32

0.76

0.30

0.05

1.72

0.69

0.87

7

23

0.94

6.28

0.30

0.64

1.69

1.01

0.35

1.06

8.89

8.16

3.65

7.14

2.70

9.52

0.99

7

47

0.94

6.16

0.04

0.08

0.69

0.12

0.28

1.02

4.52

0.77

3.15

7.38

0.81

3.08

0.99

7

58

0.94

6.02

0.84

3.08

0.65

0.16

0.42

0.70

0.37

0.12

0.54

0.54

0.81

0.32

0.84

7

389

0.93

5.80

0.58

1.41

0.81

0.37

0.06

0.32

1.48

1.15

0.25

0.87

1.19

1.41

0.83

7

8

7

7

170.84

17.95

47.89

15.99

194.68

80.66

18.15

1.00

114

0.92

5.11

0.86

3.44

0.98

1.26

0.50

0.28

9.71

4.95

2.37

1.04

2.24

7.23

0.99

7

78

0.91

4.78

0.84

3.06

0.64

0.46

0.57

0.63

3.30

1.03

1.65

1.35

1.10

1.96

0.98

7

11

0.91

4.77

0.56

1.36

0.52

0.56

0.59

0.99

0.18

0.33

0.46

0.38

1.09

0.28

0.74

7

10

0.90

4.64

0.21

0.43

0.52

0.03

0.29

0.55

0.13

0.02

0.15

0.17

0.48

0.09

0.38

154

0.90

4.59

0.89

3.86

0.70

0.35

0.38

0.27

0.72

0.53

0.94

0.44

0.34

0.22

0.92

7

38

0.89

4.28

0.91

4.34

2.43

1.12

0.12

0.31

1.55

1.02

0.19

0.28

3.56

1.80

0.95

7

57

0.88

4.22

0.01

0.03

0.30

1.33

0.01

0.44

0.14

0.80

0.01

0.23

1.63

0.45

0.68

7

14

0.88

4.19

0.79

2.58

1.41

1.63

0.08

0.17

3.38

1.26

0.13

0.20

3.04

1.94

0.96

7

93

0.88

4.10

0.51

1.17

0.39

0.53

0.14

0.10

0.19

0.43

0.15

0.07

0.91

0.31

0.54

7

12

0.87

3.95

0.28

0.59

1.73

1.98

0.00

0.02

0.38

3.81

3.72

205.65

1.00

7

61

0.86

3.80

0.51

1.18

2.12

1.62

1.89

2.71

2.11

1.70

3.49

4.25

3.75

1.93

0.99

7

40

0.86

3.79

0.46

1.03

0.91

0.00

0.55

0.22

0.56

0.00

0.99

0.23

0.91

0.33

0.75

7

70

0.86

3.74

0.39

0.84

0.20

0.03

0.30

1.19

0.15

0.03

0.65

1.40

0.23

0.11

0.92

7

81

0.86

3.72

0.78

2.53

1.55

1.09

1.49

0.68

2.23

2.74

5.13

1.23

0.46

0.53

1.00

7

45

0.86

3.70

0.69

1.91

1.92

1.03

0.04

0.46

1.64

1.63

0.07

0.53

2.95

1.79

0.93

7

35

0.85

3.68

0.64

1.66

0.36

0.35

0.80

1.47

0.46

0.60

2.23

2.45

0.71

0.54

0.97

7

8

0.85

3.62

0.43

0.95

3.96

4.09

3.34

7.26

2.93

3.93

4.56

5.07

8.05

3.88

0.98

7

90

0.85

3.55

0.67

1.79

1.34

0.61

0.07

0.26

6.07

4.89

0.77

1.70

1.94

5.87

0.99

7

82

0.84

3.43

0.15

0.31

2.16

1.13

0.81

1.24

2.08

1.99

1.91

1.82

3.29

2.12

0.83

7

31

0.84

3.42

0.72

2.06

1.23

1.76

0.59

0.20

0.98

0.50

0.27

0.13

2.99

0.65

0.75

7

155.52

275.61

569

0.83

3.32

0.32

0.67

0.93

0.64

0.26

0.11

1.39

1.68

0.95

0.23

1.57

1.55

0.89

7

15

0.82

3.24

0.74

2.23

2.26

1.27

0.47

0.09

0.83

0.48

0.55

0.07

3.53

0.67

0.78

7

17

0.82

3.23

0.32

0.69

0.87

2.52

0.26

1.72

0.39

1.54

0.33

0.81

3.38

0.89

0.95

7

39

0.82

3.22

0.14

0.27

5.32

2.53

3.26

4.08

1.61

1.16

1.98

1.77

7.85

1.51

0.91

7

Case5 11cv 02509LHK

Document424

Filed0517 13

2

Page57

of

62

Exhibit 2
Intel

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

2

Section

Change

Correlation

Coeff

Stat

Correlation

T

Coeff

3

Section

Regression

Contemp

Stat

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

T

Section

Net

Stats

Revenue

SJ

Emp

C

L

5

Section

6

Effect

T

Stat

r2

7

12

0.60

1.69

0.69

1.90

1.46

4.18

1.01

6.31

1.14

1.33

0.55

2.46

5.64

1.53

0.99

7

16

0.60

1.67

0.68

1.85

0.81

0.54

0.75

0.35

0.11

0.15

0.24

0.07

1.35

0.14

0.82

224

0.59

1.64

0.33

0.70

1.58

0.63

0.86

0.91

1.44

1.03

2.22

1.42

2.21

1.32

0.87

7

27

0.59

1.62

0.59

1.47

2.44

0.08

0.44

1.28

0.37

0.02

0.24

0.25

2.36

0.24

0.75

7

52

0.58

1.59

0.79

2.62

0.26

0.23

0.06

0.07

0.49

0.88

0.33

0.20

0.03

0.04

0.92

7

31

0.54

1.45

0.67

1.83

2.56

0.74

0.13

0.92

0.54

0.30

0.06

0.26

3.30

0.50

0.76

878

0.50

1.31

0.59

1.48

1.85

0.75

0.70

0.51

7

88

0.49

1.27

0.79

2.57

0.24

0.32

0.37

0.19

7

9

0.49

1.26

0.61

1.34

2.99

4.04

3.39

3.51

7

14

0.42

1.04

0.60

1.48

5.60

3.82

2.25

7

15

0.39

0.95

0.62

1.59

7.30

2.86

2.56

7

68

0.38

0.91

0.51

1.17

4.24

1.64

7

34

0.36

0.85

0.62

1.57

0.33

7

11

0.34

0.81

0.14

0.27

3.17

7

12

0.31

0.74

0.60

1.29

8.69

7

47

0.24

0.55

0.29

0.61

2.29

1.15

0.73

0.52

7

24

0.12

0.26

0.09

0.17

4.06

2.08

1.49

2.58

7

7

46.85

35.03

2.49

6.95

1.36

2.90

3.35

5.09

1.33

1.91

0.65

0.02

2.67

1.80

11.14

12.08

45.73

19.62

11.05

3.14

2.25

2.54

0.88

4.58

3.84

3.05

0.76

0.62

0.88

0.79

0.24

0.27

1.18

0.05

1.00

0.34

0.63

0.44

2.60

43.87

1.00

0.08

0.54

1.00

9.42

3.07

0.96

5.44

0.97

5.88

0.72

0.64

0.27

0.32

0.19

0.93

0.15

5.84

0.45

0.55

0.29

3.44

0.65

0.46

10.16

6.51

0.66
18.07

0.61

20.25

0.66
14.21

15.65

6.14

19.43

1.00

14

0.08

0.17

0.24

0.50

0.71

0.45

0.92

0.81

0.43

0.54

1.27

0.70

0.27

0.11

0.95

187

0.08

0.17

0.37

0.78

0.10

0.08

0.12

0.82

0.08

0.12

0.24

0.97

0.18

0.10

0.77

7

10

0.18

0.42

0.29

0.62

5.13

4.95

4.84

5.17

5.01

0.98

7

15

0.22

0.50

0.53

1.26

1.02

0.23

0.62

0.63

0.20

0.09

0.27

0.16

0.79

0.10

0.56

7

17

0.43

1.07

0.48

1.10

5.55

2.37

2.37

1.63

3.35

2.98

3.45

1.37

7.92

3.33

0.96

201

0.97

7.68

0.90

3.51

6

98

0.96

7.13

0.97

6.67

6

8

0.96

6.83

0.92

4.03

6

222

0.95

5.98

0.92

4.09

6

8

0.95

5.93

0.72

1.48

6

28

0.93

5.17

0.09

0.15

6

72

0.92

4.79

0.48

0.95

6

17

0.92

4.72

0.83

2.13

6

25

0.91

4.36

0.24

0.35

131

0.91

4.26

0.91

3.08

6

12

0.90

4.06

0.78

1.78

6

18

0.90

4.03

0.86

2.35

402

0.89

3.99

0.79

2.26

6

41

0.89

3.96

0.90

2.05

6

77

0.89

3.95

0.77

2.12

6

12

0.88

3.76

0.76

1.68

6

36

0.88

3.74

0.03

0.05

6

8

0.87

3.57

0.13

0.22

6

93

0.87

3.55

0.56

1.16

7

7

6

6

6

15.96

30.79

17.07

35.69

46.74

Case5 11cv 02509LHK

Document424

2

Filed0517 13

Page58

of

62

Exhibit 2
Intel

Section
Years
Job

Title

of

1

Section

Total

Level

Emp Years

Data

T

2

Section

Change

Correlation

Coeff

Stat

Correlation

T

Coeff

149

0.71

2.03

0.98

7.11

6

22

0.68

1.86

0.36

0.66

6

10

0.61

1.56

0.98

8.07

6

8

0.55

1.32

0.12

0.18

6

14

0.52

1.20

0.93

2.58

6

10

0.51

1.19

0.65

1.47

6

34

0.51

1.18

0.76

1.67

6

15

0.49

1.14

0.50

0.99

6

9

0.42

0.93

0.35

0.52

6

31

0.41

0.90

0.16

0.16

6

12

0.27

0.55

0.80

2.34

6

8

0.24

0.49

0.33

0.61

6

13

0.23

0.47

0.89

2.81

6

10

0.21

0.42

0.67

1.28

6

40

0.18

0.37

0.60

1.29

6

24

0.09

0.18

0.42

0.65

6

11

0.02

0.04

0.58

1.23

10

0.41

0.90

0.20

0.20

170

0.74

2.21

0.06

0.10

6

6

6

3

Regression

Contemp

Stat

Revenue

Section

Coefficients

Lagged

Regression
SJ

Emp

Contemp

Lagged

4

T

Section

Net

Stats

Revenue

SJ

Emp

C

L

5

Section

Effect

T

Stat

r2

6

Case5 11cv 02509LHK

Document424

Exhibit

Filed0517 13

2

Page59

62

of

2

Intuit

Section

Job

Title

of

1

Section

Total

Years

Emp

Data

Level

Years

T

Coeff

11

2981

11

2

Section

Change

Correlation

T

Coeff

Stat

0.60

2.26

0.97

597

0.59

2.18

0.95

11

293

0.54

1.91

0.97

11

150

0.40

1.29

11

140

0.26

10

170

10

1571

10

69

10

194

9

9

57
1073

Correlation

3

Section

Regression

Contemp

Stat

Coefficients

Lagged

Revenue

Regression
SJ

Emp

12.05

1.50

1.01

0.26

0.34

8.57

1.13

1.33

0.48

0.04

11.05

1.50

1.17

0.49

0.76

3.31

2.01

1.70

0.81

0.05

0.13

0.69

0.78

3.55

0.98

10.93

0.55

1.85

0.79

0.49

1.60

0.40

Contemp
10.44

Lagged

4

T

Section

Net

Stats

Revenue

SJ

Emp

C

5

T

L

Section

6

Effect
Stat

r2

2.21

1.05

1.42

2.51

4.97

0.99

8.97

3.99

3.14

0.29

2.46

5.57

0.98

0.08

8.38

2.13

1.64

0.29

2.67

3.97

0.97

0.80

0.27

4.41

1.72

1.21

0.33

3.71

2.77

0.87

1.28

0.43

1.77

1.41

2.27

0.74

2.01

1.97

2.14

0.71

1.08

0.18

0.15

0.12

4.91

0.37

0.47

0.23

0.89

1.50

0.97

3.16

1.34

1.01

0.36

0.02

6.15

3.76

0.14

2.35

0.30

0.78

0.19

0.68

0.18

0.15

0.28

1.47

0.42

0.17

0.50

0.57

0.52

1.25

0.76

2.86

1.39

1.36

0.33

0.44

1.89

0.78

0.27

0.43

2.75

1.12

0.94

0.67

2.39

0.08

0.21

0.62

0.82

0.05

0.38

0.53

0.91

0.07

0.24

1.44

0.92

0.40

0.64

2.22

0.69

2.34

1.15

0.25

0.30

0.41

3.94

0.68

1.77

0.85

1.40

2.74

0.89

13.75

11.01

0.99

9

94

0.59

1.94

0.57

1.56

1.10

0.36

0.01

1.56

2.52

0.28

0.01

2.86

1.47

1.11

0.90

9

81

0.54

1.70

0.77

2.94

1.63

1.09

0.15

0.23

4.23

1.86

0.49

0.46

2.71

4.12

0.92

0.53

1.67

0.68

2.05

0.34

0.90

0.56

0.09

0.33

0.28

0.37

0.02

0.56

0.14

0.51

0.17

0.46

0.74

2.70

2.01

0.71

0.11

0.23

2.20

0.66

0.16

0.18

2.73

2.07

0.75

9

9

758

46

9

486

0.01

0.02

0.46

1.28

1.34

1.60

0.55

0.31

4.91

3.62

2.13

0.90

2.94

4.97

0.94

8

113

0.80

3.25

0.91

4.90

0.44

0.22

1.21

2.04

1.78

0.33

2.25

5.15

0.66

0.73

1.00

8

24

0.68

2.25

0.72

2.32

1.52

2.13

0.81

0.39

0.39

0.24

0.10

0.06

3.65

0.29

0.83

8

29

0.61

1.87

0.76

2.62

2.07

2.81

1.72

0.60

1.19

0.79

0.53

0.22

4.88

0.93

0.83

0.46

1.25

0.81

3.08

1.40

1.62

1.07

0.50

0.84

0.48

0.34

0.17

3.01

0.61

0.74

0.33

0.87

0.04

0.10

0.37

0.68

0.51

1.04

0.95

0.84

0.41

0.53

1.05

0.99

0.97

8

8

114

22

8

177

0.33

0.85

0.94

5.94

2.15

2.42

2.11

1.22

2.70

1.46

1.39

0.96

4.57

1.88

0.95

8

206

0.63

2.00

0.13

0.30

1.48

5.60

4.14

2.16

1.84

1.56

1.55

1.36

7.08

1.74

0.93

7

48

0.82

3.26

0.65

1.73

2.10

0.32

0.98

3.09

6.73

1.26

4.45

4.93

2.42

8.37

0.99

7

22

0.74

2.48

0.87

3.60

2.05

1.38

0.10

0.31

1.40

1.17

0.13

0.20

3.43

2.57

0.93

7

7

0.72

2.33

0.86

3.41

3.15

0.40

0.59

0.24

1.69

0.08

0.14

0.08

3.54

0.77

0.95

7

43

0.70

2.17

0.54

1.28

0.89

1.50

0.51

0.15

2.01

1.58

1.42

0.52

2.39

1.78

0.82

0.65

1.93

0.79

2.61

1.31

2.39

0.84

0.14

6.24

3.53

3.28

0.44

3.70

5.12

0.98

0.62

1.75

0.71

2.01

0.76

3.57

1.38

2.21

0.73

1.19

0.94

1.30

4.33

1.71

0.87

7

7

354

58

7

110

0.31

0.72

0.45

1.01

0.86

1.35

0.69

2.45

2.20

2.04

1.87

3.56

0.49

0.54

0.99

7

143

0.21

0.48

0.90

4.19

1.05

0.28

0.30

0.40

5.44

0.61

1.27

1.51

0.77

1.34

0.98

0.04

0.10

0.21

0.43

1.11

1.49

0.29

2.38

0.35

0.28

0.06

0.55

2.60

0.31

0.83

0.10

0.23

0.09

0.18

1.45

2.96

1.25

0.62

2.56

3.71

3.39

1.23

4.41

3.45

0.93

0.33

0.78

0.12

0.25

0.39

1.05

1.03

0.99

0.83

1.32

2.01

2.15

0.66

0.61

0.96

0.55

1.49

0.73

2.11

1.15

4.61

3.16

0.29

1.93

1.31

1.27

0.30

5.76

1.56

0.86

7

7

7

7

26
136

16
378

7

25

0.73

2.36

0.14

0.28

0.19

0.70

0.18

0.23

0.04

0.23

0.07

0.03

0.52

0.08

0.62

7

15

0.83

3.37

0.60

1.52

0.27

0.93

0.52

2.08

0.62

1.93

1.36

4.48

1.20

1.59

0.98

6

16

0.95

6.25

0.98

8.84

0.93

5.09

0.93

4.44

6

180

Case5 11cv 02509LHK

Document424

Exhibit

Filed0517 13

2

Page60

62

of

2

Intuit

Section

Job

Title

of

1

Section

Total

Years

Emp

Data

Level

Years

T

Coeff

2

Section

Change

Correlation

T

Coeff

Stat

Correlation

Contemp

Stat

6

96

0.71

2.02

0.95

5.47

6

39

0.71

2.01

0.74

1.93

6

91

0.71

2.00

0.49

0.97

6

8

0.69

1.92

0.68

1.62

6

26

0.67

1.81

0.19

0.33

6

26

0.58

1.41

0.28

0.51

6

31

0.57

1.39

0.77

2.08

6

9

0.54

1.27

0.38

0.71

6

8

0.52

1.22

0.78

2.14

6

405

0.46

1.02

0.60

1.30

6

230

0.43

0.96

0.69

1.63

6

14

0.42

0.93

0.36

0.67

6

23

0.41

0.91

0.09

0.15

6

15

0.40

0.88

0.17

0.30

6

8

0.38

0.82

0.03

0.06

6

12

0.38

0.81

0.44

0.85

6

18

0.35

0.75

0.27

0.49

6

78

0.33

0.70

0.38

0.70

6

38

0.33

0.69

0.85

2.82

0.29

0.60

0.09

0.15

0.28

0.58

0.59

1.27

0.23

0.48

0.66

1.51

6

6

6

115

37
102

6

74

0.07

0.14

0.05

0.09

6

24

0.05

0.10

0.48

0.94

0.01

0.01

0.43

0.82

6

338

6

17

0.00

0.01

0.30

0.55

6

6

0.05

0.09

0.13

0.23

6

16

0.09

0.17

0.15

0.26

6

54

0.12

0.25

0.93

4.33

6

98

0.13

0.27

0.81

2.40

0.24

0.50

0.34

0.63

6

179

3

Regression

6

23

0.26

0.54

0.09

0.16

6

19

0.29

0.61

0.07

0.13

6

35

0.36

0.78

0.83

2.61

6

18

0.38

0.83

0.22

0.40

6

15

0.40

0.87

0.53

1.08

6

16

0.46

1.02

0.80

2.29

6

10

0.47

1.06

0.69

1.36

6

38

0.85

3.22

0.92

3.98

Lagged

Revenue

Section

Coefficients

Regression
SJ

Emp

Contemp

Lagged

4

T

Section

Net

Stats

Revenue

SJ

Emp

C

L

5

Section

Effect

T

Stat

r2

6

Case5 11cv 02509LHK

Document424

Exhibit

Filed0517 13

2

Page61

62

of

2

Pixar

Section

Job

of

Title

1

Section

Total

Years

Level

Emp Years

Data

T

Coeff

2

Section

Change

Correlation

Correlation

T

Coeff

Stat

3

Section

Regression

Stat

Contemp

Coefficients

Lagged

Revenue

Regression
SJ

Emp

Contemp

Lagged

4

Section

T Stats

Net

Revenue

SJ

Emp

C

L

5

Section

6

Effect

T

Stat

r2

TECHNICAL DIRECTOR

11

1872

0.94

8.31

0.89

5.65

0.55

0.31

0.03

0.02

3.08

0.63

0.60

0.06

0.86

1.32

0.82

ARTIST SKETCH

11

141

0.91

6.64

0.82

4.06

1.29

1.53

0.12

0.18

7.17

4.44

1.77

0.40

2.82

6.78

0.94

ENGINEER

SOFTWARE

11

503

0.91

6.41

0.93

7.25

0.95

0.70

0.01

0.25

6.38

1.64

0.14

0.62

1.65

3.78

0.91

ANIMATOR

SUPERVISING

11

70

0.82

4.35

0.89

5.41

0.23

2.42

0.22

2.26

0.18

1.94

1.18

1.85

2.65

5.34

0.89

11

772

0.81

4.21

0.78

3.53

0.55

0.48

0.06

0.82

5.27

1.97

1.47

3.57

1.03

3.32

0.92

11

44

0.77

3.57

0.89

5.59

1.79

3.71

0.06

2.65

1.16

2.92

0.44

2.22

1.92

3.94

0.92

11

129

0.75

3.37

0.79

3.68

0.91

1.27

0.15

0.47

3.97

3.23

1.90

0.79

2.18

5.50

0.92

ANIMATOR
ANIMATOR
LAYOUT

DIRECTING

ARTIST

ENGINEER

SR SOFTWARE

11

53

0.74

3.31

0.79

3.59

0.70

1.61

0.00

0.79

1.75

2.89

0.03

1.11

2.32

5.27

0.89

DESIGNER

PRODUCTION

11

62

0.73

3.20

0.86

4.86

0.52

2.50

0.22

3.16

0.22

1.55

0.97

1.44

1.98

2.14

0.83

11

73

0.72

3.10

0.75

3.21

0.53

1.60

0.05

0.10

0.86

2.81

0.33

0.10

2.12

4.47

0.83

ART DIRECTOR

11

70

0.70

2.95

0.76

3.26

1.18

0.70

0.04

1.55

4.33

1.74

0.33

1.81

1.89

3.36

0.83

ENGINEER

11

54

0.58

2.16

0.82

4.06

0.72

1.11

0.24

0.86

1.07

1.77

1.00

0.75

1.83

3.79

0.80

11

91

0.56

2.04

0.81

3.97

1.07

0.56

0.12

0.70

5.49

2.03

1.65

1.48

1.63

4.81

0.90

11

247

0.55

1.98

0.46

1.48

1.27

1.09

0.01

0.41

2.96

2.26

0.07

0.43

2.36

2.98

0.70

11

11

0.51

1.79

0.81

3.89

1.08

0.42

0.01

1.19

4.76

1.69

0.09

1.88

1.50

4.24

0.86

SYSTEMS ADMINISTRATOR

11

133

0.50

1.75

0.29

0.86

0.74

1.15

0.06

0.16

1.93

2.43

0.51

0.20

1.89

2.50

0.62

SCIENTIST

11

62

0.50

1.74

0.39

1.21

1.06

1.26

0.09

0.07

2.05

2.72

0.49

0.06

2.31

2.91

0.68

ANIMATOR

FIX

QUALITY ASSURANCE

SYSTEMS ADMINISTRATOR

SR

ARTIST STORY

MGR DESKTOP

SYSTEMS

SR

TECH DIRECTOR

SUPERVISING

11

70

0.49

1.67

0.72

2.95

1.91

0.66

0.15

3.54

4.54

1.97

0.89

3.08

2.56

4.81

0.87

MGR FINANCIAL

SYSTEMS

11

11

0.43

1.41

0.84

4.41

0.91

0.34

0.00

0.90

5.48

1.95

0.03

2.06

1.24

4.99

0.88

MANAGER

11

11

0.42

1.38

0.83

4.20

0.88

0.24

0.08

0.56

4.82

1.10

1.22

1.12

1.12

3.60

0.86

11

11

0.42

1.38

0.88

5.34

0.84

0.21

0.04

0.53

5.76

1.20

0.67

1.39

1.05

4.31

0.88

11

42

0.42

1.37

0.63

2.29

1.15

0.84

0.08

1.67

3.63

2.51

0.76

1.85

1.98

3.68

0.79

11

24

0.38

1.22

0.86

4.72

0.60

0.02

0.09

0.13

4.06

0.10

1.73

0.36

0.62

2.11

0.84

SVCS

11

11

0.34

1.09

0.84

4.35

0.95

0.24

0.06

0.73

4.89

1.21

0.87

1.37

1.19

4.01

0.86

PRODUCTS

11

11

0.21

0.63

0.79

3.66

1.01

0.25

0.03

1.20

4.52

1.44

0.42

2.01

1.25

4.30

0.85

11

44

0.19

0.59

0.26

0.75

0.57

0.92

0.18

1.39

2.12

3.91

1.80

1.63

1.49

3.88

0.85

11

22

0.17

0.52

0.41

1.29

0.84

0.35

0.07

1.70

4.85

2.20

1.10

4.11

1.19

4.57

0.92

11

35

0.12

0.36

0.12

0.35

0.77

0.92

0.01

1.08

1.17

1.57

0.04

0.60

1.69

1.58

0.39

10

35

0.50

1.62

0.71

2.65

1.47

0.68

0.03

4.53

2.67

2.62

0.15

2.08

2.15

3.58

0.85

MGR SYSTEMS

10

10

0.41

1.28

0.74

2.66

1.03

0.40

0.20

2.10

3.42

1.19

0.93

1.93

1.44

2.70

0.81

ENGINEER

10

15

0.28

0.83

0.68

2.45

1.10

0.49

0.02

0.34

2.08

1.33

0.06

0.13

1.59

2.68

0.67

VP SOFTWARE ENGINEERING

10

12

0.26

0.76

0.56

1.79

3.29

0.66

0.72

9.33

2.20

1.19

1.18

2.35

3.95

2.37

0.89

USER INTERFACE DESIGNER

10

20

0.14

0.40

0.66

2.35

0.65

0.35

0.02

0.43

1.94

1.17

0.19

0.35

0.99

2.17

0.61

9

9

0.34

0.95

0.78

3.01

1.66

0.14

0.12

2.32

3.77

0.40

0.55

1.91

1.80

3.41

0.88

9

15

0.17

0.45

0.43

1.07

1.85

1.06

0.30

1.74

5.23

12.17

6.25

4.55

2.92

7.02

0.99

8

25

0.58

1.73

0.73

2.36

0.34

1.69

0.31

2.68

0.22

2.03

0.66

1.00

1.35

1.15

0.85

8

13

0.35

0.92

0.63

1.60

0.56

0.96

0.85

6.04

17.44

10.07

16.87

1.52

8

8

0.34

0.89

0.81

3.05

1.03

0.02

0.22

2.32

6.00

0.08

2.32

3.35

1.05

3.89

0.97

8

20

0.27

0.70

0.03

0.06

0.05

0.57

0.11

1.05

0.10

2.80

0.51

0.29

0.52

0.90

0.86

ENGINEERING
ENGINEER
ARTIST

ASSOCIATE

GRAPHIC

TECH DEPT
TECH DIRECTOR LEAD CRTV
ADMINISTRATOR
DEVELOPER

RENDERMAN

TECH DIRECTOR

CRTV

SVCS

SCULPTOR
ENGINEER
PROJECT

PRODUCTION

MGR STUDIO

TOOLS

OPERATIONS
RENDERMAN
SUPPORT

DIR RENDERMAN
DESIGNER

SUPPORT

PRODUCT

DEV

ENVIRONMENTAL

ARTIST AFTER EFFECTS
TECHNICAL WRITER
TECHNICAL LEAD RENDERING
ARTIST STORY DEVELOPMENT

11.18

20.27

1.00

Case5 11cv 02509LHK

Document424

Exhibit

Filed0517 13

2

Page62

62

of

2

Pixar

Section

Job

of

Title

ENGINEER

SOFTWARE

ENGINEER

IMAGE

TECHSUPPORT

1

Section

Total

Years

Level

Emp Years

Data

T

Coeff

2

Section

Change

Correlation

Correlation

T

Coeff

Stat

Stat

7

7

0.86

3.77

0.01

0.03

6

8

0.92

4.74

0.54

1.13

6

6

0.92

4.65

0.75

1.97

6

6

0.88

3.76

0.79

2.24

6

6

0.88

3.69

0.78

2.18

6

65

0.74

2.20

0.53

1.07

6

6

0.60

1.50

0.76

2.00

6

6

0.52

1.22

0.57

0.98

FINANCIAL APPS DEVELOPER

6

6

0.46

1.03

0.80

2.31

MGR SR PROJECT

6

6

0.46

1.03

0.21

0.31

6

6

0.42

0.93

0.27

0.49

6

8

0.12

0.24

0.35

0.66

TECHNICAL
ENGINEER

LEAD TELECOM
ROOM

SCREENING

MGR IMAGE
CGI

MASTERING

MASTERING

PAINTER

DESIGNER

CAMERA

ENGINEER

APPLICATIONS

STUDIO

TOOLS

LAYOUT ARTIST LEAD
MEDIA SYSTEMS COORDINATOR

3

Section

Regression

Contemp
0.51

Coefficients

Lagged

Revenue

0.02

0.01

Regression
SJ

Emp
2.20

Contemp
0.63

Lagged
0.07

4

Section

T Stats

Net

Revenue
0.03

SJ

Emp
1.07

C

L
0.49

5

Section

6

Effect

T

Stat
0.55

r2
0.58