This document is available in two formats: this web page (for browsing content), and PDF (comparable to original document formatting). To view the PDF you will need Acrobat Reader, which may be downloaded from the Adobe site. For an official signed copy, please contact the Antitrust Documents Group.

Claude F. Scott, Esq.
Pam Cole, Esq. (CA Bar No. 208286)
Renata B. Hesse, Esq.
Phillip R. Malone, Esq.
U.S. DEPARTMENT OF JUSTICE, ANTITRUST DIVISION
450 Golden Gate Avenue, Rm. 10-0101
San Francisco, CA 94102-3478
(415) 436-6660
(415) 436-6683 (Fax)

Attorneys for Plaintiff the United States of America

Also filed on behalf of 10 Plaintiff States (See signature block)


UNITED STATES DISTRICT COURT

NORTHERN DISTRICT OF CALIFORNIA

SAN FRANCISCO DIVISION



UNITED STATES OF AMERICA, et al.,    

                  Plaintiffs,

                  v.

ORACLE CORPORATION

                  Defendant.


|
|
|
|
|
|
|
|
|
|
|
|         


CASE NO. C 04-0807 VRW

Filed July 8, 2004

PLAINTIFFS' PROPOSED
FINDINGS OF FACT

     PUBLIC VERSION


TABLE OF CONTENTS

  1. Background
  2. 1.1 Oracle Corporation

    1.2Oracle Sells High Function HRM and FMS Software

    1.3PeopleSoft, Inc

    1.4Oracle's Unsolicited Offer for PeopleSoft

    1.5Jurisdictional Requirements

    1.6Witnesses

  3. Product and Industry Characteristics
  4. 2.1Product and Industry Definitions

    2.2"High-Function Software" Is a Software Product That Is Capable of Executing a Wide Array of Business Processes at a Superior Level of Performance

    2.3Characteristics of Vendors of High-Function Software Are Important to Purchasers of High-Function Software

    2.4The Vendors' High Function Software Is Differentiated, i.e., Not All Alike, and Different Vendors Have Different Strengths and Weaknesses

    2.5After a Review of 148 Vendors, Professor Iansiti Concluded That from the Standpoint of Functionality, Only Oracle, PeopleSoft and SAP Are Able to Satisfy the High Function Requirements of Enterprises

    2.6Mid-Market Software

    2.7Sales Process for High-Function Software: the Sales Process Is Lengthy and Involves Extensive Sharing of Information in Order to Identify a Product That Can Meet the Customer's Functional Requirements

  5. Product Market
  6. 3.1Definitions

    3.2Customers Consistently Identify Oracle, PeopleSoft, and SAP as the Only Vendors of HRM or FMS Software That Can Automate Complex Human Resource and Financial Management Processes

    3.3Oracle, Peoplesoft, and SAP Recognize That They Are Competing in a Three Firm Market to Supply High-Function Software to Organizations with Complex Requirements

    3.4Organizations with Complex Human Resource or Financial Management Requirements Do Not View "Mid-Market" HRM or FMS Software Applications as a Substitute for the High-Function Software Sold by Oracle, PeopleSoft, and SAP

    3.5SAP, PeopleSoft, and Oracle Also Compete in the Market for Small Business and Mid-Market Accounts

    3.6Customers Would Not Switch to a "Best of Breed" Solution (Or "Point Solution") in Response to a Small, but Significant, Nontransitory Price Increase in High Function FMS or HRM Software Because These Solutions Lack the Functionality Required by High-function Software and Are Costlier to Implement and Maintain

    3.7Customers Would Not Develop FMS or HRM Software in Response to a Small, but Significant, Nontransitory Price Increase in High Function FMS or HRM Software

    3.8Customers' Threats of Using Their Incumbent HRM or FMS Software Would Not Prevent a Significant Price Increase in High Function FMS or HRM Software

    3.9Large and Complex Customers Would Not Outsource in Response to a Small, but Significant, Non-Transitory Price Increase in High Function HRM Software

    3.10Outsourcing Vendors Using Their Own Proprietary Software Do Not Compete with Oracle, PeopleSoft, and SAP for Organizations with Complex HR Requirements

    3.11All the Theoretically Available Options (Listed Above) to High Function Software, Taken Together, Would Not Make a SSNIP in High Function Software Unprofitable

  7. Geographic Market
  8. 4.1Selling High Function HRM and FMS Software in the United States Requires That a Vendor Have a Presence in the United States

    4.2Nothing Links Together Customers of High Function HRM and FMS Software, Either Within the United States or Between the United States and Elsewhere

    4.3Oracle and PeopleSoft Ensure That Their High Function HRM and FMS Software Contains Up-to-date Country-specific Functional Characteristics, Making Arbitrage More Difficult and Price Discrimination Possible

    4.4The Effects of the Proposed Acquisition on Customers Would Differ Significantly Both among Customers in the United States and Between Customers in the United States and Elsewhere

    4.5Defining a United States Geographic Market Does Not Exclude Any Relevant Sellers on the Market's Supply Side

    4.6Other Than SAP, There Are No Vendors Outside Of The United States That Would Be Positioned To Serve Customers' Needs For High Function HRM and FMS Software in The Face Of A SSNIP

  9. Firms in the Market
  10. 5.1Oracle, PeopleSoft, and Sap Are in a Different Class from All Other Vendors of Enterprise Resource Planning Software in Their Ability to Meet the Needs of Large and Complex Enterprises

    5.2Expert Testimony Confirms That Oracle, PeopleSoft and Sap Are the Three Rivals That Determine Pricing in the Relevant Markets

    5.3Oracle, Peoplesoft, and SAP Primarily View Each Other as Competitors for the Sale of High-function Software

    5.4Other Firms are Not in the Market

    5.5Lawson Lacks Functional and Vendor Characteristics to Serve as a Competitive Restraint in the High Function Software Market

    5.6Lawson Counts Several Large Organizations as Customers but a Number of Those Organizations Are Moving off Lawson Having Made the Decision to Acquire High Function Software from Oracle, Peoplesoft and SAP

    5.7Lawson Products Have Problems When Used by Large or Complex Customers

    5.8SSA/Baan is a Mid-Market Firm and Not a Seller of High- Function Software

    5.9Microsoft is a seller of software for the mid-market (and smaller firms), not high function software for large and complex enterprises

    5.10Oracle, PeopleSoft and SAP View Microsoft as Solely a Small to Mid-Market Competitor

    5.11Customers Do Not Believe Microsoft Provides High Function Software

    5.12Microsoft Does Not Provide Customers With High-Function Software

    5.13AMS is a Supplier of FMS Software Solely Tailored for the Public Section

    5.14Hyperion's Software is Complementary, Not Competitive, to Oracle

    5.15IBM Is Not a Participant in the Market

  11. Market Concentration
  12. 6.1Computing Market Shares and Concentration Statistics Within Each Relevant Market is a Standard Feature of Merger Guidelines Analysis That Professor Elzinga Has Employed

    6.2In Estimating Market Shares and Concentration Statistics, Professor Elzinga Has Relied on Non-public Sales Data From Oracle, PeopleSoft and Third Party Vendors, Obtained Through Compulsory Process

    6.3For the U.S. High-function FMS Software Market in 2003, Professor Elzinga Calculates Market Shares of 39% for SAP, 31% for PeopleSoft, and 17% for Oracle. Oracle and PeopleSoft Thus Have a Combined Share of Nearly 50%

    6.4Professor Elzinga Calculated Shares and Concentration Statistics For the U.S. High-function HRM Software Market in Similar Fashion to His Treatment of the FMS Market

    6.5Even If the Appropriate Geographic Market Were Worldwide, The Merger of Oracle and PeopleSoft Would Result in Increased Concentration in an Already Highly Concentrated Market

  13. Anticompetitive Effects - Concentration and Likely Consumer Harm. The Merger of Oracle and PeopleSoft May Tend to Lessen Competition Substantially
  14. 7.1The Proposed Merger Will Eliminate Head-to-Head Competition Between Oracle and PeopleSoft And, from the Vantage Point of Many Customers, Will Constitute a "Three-to-Two" Merger: Reducing the Number of Meaningful Competitors from Three Independent Centers of Initiative to Two

    7.2Customers Benefit from Competition Between Oracle and PeopleSoft Because it Leads to Innovation, Quality Service, and Lower Prices. The Proposed Merger Will Eliminate That Competition and Thus Directly Harm Customers

    7.3The Merger Is Likely to Produce Harmful Unilateral Effects

    7.4The Merger Raises Concern for Possible Additional Harmful Anticompetitive Effects

    7.5The Predicted Harmful Effects Are Likely to Be Substantial. The Volume of Commerce in FMS and HRM Software Affected By the Merger of Oracle and PeopleSoft Would Be Large

  15. Anticompetitive Effects -- Entry/Repositioning
  16. 8.1JD Edwards' (JDE) Failed Attempt to Enter The High-Function Software Markets Shows That Entry is Difficult

    8.2Oracle's, PeopleSoft's and SAP's Entry into and Repositioning Within the High-function Software Markets Has Been Time-Consuming and Expensive

    8.3Microsoft Has No Plans to Enter the High-function HRM or FMS Markets, And, Even Were it to Develop Such Plans, Its Entry Could Not Be Timely or Sufficient to Alleviate the Anticompetitive Effects of the Merger

    8.4Lawson Is Not Likely to Enter the High-Function Software Market

    8.5Other Firms Are Not Likely to Enter Into the High-Function Software Markets

    8.6Further Evolution of the "Infrastructure Layer" Will Not Make New Entry Into the High-Function Applications Markets Timely, Likely, or Sufficient

    8.7Expert Testimony Supports the Conclusion That Entry or Repositioning Is Unlikely to Stem the Anticompetitive Effects of the Proposed Merger

  17. Anticompetitive Effects - Efficiencies. Oracle Cannot Demonstrate That Verifiable, Merger-Specific Efficiencies Will Reverse the Anticompetitive Effects of the Merger
  18. 9.1Oracle's Alleged "Efficiencies" Defense Is Not Based on an Actual Efficiency, i.e., "Something That Expands Output."

    9.2Oracle's Efficiencies Claims Are Not Verifiable

    9.3Oracle Used the Wrong Cost Baseline In Computing the Alleged Efficiencies

    9.4Oracle Has Not Shown That the Alleged Cost Savings Efficiencies Are Independent of the Substantial Reductions in Output Projected in the July 2003 Model

    9.5No Third Party Verified Oracle's Efficiencies Claims

    9.6Oracle's Efficiencies Claims Do Not Specifically Account for Substantial Merger Integration Costs or Potential Integration Risks

    9.7Oracle's Innovation Claims Are Not Merger Specific

    9.8Oracle's Various Efficiencies Claims Related to Innovation Improvements Are Speculative

  19. Oracle's "Stack" or "Infrastructure Layer" Arguments Do Not Alter the Antitrust Analysis of the Proposed Acquisition or Mitigate Its Anticompetitive Impact in the Relevant Markets
  20. 10.1Background

  21. Many of the Opinions Offered by Oracle's Expert Witnesses Are Unfounded, Contradicted by Other Oracle Witnesses or the Trial Record, or Otherwise Not Credible
  22. 11.1Dr. David Teece

    11.2Dale Kutnick

    11.3Dr. Jerry Hausman

    11.4Dr. Thomas Campbell

Appendix


1 Background.

    1.1 Oracle Corporation

      1.1.1 Oracle is a Delaware corporation with its principal place of business in Redwood City, California. Oracle's Answer, at Allegation 7; Joint Stipulations of Fact, at ¶ 1.

      1.1.2 Oracle provides organizations with database technology, enterprise software applications and related consulting services, in the United States and abroad. Oracle's Answer, at Allegation 7; Joint Stipulations of Fact, at ¶ 2; Ellison, 6/30/04, 4244:18-4245:1.

      1.1.3 In 2003, Oracle earned over $9 billion in revenues, including over $2 billion of revenues related to enterprise software applications. Oracle's Answer, at Allegation 7.

      1.1.4 Oracle sells human resource management software ("HRM") and financial management software ("FMS") solutions throughout the United States and around the world. Oracle's Answer, at Response to Allegation 26; Joint Stipulations of Fact, at ¶ 16; Ellison, 6/30/04, 4244:18-4245:1.

        1.1.4.1 The E-Business Suite is a fully integrated suite of more than 70 software modules for financial management, Internet procurement, business intelligence, supply chain management, manufacturing, project systems, human resources and sales and 16 service management. P2209, at xiv.

        1.1.4.2 As of December 2002, Oracle had over 5000 customers live on its E- Business Suite, Release Hi. P2208, at ORUT-EDOC-00244117.

        1.1.4.3 Oracle's Release Hi and its E-Business Suite are the same products. P3038, Godwin dep., 1/20/04, 52:22-53:09.

    1.2 Oracle Sells High Function HRM and FMS Software. Iansiti, 6/17/04, 2025:02-13; See 2.2. infra.

        1.2.1.1 High function software is capable of executing a wide array of business processes at a superior level of performance. Iansiti, 6/17/04, 2035:01-17; See 2.2. infra.

        1.2.1.2 High function software is scalable. See 2.2. infra.

        1.2.1.3 High function software is configurable. See 2.2. infra.

        1.2.1.4 High function software must be able to perform a variety of related transactions seamlessly. See 2.2. infra.

        1.2.1.5 High function software must be able to handle international aspects of an enterprise: multiple languages, multiple currencies, multiple legal regimes. See 2.2. infra.

        1.2.1.6 High function software must be able to handle the acquisition of new enterprises and the reorganization of existing ones. See 2.2. infra.

        1.2.1.7 High function software must be able to consolidate information across multiple divisions or organizations. See 2.2. infra.

    1.3 PeopleSoft, Inc.

      1.3.1 PeopleSoft is a Delaware corporation with its principal place of business in Pleasanton, California. Oracle's Answer, at Allegation 8; Joint Stipulations of Fact, at ¶ 7.

      1.3.2 PeopleSoft provides organizations with enterprise software applications and offers related consulting services in the United States and abroad. Oracle's Answer, at Allegation 8; Joint Stipulations of Fact, at ¶ 8.

      1.3.3 PeopleSoft earned over $2 billion in revenues in 2003, comprised entirely of enterprise software applications-related revenue. Oracle's Answer, at Allegation 8.

      1.3.4 PeopleSoft sells HRM and FMS solutions throughout the United States and around the world. Oracle's Answer, at Response to Allegation 26; Joint Stipulations of Fact, at ¶ 17.

      1.3.5 PeopleSoft sells application software that runs commercial and governmental enterprises, whether large, complex enterprises or those in the middle market. Bergquist, 6/08/04, 260:9-17.

      1.3.6 PeopleSoft sells four categories of application software: Human Capital Management (HCM), Financial Management (FMS), Supply ChainManagement (SCM), and Customer Relationship Management (CRM). Bergquist, 6/08/04,261:02-06.

      1.3.7 PeopleSoft was formed in the late 1980's to develop an HR product for large and complex customers. Bergquist, 6/08/04, 262:24-263:06.

        1.3.7.1 This focus on the needs of large and complex customers affected the way PeopleSoft designed its human resources product: it had to be highly functional, had to scale well, had to support multiple industries, and had to have a global orientation. Bergquist, 6/08/04, 263:13-23.

        1.3.7.2 PeopleSoft's Enterprise product is primarily sold to large and complex enterprises because it has the functionality to meet their needs. More specifically, it has the breadth and depth of functionality required by such firms, industry expertise, global capabilities and support, high levels of scalability and reliability, and the appropriate technology. Bergquist, 6/08/04, 274:01-15.

      1.3.8 PeopleSoft offers suites both in the sense of multiple products within a single pillar (a pillar being a single product line; PeopleSoft has four pillars: HRM, FMS, SCM, and CRM) and multiple products across different pillars because customers are looking for full functionality from a single vendor to minimize both the number of vendors they have to deal with and the cost of integrating multiple products together. For those customers who prefer to buy separate products, PeopleSoft will also sell them module by module. Bergquist, 6/08/04, 262:10-19; 265:10-266:18.

      1.3.9 Oracle regards PeopleSoft as a Tier One competitor for ERP software, including FMS and HRM enterprise software. P3009, Block dep., 1/05/04, 386:12-387:9.

        1.3.9.1 Tier One is another term for high function software. Elzinga, 6/18/04, 2180:22-2181:05.

      1.3.10 PeopleSoft sells high function HRM and FMS software. Iansiti, 6/17/04, 2025:02-13; See 2.2. infra.

        1.3.10.1 High function software is capable of executing a wide array of business processes at a superior level of performance. Iansiti, 6/17/04, 2035:01-17; See 2.2. infra.

        1.3.10.2 High function software is scalable. See 22. infra.

        1.3.10.3 High function software is configurable. See 2.2. infra.

        1.3.10.4 High function software is able to perform a variety of related transactions seamlessly. See 2.2. infra.

        1.3.10.5 High function software is able to handle international aspects of an enterprise: multiple languages, multiple currencies, multiple legal regimes. See 2.2. infra.

        1.3.10.6 High function software is able to handle the acquisition of new enterprises and the reorganization of existing ones. See 2.2. infra.

        1.3.10.7 High function software is able to consolidate information across multiple divisions or organizations. See 2.2. infra.

    1.4 Oracle's Unsolicited Offer for Peoplesoft

      1.4.1 Oracle's unsolicited $16 per share cash tender offer for PeopleSoft was announced on June 6, 2003. P3171, Ellison dep., 1/20/04, 53:7-54:10 (referencing P2039 (Oracle's Form 8-K including Oracle press release)).

      1.4.2 Pursuant to its public tender offer, Oracle plans to purchase PeopleSoft. Oracle's Answer, at Allegation 38; Joint Stipulations of Fact, at ¶ 9.

        1.4.2.1 Oracle CEO Lawrence Ellison decided to initiate a tender offer for PeopleSoft the morning he learned of PeopleSoft's acquisition of J.D. Edwards. Ellison, 6/30/04, 4276:12-16; see also P3327 at ORCL-EDOC-00030819.

        1.4.2.2 Oracle's decision to initiate a hostile take-over for PeopleSoft was triggered by a 6/02/03 e-mail from Ms. Catz to Mr. Ellison. Ellison, 6/30/04, 4278:08-20; P3327 at ORCL-EDOC-00030819 (stating "Now would be the time to launch on PSFT").

        1.4.2.3 Oracle had considered acquiring J.D. Edwards in April, 2003. P0035 at 26; P3033, Henley dep., 5/4/04, 88:18-89:05; 110:15-111:09.

      1.4.3 Oracle's reasons for launching the hostile tender offer included disrupting PeopleSoft's acquisition of J.D. Edwards and winning competitions against PeopleSoft.

        1.4.3.1 Ellison launched the tender offer the morning he heard about PeopleSoft's deal with J.D. Edwards. Ellison, 6/30/04, 4276:12-16; see also P3327 at ORCL-EDOC-00030819.

        1.4.3.2 In justifying the tender offer to Oracle's Board of Directors, Ellision and Catz solicited the help of Mr. Reece of Credit Suisse First Boston, who contributed to the presentation to the Board. Ellison, 6/30/04, 4279:4-16, see also P0092.

        1.4.3.3 Mr. Reece advised that the "[k]ey element is the one you focused on, time is on our side. Unless, and this is a big one, unless they change their bid for JD to an all cash offer or at least enough cash to accelerate and avoid a PST shareholder vote, time is our ally! Right now we are calling our strategy the Twist-in-the-Wind-Strategy. No need to change our bid! Standing firm will begin to create doubts in the minds of the market. Over time, we should see a decline in the price of PSFT in the market. Their quarterly results should, we certainly hope, dampen the markets enthusiam (sic). Of course, if they change their bid, AA and I have worked out a couple of strategy and ad scenarios where we completely BLAST our friends at PSFT." P0092 at CSFB 0028832.

      1.4.4 Oracle's Plans Post-Merger

        1.4.4.1 Oracle is not pursuing PeopleSoft in order to acquire PeopleSoft's technology. Ellison, 6/30/04, 4315:19-22; 4332:2-15.

        1.4.4.2 Oracle will not be actively selling PeopleSoft products to new customers but will provide some support for all PeopleSoft products. P3171, Ellison dep., 1/20/04, 54:11-56:02 (referencing P2039 (Oracle's Form 8-K including Oracle press release)); Ellison, 6/30/04, 4306:21-4307:8.

        1.4.4.3 Post-acquisition, Oracle will not have a separate sales force selling PeopleSoft products. Oracle's sales force is trained in selling Oracle's products, and those are the products they will continue to sell; Oracle's sales force will not be trained to sell the PeopleSoft products. P3171, Ellison dep., 1/20/04, 56:03-59:13; Catz, 6/25/04, 3540:12-3542:17.

        1.4.4.4 After acquiring PeopleSoft, Oracle will not develop future generations of PeopleSoft software. Catz, 6/25/04, 3540:12-25.

        1.4.4.5 Oracle has no plans to integrate the PeopleSoft and Oracle product lines. Ellison, 6/30/04, 4300:21-4301:11; see also P3270 at 5.

        1.4.4.6 PeopleSoft customers have no assurances that they will receive the same level of support and enhancements that they thought they were purchasing from PeopleSoft pre-acquisition. Ellison, 6/30/04, 4307:13-4308:11.

      1.4.5 PeopleSoft difficulties resulting from Oracle's tender offer

        1.4.5.1 Oracle believes that Oracle's hostile takeover bid for PeopleSoft will make customers reluctant to purchase its software products that will be discontinued post- merger. P3192, Block dep., 1/5/04, 463:19-464:03.

        1.4.5.2 Oracle has used the pendency of the PeopleSoft deal to hurt PeopleSoft's business and win accounts from PeopleSoft. Ellison, 6/30/04, 4286:23-4287:4; P2280 at ORCL-EDOC-0030398.

        1.4.5.3 After AIMCO heard about Oracle's proposed acquisition of PeopleSoft, it was apprehensive about making a purchase from PeopleSoft, because PeopleSoft might not be around in the future. Wesson, 6/14/04, 1145:07-13.

        1.4.5.4 AIMCO purchased PeopleSoft software only after receiving a guarantee to cover its costs in the event of a takeover of PeopleSoft. Wesson, 6/14/04, 1145:07-1146:22, at 1146:05-07.

    1.5 Jurisdictional Requirements

      1.5.1 Oracle sells its products throughout the United States and such sales in the United States, and in each of the Plaintiff States, represent a regular, continuous and substantial flow of interstate commerce, and have had a substantial effect upon interstate commerce as well as commerce with and in each of the Plaintiff States. Oracle's Answer, at Allegation 4; Joint Stipulations of Fact, at ¶¶ 3-5.

      1.5.2 Oracle transacts business and is found within the Northern District of California. Oracle's Answer, at Allegation 5; Joint Stipulations of Fact, at ¶ 6.

    1.6 Witnesses

        1.6.1.1 Appendix A contains information on some of the witnesses relied upon by the parties during the trial.

        1.6.1.2

2 Product and Industry Characteristics

    2.1 Product and Industry Definitions

      2.1.1 Function: A function is a business process that can be performed manually and/or can be automated in whole or in part by enterprise applications software. Examples of business functions include paying payroll or preparing a general ledger. Function is related to functionality, but is not used interchangeably. Joint Submission of Definitions, ¶

      2.1.2 Functionality: Functionality refers to the ability of enterprise applications software to automate particular business functions. Functionality can also be thought of as the capabilities of a software program. Joint Submission of Definitions, ¶ 3.

      2.1.3 Core financial management services ("FMS") software: The term "core financial management services ("FMS") software," as generally used in the software industry and by the parties in this case, refers to the software necessary to automate an organization's basic financial functions, including the following modules: General Ledger, Accounts Receivable, Accounts Payable, and Cash and Asset Management. Joint Submission of Definitions, ¶ 4.

      2.1.4 Core human resource management ("HRM") software: The term "core human resource management ("HRM") software," as generally used in the software industry and by the parties in this case, refers to the software necessary to automate an organization's basic human resource management functions, including the following modules: Payroll, Employee Tracking and Benefits Administration business processes. Joint Submission of Definitions, ¶ 5.

      2.1.5 Legacy system; Incumbent system: A legacy system refers to an installed software system that was never or is no longer available for purchase as a new product or platform. An incumbent system refers to whatever software system a customer currently has installed. Legacy systems are one subset of incumbent systems. Joint Submission of Definitions, ¶ 6.

      2.1.6 Enterprise resource planning ("ERP") software: Enterprise resource planning ("ERP") software refers, at a minimum, to the set of business applications software that is designed to automate an organization's back-office functions, such as payroll, general ledger, accounts payable and receivable, benefits management, and employee record keeping. It traditionally encompasses HRM and FMS. Some also include one or more other pillars in the definition of ERP, such as Supply Chain Management (SCM) or Customer Relationship Management (CRM). Joint Submission of Definitions, ¶ 7.

      2.1.7 Enterprise application software ("EAS"): In contrast to ERP, enterprise applications software ("EAS") refers to the broader universe of software applications that are used by an enterprise to automate its overall business processes. In addition to the Human Resource Management (HRM) and Financial Management Services (FMS) pillars that traditionally comprise ERP, EAS also encompasses the Supply Chain Management (SCM), Customer Relationship Management (CRM), Product Life Cycle Management, Student Information Systems (SIS), and Business Intelligence (BI) pillars. Joint Submission of Definitions, ¶ 8.

      2.1.8 Integration: In the context of this case, the term "integration" refers to the ability of a software application in one module to communicate with, work with and update software and data in another module. Most companies, particularly large ones, have a large number of software products that need to be integrated so they can communicate with each other. Joint Submission of Definitions, ¶9.

      2.1.9 Analytics: Analytics is software that enables a company to use, analyze, report and aggregate data collected through one or more other business applications. Joint Submission of Definitions, ¶ 10.

      2.1.10 Module: A module is software designed to automate one or more functions. Joint Submission of Definitions, ¶ 11.

      2.1.11 "High-function software" is a term adopted by Plaintiffs in order to describe more precisely the features of a category of software products that industry participants call "enterprise" software, "up-market" software, or "Tier One" software." See, e.g., Bergquist, 6/8/04, 274:24-275:7; Wilmington, 6/16/04, 1771:5-1772:1; Iansiti, 6/18/04, 2180:22-2181:5.

        2.1.11.1.1 In today's global economy, the ability to reduce the costs inherent in running an organization is vital to an organization's success. Most organizations (including corporations, federal, state, and local government agencies, and non-profit organizations) therefore automate their financial management and human resource functions in search of operational efficiencies. Oracle's Answer, at Response to Allegation 9; Joint Stipulations of Fact, at ¶ 11.

        2.1.11.1.2 There are a variety of enterprise software products that organizations use to automate different types of business functions. Among others, enterprise software can be used to (1) manage employees through HRM software and (2) maintain financial records through FMS software. Oracle's Answer, at Allegation 10.

        2.1.11.1.3 Less sophisticated software require organizations to perform more manual processes that are labor intensive, subject to errors and can require significant additional server space. [REDACTED TEXT]

        2.1.11.1.4 Although a small to medium sized company might face complicated tasks, because the operations are less diverse and dispersed it is easier to use traditional means - phone, fax, meetings - rather than automation to manage the organization. Iansiti, 6/17/04, 2031:2-16.

        2.1.11.1.5 As an organization grows, it's increasingly difficult to use the traditional means to understand even the most basic functioning of the organization. Iansiti, 6/17/04, 2031:2-16

        2.1.11.1.6 There are two different types of tasks that need to be accomplished in order to manage a large enterprise: one is to process transactions and the other is understanding critical information necessary in order to implement basic managerial tasks. Iansiti, 6/17/04, 2028:22-2030:12.

        2.1.11.1.7 In order for a large complex enterprise to process the transactions and enable the information to be used to manage there are certain high-function requirements that software must meet. Iansiti, 6/17/04, 2033:16-2034:1.

        2.1.11.1.8 A sophisticated product can provide enhanced functionality and an ability to streamline all Human Resources, benefits and payroll business processes. [REDACTED TEXT]

        2.1.11.1.9 High function HR software allows companies to run their businesses more efficiently and effectively. High function HR software allowed [RED.] to automate a number of manual processes and eliminate costly customization. [RED. TEXT]

        2.1.11.1.10 High function software needs to be able to capture the organization and all of its complexity. Where necessary it needs to operate in multiple languages and currencies, across multiple jurisdictions so as to correctly reflect the actual structure and organization of the firm. Iansiti, 6/17/04, 2035:1-17

    2.2 "High-Function Software" Is a Software Product That Is Capable of Executing a Wide Array of Business Processes at a Superior Level of Performance. See, e.g., Iansiti, 6/17/04, 2035:1-17. Thus,"high-function" software must have at least the following advanced performance capabilities, not all of which each user may demand:

      2.2.1 Scalability. The Software must Be Scalable, Such That it Can Track Thousands of Transactions and Support Thousands of Concurrent Users. See, E.g., Burgum, 6/23/04, 3011:19-3013:3. Bergquist, 6/8/04, 289:22-290:7; Johnsen, 6/16/04, 1755:11-24.

        2.2.1.1 High-function software generally permits the customer to establish unlimited distinct units and employees within each unit that can share data seamlessly and can be moved -- or reconfigured -- easily (such as with a click and a drag) without causing the loss of historical data. In contrast, the Lawson products permit only 5 levels of organization in FMS and only 3 in HRM and do not enable extensive data sharing across entities. Iansiti, 6/17/04, 2046:19-2047:5.

        2.2.1.2 High-function software has the capability (i) to accept thousands of invoices in different currencies and to remit payment in a different currency while automatically recalculating the exchange rates on a real-time basis; (ii) High-function software has the capability of generating reports from hundreds of separate business units while permitting the user to quickly "drill down" to the underlying source data; (iii) the ability to model reorganizations of the business and to generate forecasts; and (iv) the ability to generate reports on costs associated with a product or component that may be drawn from dozens or even hundreds of distinct entities within the larger organization. Iansiti, 6/17/04, 2039:2-2040:7.

        2.2.1.3 Enterprise software must meet more demanding scalability and reliability requirements than those required of mid-market software. Bergquist, 6/8/04, 289:8-290:4.

        2.2.1.4 Changes in hardware and hardware architecture have not resolved the problems of scalability and the software must continue to address it. Bergquist, 6/8/04, 290:8-14.

        2.2.1.5 The software Verizon uses for its financial management and human resources needs must be robust (relatively defect free) and scalable (able to accommodate a large number of users and a large number of transactions). Bradley, 6/9/04, 598:4-600:8. Scalability allows the software to work whether there are hundreds or thousands of people interacting with the system. For Verizon that means processing several billion transactions per month. Bradley, 6/9/04, 599:16-600:8.

        2.2.1.6 SAP's high function software can handle 47,000 concurrent users and give a response time of less than two seconds. P3036, Knowles dep., 12/3/02, 103:20-104:2.

      2.2.2 Configurability. High Function Software must Be Highly Configurable (e.g., it must Enable Unlimited Levels of Organization), Such That the User Can Mold the Software to Meet its Business Needs Without Having to Undertake Expensive and Inefficient Customization of the Software. See Johnsen, 6/16/04, 1753:9-1754:8; Iansiti, 6/17/04, 2046:19-2048:5, 2036:7-20; Bergquist, 6/8/04, 284:17-285:6, 287:9-17.

        2.2.2.1 Configuration is built into the software and allows for a user to select between various options without rewriting code or merely accepting a standardized version of software that does not fit its specific business practices. Iansiti, 6/17/04, 2121:16-2123:4.

        2.2.2.2 Configuration or tailoring is defining values in data that indicate how software should behave for a specific business practice. P3038, Godwin dep., 1/20/04, 131:2-132:14.

        2.2.2.3 Software package configurations are a series of decisions end-users and management make upon installing software that allows the software to fit more closely with an enterprise's needs. P2896 at ORLITF0091413.

        2.2.2.4 Customization is when a customers take the code that Oracle or another vendor has supplied and changes the code to have it work another way. When Oracle updates its software, the customer is required to re-write the code they have changed to update the software. P3038, Godwin dep., 1/20/04, 131:2-132:14.

        2.2.2.5 In contrast to configuration, customization, while sometimes necessary, is disfavored as expensive to do and difficult to maintain. For that reason, up-market customers look for a high degree of configurability in the software - where its options are pre-built. Bergquist, 6/8/04, 285:23-287:17.

        2.2.2.6 "Well, the simplest way is, a configuration is, the ability to make a change via configuration is a good thing, and customization is a bad thing. Johnsen, 6/16/04, 1753:9-1754:8.

        2.2.2.7 Customization or modifications involves changing the software code and are not supported by the vendor. P2896 at ORLITF0091414.

        2.2.2.8





        [REDACTED TEXT]


        2.2.2.9 Software that has configuration flexibility can to be adapted to specific industry needs or different needs within divisions or parts of an enterprise. Configurable software will allow for software to reflect the differences across industry verticals, within industry verticals and even between divisions of an enterprise. Iansiti, 6/17/04, 2035:18-2036:6.

        2.2.2.10 By making their applications as configurable as possible customers do not have to go in and invasively modify the code that Oracle has shipped to make the applications work for their business. Oracle puts a lot of effort into configuring capabilities into its software so that a customer may set up the software to work the way they want it to for their business without incurring a high level of expense. P3038, Godwin dep., 1/20/04, 128:13-129:18.

        2.2.2.11 Oracle writes it applications to be highly configurable so that a customer will not have to actually write code to make the program work with their business. P3038, Godwin dep., 1/20/04, 129:19-130:20.

        2.2.2.12 The flexibility of the application is also an important distinction between up-market and mid-market vendors. Bergquist, 6/8/04, 283:8-284:2.

        2.2.2.13 Unlike up-market firms, mid-market entities are less able to afford or operate highly flexible software products and are willing to change their business processes to conform to the software. Bergquist, 6/8/04, 284:3-8.

        2.2.2.14 Software flexibility in this sense is the result of the vendor building in "configurability" options. Bergquist, 6/8/04, 284:9-285:6.

        2.2.2.15 The flexibility built into high function enterprise software allows the customer to select options at the initial implementation stage and, later, to accept upgrades without customization and the substantial attendant costs. Wolfe, 6/16/04, 1565:21-1568:12; 1600:02-14.

        2.2.2.16 Oracle's enterprise application software is highly configurable, which allows it to meet the diverse needs of its customers. Its inherent flexibility makes it adaptable to a variety of business practices. P3033, Henley dep., 5/4/04, 58:15-59:6.

        2.2.2.17 Organizations handle transactions differently depending on the industry and business model, these differences should be reflected in the software. P3037, Knowles dep., 5/3/04, 73:18-21.

        2.2.2.18 Oracle build into its ERP software more features and more configuration options to allow larger customers to fit the software to their business processes without having to engage in as much customization. P3171, Ellison dep., 1/20/04, 271:14- 274:12; P3038, Godwin dep., 1/20/04, 131:2-132:14.

        2.2.2.19 At the broad level industries are different, but even within industries companies are very different. "And the software has to be able to be configured in such a way to solve what the company thinks is the more important elements to run its business, to try to create some element of competitive advantage from the other, their own competitors, within their given industry." P3037, Knowles dep., 5/3/04, 73:18- 75:8.

      2.2.3 High Function Software must Be Able to Perform a Variety of Related Transactions Seamlessly (e.g., Provide a "Very Tight Integration Between" Business Processes Without Having to Write Customized Code) and with a High Degree of Ease and Sophistication. Iansiti, 6/17/04, 2039:2-2040:7.

        2.2.3.1 Integration: high-function software is sold in integrated suites, which offers many benefits including lower TCO. An integrated suite has a lower TCO because the customer does not have to pay to revise the interface between pillars when a new version or upgrade is released. P3050, Kender dep., 5/11/04, 79:11-81:17.

        2.2.3.2 Nextel wanted an integrated solution for HR and FMS; Nextel was concerned that if it acquired two separate systems and attempted to integrate them, Nextel would never reach optimal efficiency. These concerns would also apply to future upgrades. Cichanowicz, 6/14/04, 1069:23-1071:12.

        2.2.3.3 Oracle applications allow financial information to be rolled up from different parts of the organization using different languages and then would allow users to view the results of the roll-up using descriptions in their own language. P3038, Godwin dep., 1/20/04, 180:11-181:17.

        2.2.3.4 The ability to capture data, roll-up information, drill down and accomplish it in near real time is important to allow an organization to have critical data be as accurate a possible when management decisions are made. Iansiti, 6/17/04, 2040:19-2041:3.

        2.2.3.5 Basic software packages allow the user to track employee names, addresses, salaries and job levels. In contrast, the high-function human resource capital management software sold by PeopleSoft, Oracle and SAP goes beyond the basics and allows an organization to track employees by skills, job classes, job codes, union affiliations, and any number of other criteria. Glover, 6/15/04, 1484:8-1485:3.

        2.2.3.6 Verizon needs software that can accommodate a workforce that has significant variation in its contractual relationship with Verizon by virtue of differences in the union contracts with Verizon. Bradley, 6/9/04, 582:24-583:15. Verizon's software needs to be able to accommodate Verizon's need to account for differences in financial reporting requirements for regulated affiliates and its unregulated affiliates. Bradley, 6/9/04, 581:5-23. Verizon's FMS must be able to quickly adopt to Verizon's frequent corporate reorganizations. Bradley, 6/9/04, 581:24-582:19.

        2.2.3.7 Managing a tangle of connections between point solutions is extremely costly and complex. Any time you update a point solution, you must go back and review all integration points and potentially update the integration software. P2208, at ORLIT-EDOC-00244083.

      2.2.4 International Operations. High Function Software must Have the Capability to Handle International Aspects of a Business, Such as Multiple Currencies, Multiple Languages, and Multiple Legal Regimes (e.g., "Enhanced Multi-currency Transaction Support, Invoicing in Any Currency, [And] Cross-charging and Inter-company Accounting" and the Support of "30 Languages, All Currencies, and Many Regulatory Requirements in a Single Database" with All Languages "Installable in the Same Unicode Instance"). See P2208 at ORCL-EDOC-00244092- 95; ORCL-EDOC-00244113; see also Bergquist, 6/8/04, 290:15-291:6.

        2.2.4.1 In order to have broad multi-language support, a vendor must have a Unicode character set, which allows for applications to be written in both Western and Eastern characters. P3038, Godwin dep., 1/20/04, 179:3-180:5.

        2.2.4.2 Oracle's multiple language support is not separate installations of its applications in separate languages, but rather allows users that are using different languages to access and modify the same corporate data. P3038, Godwin dep., 1/20/04, 182:9-12.

        212.2.4.3 "Oracle supports 30 languages, all currencies and many regulatory requirements in a single database. All 30 are installable in the same Unicode instance. Trading partner receive business documents in the language of their choice and user view and enter date, number and currencies in their choice of format. There is no separate U.S., Japanese or French software version of the E-Business Suite. Support for local business practices, address formats and reports is all part of the standard product." P2208 at ORLIT-EDOC-00244113.

        2.2.4.4 Multiple language support must be developed by the vendor so that international users can interact with the software in their native language. P3036, Knowles dep., 12/3/02, 84:2-85:15.

        2.2.4.5 Oracle, Peoplesoft and SAP are able to deal with currency fluctuations on a centralized basis that allows for minute-by-minute, real-time changes to be reflected in the software. Iansiti, 6/17/04, 2057:7-2058:3.

        2.2.4.6 In Oracle 11i software applications it is possible to use multiple reporting currencies, which is an Oracle General Ledger feature that allows you to report your functional currency in one or more foreign currencies. P2208 at ORLIT-EDOC- 00244092; P2209 at Appendix, p. 2.

        2.2.4.7 High-function financial management software supports multiple currencies so that the organization can report in the local currency but it also rolls up financial reports to the corporate headquarters and automatically converts it to the local currency of the headquarter country. P3036, Knowles dep., 12/3/03, 82:25-84:1.

        2.2.4.8 Enterprise software must have global capabilities. Global in this sense means the ability to handle multiple countries, multiple currencies, and multiple languages and to do so concurrently. Bergquist, 6/8/04, 290:15-291:6.

        2.2.4.9 Mid-market vendors lack the geographic scope required by many up-market firms. Bergquist, 6/8/04, 291:7-24.

        2.2.4.10 Software must account for the differences in reporting certain information from country to country. P3036, Knowles dep., 12/3/02, 84:7-85:15.

      2.2.5 Acquisitions and Reorganizations. Software must Be Flexible and Able to Accommodate Rapid Growth, Acquisitions and Reorganizations.

        2.2.5.1 High-function software captures data and can use it to model the organization for purposes of forecasting and reorganization. Iansiti, 6/17/04, 2039:2-2040:18.

        2.2.5.2 PeopleSoft allows for reorganizations to be modeled through the use of a drag and drop function that makes it very easy. Software that cannot do so has enormous limitations. Iansiti, 6/17/04, 2051:18-2052:4.

        2.2.5.3 Oracle Application 11i allows customers to model or do reorganizations without losing historical organizations and transactions. "At some point, your company may need to make organization changes. Companies usually make organization changes to better adapt themselves to the new business needs and challenges of the market. One type of organization change you can make is to discontinue one organization...If you create a new organization, you can still access your historical transactions by logging to the responsibility tied to the disabled organization." P2209 at 2-23.

        2.2.5.4 Verizon, for example, has 120 unregulated entities, which for business reasons they acquire and sell fairly often. This requires that Verizon be able to separate out the numbers for the companies that they are selling. Bradley, 6/9/04, 581:24-582:19.

      2.2.6 Multiple-organizations. High Function Software Allows for Information to Be Consolidated Across the Organization to Generate Reports, but Also Allows for a Manager to Drill Back down to Understand Where the Data in a Specific Entry Came From. Iansiti, 6/17/04, 2039:2-2040:7.

        2.2.6.1 High function software can be set up so that it reflects an actual unit of business, rather than a textbook example, and link the data so that it is useful. Iansiti, 6/17/04, 2048:9-25

        2.2.6.2 It is important for an organization to be able to run a single instance of the product, that will allow for users in different locations, different jurisdictions, perhaps different countries speaking different languages, to have access to the same information at the same time. Iansiti, 6/17/04, 2042:1-12.

        2.2.6.3 High function software allows for interorganizational journal transactions which allow for different organizations to be linked to transactions, so that they can be correctly captured in the software accounting can be done correctly, and the books can be closed. Iansiti, 6/17/04, 2049:16-2050:5.

        2.2.6.4 Oracle states about its single global instance strategy, "[c]ompanies can manage their worldwide operations using an single standard installation providing consolidated data globally while ensuring fiscal and statutory compliance around the world. Oracle offers a truly global product that supports multiple languages, multiple currencies, unique global business practices, and local statutory and regulatory requirements." P2208, at ORLIT-EDOC-00244083.

        2.2.6.5 Oracle's software applications allow customers to separately track and report multiple organizations within a single application. P3038, Godwin dep., 1/20/04, 202:19-203:2.

        2.2.6.6 Large, complex organizations typically have multiple legal entities, ranging anywhere from maybe 200 to 2000. Thomas, 6/9/04, 475:3-14.

        2.2.6.7 In Oracle, Peoplesoft and SAP it is possible to model unlimited numbers of levels of an organization and match the software to them so that it is useful, accurate and visible to management. Iansiti, 6/17/04, 2047:17-2048:5; P3038, Godwin dep., 1/20/04, 215:18-216-5.

        2.2.6.8 Oracle's software applications allow for a customer to model multiple organizations within a single installation of its software, which allows for customers to account for different subsidiaries or different operating units or to subdivide its business along other lines. P3038, Godwin dep., 1/20/04, 206:10-207:17.

    2.3 Characteristics of Vendors of High-Function Software Are Important to Purchasers of High-Function Software.

      2.3.1 Reference Customers: References from Similarly Situated Customers Are Used by Enterprise Software Vendors in Marketing FMS and HRM Solutions. Oracle's Answer, at Response to Allegation 17; Joint Stipulations of Fact, at ¶ 15.

        2.3.1.1 Enterprise software vendors compete to secure references from large enterprises. Oracle's Answer, at Response to Allegation 17.

        2.3.1.2 In addition to high functional capabilities, vendors of high function software must demonstrate credibility and business capabilities. Financial stability is critical. Iansiti, 6/17/04, 2037:1-2038:6.

        2.3.1.3 In addition to the capabilities of the products themselves, vendors who wish to sell to the up-market must establish not only their fiscal stability but also a track record of successful implementations. Bergquist, 6/8/04, 296:7-297:15; 297:20-298:10.

        2.3.1.4 Large, complex companies like DaimlerChrysler only purchase HR software from vendors that can provide a reference from a company in the same industry. Gorriz, 6/15/04, 1375:9-15.

        2.3.1.5 Reference customers and an extensive network of contacts with similarly situated companies - those with rich complexity and regulatory issues similar to those faced by Verizon - are important to Verizon before it makes a selection. Bradley, 6/9/04, 600:9-602:3.

        2.3.1.6 Customers look at functionality, credibility, reputation, and ability to sustain a long-term relationship. P3171, Ellison dep., 1/20/04, 45:2-46:8.

        2.3.1.7 Vendors of high-function software to large, complex state and local customers must be able to demonstrate large, credible references that look a lot like the customer. P3171, Ellison dep., 1/20/04, 286:14-287:8.

      2.3.2 Research and Development by a Vendor Is Critical to Purchasers of High- Function Software.

        2.3.2.1 Vendors must be able to demonstrate a level of research and development investment and spending that will ensure that the software is constantly upgraded and maintained to make certain that software improves as technology improves. Iansiti, 16 6/17/04, 2037:1-2038:6.

        2.3.2.2 Oracle has 6000 R&D personnel worldwide working on enhancements to the 11i E business suite applications. P3038, Godwin dep., 1/20/04, 88:6-92:8.

        2.3.2.3 Since first releasing the 1li application, the number of R&D personnel making enhancements to the product has grown, because Oracle has been investing more money into its applications to grow its applications business. P3038, Godwin dep., 1/20/04, 89:18-90:6.

        2.3.2.4 With the release of 11i, Oracle released new human resources and financials modules. P3038, Godwin dep., 1/20/04, 106:4-108:4.

        2.3.2.5 Since Oracle 11i applications were released in 2000, more person years of new development have occurred than had been devoted to the development of all of Oracle applications to that point. P3038, Godwin dep., 1/20/04, 106:4-1-108:4.

        2.3.2.6 Vendors engage in R&D to serve their largest customers. Hausman, 6/29/04, 4053:16-4054:11; 4058:7-4059:4.

        2.3.2.7 Many customers consider upgrades to the software important. Campbell, 6/22/04, 2749:25-2752:2.

        2.3.2.8 Vendors of high function software must invest sufficient development resources to innovate the software and to meet the evolving needs and demands of customers. Knowles, 6/23/04, 2870:24-2871:12.

        2.3.2.9 Product integrity and product development is related in part to research and development spending and the stability of that investment over the long term. P3037, Knowles dep., 5/3/04, 86:4-87:16.

        2.3.2.10 Reductions in research and development spending signal the marketplace that the software vendor is no longer improving the product. P3037, Knowles dep., 5/3/04, 86:4-87:16.

        2.3.2.11 Software evolves over time, it becomes better, it can do more, and it can solve more issues. P3037, Knowles dep., 5/3/04, 142:18-144:4.

        2.3.2.12 Customers do not just want better functionality, they want it from a vendor that will continue to improve the product over a five-to-ten year period. P3171, Ellison dep., 1/20/04, 48:8-49:8.

        2.3.2.13 Web-enablement is a critical feature for high function software because it assists in accurately entering data, in deploying an application across many different users, in implementing that data and in having the data be visible to all of those who need to use it. Iansiti, 6/17/04, 2043:2-13.

        2.3.2.14 Oracle went to a full web-enabled version of its products when it released the 11i product in May 2000. P3038, Godwin dep., 1/20/04, 55:15-21; 64:9-65:12.

        2.3.2.15 Laurette Bradley of Verizon believes that Oracle, PeopleSoft and SAP are "right at the forefront of embracing web services. So that if this is a new foundational technology, we don't have to worry that they'll lag 10 years behind. We know that if it counts, they will develop it, they'll integrate it, they'll a super job of it." Bradley, 6/9/04, 598:9-599:5.

      2.3.3 Dedicated Sales Forces Are Necessary to Assist Customers in Selecting a Product That Will Meet Their Complex Functional Requirements.

        2.3.3.1 Vendors of high-function software need a dedicated sales force that understands the enterprise's needs and can work to match the software as much as possible to the needs of the enterprise. Iansiti, 6/17/04, 2037:1-2038:6.

        2.3.3.2 A dedicated sales force is one of the attributes that distinguishes most mid- market vendors, whose distribution model is to have sales made through partners and/or independent distributors, from a dedicated sales force. Iansiti, 6/17/04, 2037:1-2038:17.

        2.3.3.3 Peoplesoft, Oracle and SAP have very large, targeted sales forces that are specialized by industry verticals, which sell directly to large customers. Iansiti, 6/17/04, 2054:22-2055:7.

        2.3.3.4 High-function software vendors employ large and sophisticated sales forces. The sales force must explain to clients the complexity of the products and how those products will meet an organization's particular needs. Knowles, 6/23/04, 2903:5-2903:9.

      2.3.4 Strong Support Organizations Are Needed to Provide Technical Expertise to Buyers of High-Function Software.

        2.3.4.1 Enterprises require that vendors have strong support organizations. Inansiti, 6/17/04, 2037:1-2038:6, at 2037:21.

        2.3.4.2 The ability to provide global support for the software is an important criterium for international organizations. P3155 at 3.

      2.3.5 Financial Stability of the Related Vendor Is Critical to Customers Making the Long Term Investment in High-function Software.

        2.3.5.1 Purchasers of high-function software are typically concerned that the vendors they use be financially robust, "so they won't disappear after next quarter's results." Bradley, 6/9/04, 598:9-599:5.

        2.3.5.2 Large, complex companies like DaimlerChrysler only purchase HR software from vendors that have a solid financial background. Gorriz, 6/15/04, 1375:2-8.

    2.4 the Vendors' High Function Software Is Differentiated, i.e., Not All Alike, and Different Vendors Have Different Strengths and Weaknesses.

      2.4.1 Oracle, SAP, and Peoplesoft Perform Many of the Same Functions, but They Have Different Heritages Based on the Historic Strengths of Their Respective Products. There Are Architectural Differences Between the Products and the Products Operate Differently. Keating, 6/10/04, 897:23-898:7.

      2.4.2 Oracle, SAP, and Peoplesoft Have All Had Different Strengths Historically; Oracle and Sap Had Taken the Major Market Share in Those Industries in Which Their Respective Strengths Were Important to the Industry. Oracle Had Very Rich Financial Functionality and Won a Majority of Sales in That Area. SAP Had Very Rich Supply Chain and Manufacturing Capability and Had Won a Majority of the Market Share in Industries in Which Those Characteristics Were Important. Peoplesoft Had the Best Human Resources Functionality and Had the Dominant Market Share in the Market for Human Resource Software. Keating, 6/10/04, 898:8-899:3.

      2.4.3 Oracle and PeopleSoft have historically sold their products in modules that could be separately licensed and implemented, while SAP, at least until more recently, was all-encompassing and not licensed in separate modules. Keating, 6/10/04, 901:6-19. The customer had to buy the entire SAP suite. Keating, 6/10/04, 946:18-20. The modular architecture gave Oracle and Peoplesoft advantages in industries, such as telecommunications, in which the customers had other key systems that they needed to interface with the Oracle and PeopleSoft products. SAP took the lead in industries in the all-in-one software solution best matched the customer's requirements. Keating, 6/10/04, 901:20-902:15.

      2.4.4 Oracle took a major market share in the telecommunications industry because many telecommunications companies have billing systems that must be interfaced with a strong general ledger. Oracle also took a major share of the financial services industry because those customers had other financial systems that had to interface with the general ledger and other core financials. In contrast, SAP had a dominant position in industries that required sophisticated supply chain functionality, including aerospace and oil and gas. Keating, 6/10/04, 899:9-900:19. Some of the financial systems that financial services customers need to integrate with the general ledger include brokerage and lending systems. Keating, 6/10/04, 947:10-21.

      2.4.5 In comparison to Oracle and PeopleSoft, SAP was tightly integrated and had closed interfaces. Keating, 6/10/04, 947:4-9. because of its all-in-one architecture, SAP historically was more expensive to implement than Oracle and PeopleSoft. Now that SAP has begun to offer its product in modules, its implementation costs are getting more in line with Oracle and PeopleSoft's implementation costs. Keating, 6/10/04, 903:6-15. However, the gap has not been closed completely between SAP and Oracle and PeopleSoft in terms of modularity and openness. Keating, 6/10/04, 993:16-994:2.

      2.4.6 Keating testified that the differences that continue to exist between SAP and Oracle and PeopleSoft continue to provide a competitive advantage for Oracle and PeopleSoft for some customers. Keating, 6/10/04, 994:3-6.

      2.4.7 SAP has unsuccessfully attempted for years to sell its software to the North American banking industry. Keating, 6/10/04, 993:3-10. SAP's recent announcement about its product modification plans is SAP's most recent attempt to make its software attractive to North American financial services customers. Keating, 6/10/04, 992:23-993:7.

      2.4.8 Keating testified that SAP has been more successful at selling its software to financial institutions in Europe than in North American, but the banks in the U.S. must satisfy additional requirements that European banks do not have. In addition, the U.S. market for Tier One core ERP software in financial services is much bigger than the European market. Keating, 6/10/04, 996:20-997:15.

        2.4.8.1 CH2M Hill rejected SAP as an option because it had a reputation of being costly and complex, and it lacked reference clients in the engineering & construction industry. Bullock, 6/7/2004, 209:13-209:18.

        2.4.8.2 In certain areas like manufacturing, supply chain, and financial management, Oracle's product is more feature-rich than PeopleSoft's. P3171, Ellison dep., 1/20/04, 106:21-109:2.

    2.5 After a Review of 148 Vendors, Professor Iansiti Concluded That from the Standpoint of Functionality, Only Oracle, PeopleSoft and SAP Are Able to Satisfy the High Function Requirements of Enterprises. Iansiti, 6/17/04, 2025:2-206:6.

      2.5.1 Oracle, PeopleSoft and SAP provide a different class of software, different products, aimed at solving different requirements than other software providers. Iansiti, 6/17/04, 2085:3-19.

      2.5.2 The State of Ohio evaluated five ERP vendors against requirements for functionality, scalability, technical fit, R&D spending, and public sector spending. The State of Ohio short-listed Oracle, PeopleSoft, and SAP as most closely fitting its needs. P1944, at ORLITF 0149246.

      2.5.3 BearingPoint's Global Enterprise Solutions Group clients require software that allows them to maintain multiple charts of account so that they can track revenue by all their different products and services and combinations thereof. No software vendor provides multiple chars of account functionality that is comparable to Oracle, SAP or PeopleSoft. Keating, 6/10/04, 871:3-20.

      2.5.4 The largest commercial customers and public services clients have software requirements that has predominantly caused them to choose Oracle, PeopleSoft, and SAP. Keating, 6/10/04, 867:5-14.

    2.6 Mid-Market Software

      2.6.1 Mid-market software customers are often identified by proxies of size such as company revenue. Knowles, 6/23/04, 2814:10-21; Wilmington, 6/16/04, 1765:8-1766:8.

      2.6.2 Firms such as Oracle, PeopleSoft, and SAP have products and sales forces directed to serving mid-market customers. Knowles, 6/23/04, 2814:10-21; 2905:3-2906:21; Wilmington, 6/16/04, 1765:8-1766:8; 1767:5-1768:2; P3033, Henley Dep, 5/4/04, 53:22-55:8; 115:1-116:11.

      2.6.3 Customers of any size may require high function software depending on the characteristics of that customer. A smaller company may have complex processes that require high function software. Knowles, 6/23/04, 2908:20-2908:25.

      2.6.4 JD Edwards recognized the difference between high function or upmarket and mid-market customers. It identified mid-market as opposed to upmarket customers not by revenues or employee count but by various "buyer characteristics." Thus, for example, mid-market customers have a keen focus on costs and TCO; their IT organizations are small and unsophisticated; and their personnel tend to be generalists. Allen, 6/10/04, 751:25-752:16.

      2.6.5 JDE's software was originally designed for the mid-market. When it was able to sell to up-market firms it was almost always to divisions and subsidiaries of large firms, such as Johnson & Johnson, Walmart and Dole Foods. Allen, 6/10/04, 767:7-768:8.

      2.6.6 Mid-market buyers are looking for software that is highly integrated, possess features and functions appropriate to their specific industry, and, most of all, are easily configured and rapidly implemented. Allen, 6/10/04, 753:4-20.

      2.6.7 Mid-market firms are also interested in tight integration. They do not want to incur the burden of integrating products from different vendors into a suite, to customize the interfaces between various point solutions. Such interfaces are very costly to create and burdensome to maintain. Allen, 6/10/04, 755:2-16.

      2.6.8 What distinguishes up-market from mid-market is not employee count or revenues, but the complexity. Thus a large firm with a distributed structure - where division and subsidiaries make their own decisions on IT purchases - might have the buying characteristics of a mid-market firm. Allen, 6/10/04, 760:3-761:3.

      2.6.9 The differences between mid-tier and high function software are built right into the code. Allen, 6/10/04, 770:3-16.

      2.6.10 PeopleSoft's Enterprise product offers its up-market customers broader and more extensive configuration options than those offered to mid-market customers because the former have broader and more intensive needs for configuration. Bergquist, 6/8/04, 288:15-289:3.

      2.6.11 A customer with 1000 seats (users) will configure its software differently than a customer with 10,000 users. Hausman, 6/29/04, 4060:20-4061:16.

      2.6.12 The mid-market customers interest in easy configurability reflects its concern with cost. Easy configurability does not mean that the customer wants no configurability but that it does not want to make numerous complex decisions when the software is implemented. They want some choices but also a quick and less costly implementation. The more numerous the choices the more complex the software and the more costly the implementation. Allen, 6/10/04, 754:11-22.

      2.6.13 Firms will not change their structure from a centralized to a distributed structure, in order to be able to buy software from a mid-tier vendor. Allen, 6/10/04, 769:2-18.

      2.6.14 Although companies send out the same disks and the same code to all customers, smaller customers will not "turn on" the switch for the greater functionality, and therefore, large customers are getting a different product than smaller customers. Hausman, 6/29/04, 4057:3-4060:19.

      2.6.15 Large customers need a better product that can handle more transactions. Hausman, 6/29/04, 4053:16-4054:11.

      2.6.16 Even though the disk that goes to customers is the same, even though the product is "physically the same, but they [the vendors] know that it's going to be used in a different way, and it's like buying a different product. Hausman, 6/29/04, 4061:4-16.

    2.7 Sales Process for High-Function Software: the Sales Process Is Lengthy and Involves Extensive Sharing of Information in Order to Identify a Product That Can Meet the Customer's Functional Requirements.

      2.7.1 A sale is much more complex than just showing a customer a product's features and functionality. Customers "look at the total cost of ownership, which is the license, the business benefit provided by the application, the cost of ownership related to the technology, and the relationship with the company that they will have into the future." Wilmington, 6/17/04, 1953:8-1954:1.

        2.7.1.1 Organizations will identify a need for a new software system. A customer may be struggling with a business process. P3036, Knowles dep., 12/3/03, 64:9-65:8.

        2.7.1.2 Most customers will engage some third party consultants to aide them in their selection process. Bullock, 6/7/2004, 203:14-205:1; P4075R, at NEXTEL-000030-NEXTEL-000034; Cichanowicz, 6/14/04, 1060:10-18; P3062, Elliott dep., 5/20/04, 58:24-59:16.

        2.7.1.3 Prior to evaluating vendors, customers may form a selection team composed of technical and functional specialists who will identify all the functions they want in a software package. Bullock, 6/7/2004 202:24-203:13.

        2.7.1.4 Customers may form project teams composed of as many as 50 representatives to evaluate software alternatives. P3155, at 4.

        2.7.1.5 The sales process provides the information that permits the vendor to distinguish between firms that need high function software and those which don't. Indeed, it is the sales person's responsibility to "qualify" a prospect, to not spend time or effort on an up-market account that JDE could not satisfy. Allen, 6/10/04, 761:4-763:15; 764:6-765:15; 766:4-24.

        2.7.1.6 Customers often develop a list of candidate vendors before consulting of the Big Five systems integrators, based on market research obtained from firms such as Gartner, AMR, and Meta Group. Keating, 6/10/04, 890:17-891:23.

        2.7.1.7 BearingPoint will often assist its Tier One clients that are selecting software in preparing the solicitation documents that will be issued to the potential vendors. The documents typically have a list of requirements that the client needs for regulatory or business reasons, a list of technical requirements, and questions about the stability of the potential vendor, the vendor's service and support, and the vendor's training programs and capabilities. Keating, 6/10/04, 891:24-893:11.

      2.7.2 Customers typically issue detailed requests for proposals that outline the functions that the customer is seeking to automate. P4866, "Request for Proposal: CH2M Hill Human Resources Management System;" P3061, DeSimone dep., 5/19/04, 109:2-4, 139:9-18; Cichanowicz, 6/14/04, 1067:25-1068:03; P3062, Elliott dep., 5/20/04, 76:9-77:7.

        2.7.2.1 During the sales process for high-function software, the customer typically will issue a high level requirements definition. P3009, Block dep., 12/16/03, 166:12-17; 168:3-21.

        2.7.2.2 Typically customers will define issues such as the functionality required, the scope of work, and where the software will be deployed. P3009, Block dep., 15.12/16/03, 166:22-167:8.

        2.7.2.3 Vendors will follow-up with the customer to learn as much as possible about the customers requirements. P3009, Block dep., 12/16/03, 169:8-17.

      2.7.3 Second, high-function enterprise software is time consuming to purchase. The procurement process for some customers lasts from six to eighteen months and involves extensive interaction with software vendors and third-party consultants. Oracle's Answer, at Response to Allegation 18; P3191, Block dep., 12/16/03, 173:17-174:3.

        2.7.3.1 From Cox Communications' identification of the need to replace its FMS (formation of "Project Catalyst") until signing a contract (end of "Project Cornerstone") took from the Spring of 2003 until the summer of 2004. Hatfield, 6/7/04, 123:8-124:6.

        2.7.3.2 Emerson's ERP vendor selection process took 21 months. Peters, 6/14/04, 1264:24-1265:6.

        2.7.3.3 The sale of high function software is a long protracted process that may take 6 months to over a year to complete. P3036, Knowles dep., 12/3/03, 170:23-171:5.

        2.7.3.4 Nextel kicked off its "Financial, HR, Procurement Systems Overhaul Project" in October, 2002, and had a scheduled delivery of 2004. P4074R, at NEXTEL- 000074.

        2.7.3.5 When selecting enterprise software, large companies go through a much more detailed evaluation process - they do a deep-dive technical evaluation of your product. Smaller companies rely more heavily on references. P3171, Ellison dep., 1/20/04, 258:2-259:21.

      2.7.4 Demonstrations are designed to identify software products that can be configured to meet a customers business requirements .

        2.7.4.1 Two to four vendors are typically invited to do demonstrations of their products for a potential customer. Knowles, 6/23/04, 2911:12-15; P3036, Knowles dep., 12/3/03, 168:11-169:5.

        2.7.4.2 Customers will provide vendors detailed scripts that they expect the vendor to follow during the demonstrations. These are often very detailed because the customer does not want "any amazing discoveries." Bullock, 6/7/2004, 212:24-213:16.

        2.7.4.3 Customers expect vendors to demonstrate the abilities of the software to meet their functional requirements. Knowles, 6/23/04, 2911:6-2911:9.

        2.7.4.4 Nextel sent RFPs to Oracle and PeopleSoft. Both responded and performed demonstrations for Nextel. Nextel also did reference checks and had executive interviews with both Oracle and PeopleSoft. Cichanowicz, 6/14/04, 1071:13-24.

      2.7.5 Substantial information is exchanged between customer and vendors in the sales process. Customers provide vendors extensive information about their needs in order to ensure that they select the product that is the closest functional fit.

        2.7.5.1 Unless the vendor can demonstrate its ability to meet a customers functional requirements, there is no need to even discuss a price for the software. P3009, Block dep., 12/16/03, 224:7-18, 22-225:8.

        2.7.5.2 Oracle learns the potential customer's business processes and how its software meets the customer's business needs through a series of meetings, and demonstrations. Potential customers provide information regarding how its business operates so that Oracle can configure its software to run business scenarios for the potential customer during software demonstrations. P3009, Block dep., 12/16/03, 192:15-195:13.

        2.7.5.3 It is in the customers best interests to be forthcoming about their functional needs when purchasing enterprise software. P3009, Block dep., 12/16/03, 175:6-10.

        2.7.5.4 Oracle's dedicated North American sales force includes "SWAT sales teams" that are highly knowledgeable specialists who are expertise in the products, can explain the qualities to potential customers, and "understand how to beat the competition." P2054, at ORCL-EDOC-00049907.

        2.7.5.5 Many customers conduct a detailed assessment of their functional requirements as part of the software procurement process. Oracle's Answer, at Response to Allegation 19.

        2.7.5.6 The sales process allows a vendor to identify a prospect's requirements through a series of conversations with the customer and meetings over a period of time. Often in the sales cycle a vendor will send a team to a customer to find out the prospect's needs and capabilities, an interaction that gives the vendor a very good idea of the problem they're trying to solve and if their needs match the criteria of large and complex. The demonstration process likewise gives the vendor substantial amounts of information about the customers needs and capabilities. Bergquist, 6/8/04, 276:6-24.

        2.7.5.7 The sales process gives the vendor information about whether the prospect is attempting to solve the whole organization's problems or just a division; whether the solution is to be deployed worldwide or for a specific geographic area; whether the prospect needs a specific functionality or a broad suite of functionality to run its whole organization. Bergquist, 6/8/04, 276:25-277:9.

        2.7.5.8 Vendors obtain detailed information about the customers' business and unique requirements. P3050, Render dep., 5/11/04, 29:3-18.

        2.7.5.9 Substantial information is exchanged between vendors and customers concerning their requirements, so that it is "an iterative dynamic process." P3059, Johnson dep., 01/13/04, 87:5-88:5.

        2.7.5.10 High function software vendors will do an assessment up-front of whether the prospective client has the budget for the software and based on that assessment they will decide whether to pursue the client. P3036 Knowles dep., 12/3/03, 187:11-188:1.

        2.7.5.11 A vendor of high function software must collect a great deal of information about a potential customer's business and the issues faced by the client in order to package a software solution to meet that client's needs. The sales process becomes a collaborative effort between customer and vendor. P3036, Knowles dep., 12/3/03, 64:9-65:8.

        2.7.5.12 A customer reveals information about its business operation that enable vendors to make an informed judgment about the set of vendors that can meet the demand, as well as the customer's probable ranking of vendors. P3036, Knowles dep., 12/3/03, 168:11-168:23.

        2.7.5.13 During the sales process Peoplesoft has many meetings and conversations with a potential customer who conveys their business requirements. Customers try to present a complete picture of what they are looking for. Wilmington, 6/16/04, 1785:24-1786:24.

      2.7.6 Vendors seek and obtain information about how customers value the high- function software they are seeking to buy.

      2.7.7 Oracle and Peoplesoft attempt to learn how much the product is worth to the customer. Campbell, 6/22/04, 2785:11-25.

        2.7.7.1 Oracle monitors PeopleSoft's product line and knows the software's strengths and weaknesses. Oracle ensures that its sales force is kept up-to-date on this information. P2114, at OCRL-EDOC-00037470.

        2.7.7.2 A vendor of high function software must collect a great deal of information about a potential customer's business and the issues faced by the client in order to package a software solution to meet that client's needs. The sales process becomes a collaborative effort between customer and vendor. P3036, Knowles dep., 12/3/03, 64:9-65:8.

        2.7.7.3 Starting from the very beginning of a sales cycle, account executives spend a great deal of time with a prospective customer to understand the key decision drivers behind going to market for a new software solution. P3036, Knowles dep., 12/3/03, 164:20-166:3.

        2.7.7.4 Vendors seek and obtain information about how customers value the high-function software they are seeking to buy. P3036, Knowles dep., 12/3/03, 168:11-168:23.

        2.7.7.5 Customers have no incentive not to disclose all their requirements. The reason for this is that if a customer does not select the right application the customer might have to customize or write application code outside the application selected and this will add to the total cost of ownership. Wilmington, 6/16/04, 1786:25-1787:20.

        2.7.7.6 To understand the business processes that a customer has, Peoplesoft relies on its 'solution consultants.' These are experts who specialize in specific industries, or in specific software products, human capital management or financial management for example, or in both. Wilmington, 6/16/04, 1788:14-1789:16.

        2.7.7.7 Oracle's sales staff seek information from the customer about the price the customer expects to pay for the software. P3171, Ellison dep., 1/20/04, 289:20-290:13.

        2.7.7.8 Vendors seek information regarding business scenarios that the client needs the software to support so that they can run demos and workshops to show that the software can perform as desired. P3009, Block dep., 12/16/03, 169:22-170:16.

        2.7.7.9 In a competitive bid at Merrill Lynch for HR, an Oracle sales representative observed that "[t]his is for a Peoplesoft competitive replacement for the HR Suite and complete ERP commitment to Oracle. The price of PSFT migration to V8.8 is seen as expensive and considered a new implementation for Merrill. We have verified that this price point puts us in a competitive position.... Oracle is the only other vendor they are considering." Later, the representative came back with a new discount because "[a]fter discussions with Tony Kender and License sales management, we need to request an additional 3% discount for this PSFT competitive HR replacement.... PSFT has come in to protect their HR install at Merill and undercut our price ...." P1176, at ORCL-EDOC-00403221 (00403219-27).

      2.7.8 Vendors generally learn who else is on the short list of competitors being considered.

        2.7.8.1 Oracle typically knows who they are competing against when a potential sale is at the point where price is being discussed. P3009, Block dep., 12/16/03, 218:17-22; 219:1-9.

        2.7.8.2 Keating testified, based on his experience working for Oracle and PeopleSoft and his dealings with the software vendors, he believes that Oracle, SAP and PeopleSoft are fully aware of the functional capability of the other vendor's products. Keating, 6/10/04, 895:22-896:20.

        2.7.8.3 Customers sometimes "identify competing vendors and the prices that they are offering in an effort to encourage price competition." Oracle's Answer, at Response to Allegation 20.

        2.7.8.4 Some customers will explicitly tell the vendors who their competition is for the customer's business. Bullock, 6/7/2004, 214:18-214:25.

        2.7.8.5 In the selection of its FMS and its HRMS software, NM told OR and PS that they were the only remaining finalists after NM had eliminated SAP from the competition. OR and PS also knew that NM did not plan to retain its legacy system and did not intend to develop a system internally. Maxwell, 6/9/04, 670:19-671:7, 687:17-688:3.

        2.7.8.6 During head-to-head competition, Oracle and PeopleSoft did what they could to determine the status of the evaluation process and where they ranked against each other. Wolfe, 06/16/04, 1560:6-14; 1585:18-1586:10.

        2.7.8.7 Vendors will know the identity of the other vendors participating in the demonstrations. Knowles, 6/23/04, 2911:16-2911:19.

        2.7.8.8 Vendors try to learn who is on the short list. P3036, Knowles dep., 12/3/03, 168:11-168:23.

        2.7.8.9 During the sales process Peoplesoft learns who else is competing. Many times the customer identifies the competition. Based on industry knowledge, the customers business requirements, and feedback from the customer, Peoplesoft is able to determine who the competitors are. Wilmington, 6/16/04, 1789:17-1790:18.

        2.7.8.10 KMG disclosed to PS and OR that they were the finalists in it's KMG HRMS selection process. P3062, Elliott dep., 5/20/04, 118:14-20.

        2.7.8.11 Oracle salesmen are strongly encouraged to learn if there are other vendors competing for a customers business. P3009, Block dep., 12/16/03, 175:17-176:2.

        2.7.8.12 Knowledge of competitors in an account permits a vendor to package his offer to win against the competitor that is being faced. P3009, Block dep., 12/16/03, 180:20-181:9.

        2.7.8.13 The vendors also learn the current software footprint of the customer, which may include a split footprint of Oracle and PeopleSoft products. P1002, at ORLITE0086650-51; P1004, at ORCL-EDOC-00128424; P1019, at ORCL-EDOC-00224858.

      2.7.9 Different customers have different (heterogenous) needs. Vendors expect their salesmen to work with a prospective client understand that client's business problems or the business process issues. Knowles, 6/23/04, 2910:18-2911:5.

        2.7.9.1 Each customer for enterprise software is unique from the standpoint of how they procure the software, how they evaluate the software, and the business rules that need to be supported. P3009, Block dep., 12/16/03, 204:10-205:5

        2.7.9.2 Although the code provided to customers is the same, the use of the software by the client may be unique. P3009, Block dep., 12/16/03, 205:11-22; 206:1-19.

        2.7.9.3 The individualized needs of a customer may affect the vendors that it believes can satisfy its requirements. P3009, Block dep., 12/16/03, 208:5-10.

        2.7.9.4 Small business customers are "very different than other types of organizations" because "they have very simple needs." P3191, Block dep., 12/16/03, 83:20-84:10.

        2.7.9.5 Certain customers have unique needs that require highly configurable software. Therefore, they cannot purchase a solution in a box, which is a software version with limited options that comes pre-configured. P3191, Block dep., 12/16/03, 132:16-133:8.

        2.7.9.6 Individual customers have differing needs because of their different business practices and customers are to some extent unique. P3038, Godwin dep., 1/20/04, 126:1-22.

        2.7.9.7 Oracle witness Harris testified that in his experience as CIO at various divisions of Pepsico, Nike, and Gap, every customer has different needs and seeks different solutions to meet those needs. Harris, 6/24/04, 3196:5-19; 3228:3-25, 3229:1-9.

        2.7.9.8 Different customers have different (heterogenous) needs. P3036, Knowles dep., 12/3/03, 65:9-66:17.

        2.7.9.7 Up-market customers have different business issues and different business requirements. Peoplesoft and its competitors have experts who put on detailed demonstrations for customers to show how the application meets their business requirements. Wilmington, 6/16/04, 1787:21-1788:13.

      2.7.10 Prices are individually determined, customer by customer.

        2.7.10.1 Oracle uses a variety of pricing matrices for its application software, including pricing on a suite basis or according to the number of employees. P3191, Block dep., 12/16/03, 139:20-140:2.

        2.7.10.2 Typically, pricing is not discussed until the vendor understands the functional requirements of the customer. P3009, Block dep., 12/16/03, 212:18-213:16.

        2.7.10.3 Pricing (through the approval of discounts) is based on the circumstances of each transaction. P3171, Ellison Dep, 01/20/04, 44:4-46:8.

        2.7.10.4 Vendors of high function software compete on price and capability. P3171, Ellison dep, 01/20/04, 45:9-18.

        2.7.10.5 When pricing (discounting) high-function software, Oracle considers whether competing products can meet the functional requirements of the customers. P3171, Ellison Dep, 01/20/04, 44:4-45:8.

        2.7.10.6 Vendors such as SAP have good information about the functional capability of competitors such as Oracle and PeopleSoft and will not discount as heavily against vendors that cannot meet the needs of a particular industry. Knowles, 6/23/04, 2934:19-2935:7.

        2.7.10.7 A vendor of high-function software may not be forced to meet a price offered by a competitor if the vendor's product is a better functional fit to the customer's requirements. Knowles, 6/23/04, 2934:5-18.

      2.7.1l Vendors submits bids and prices are negotiated from those bids, often during a multiple-round negotiation process.

        2.7.1l.1 Usually two or three vendors are involved at the end when more intense price discussions occur. Wilmington, 6/16/04, 1790:19-1791:4.

        2.7.11.2 Keating testified that, when negotiating price, Tier One firms that are selecting software will always negotiate with 2 or 3 finalists and try and get the best deal possible. Tier One firms rarely try to negotiate with only one firm. Keating, 6/10/04, 895:6-21.

        2.7.11.3 Price is typically not discussed until after a customer is satisfied a particular vendor can meet their minimum requirements. Once the demonstrations have been completed vendors will have a good idea about the solution the customer is looking for and will be in a better position to provide pricing information. P3037, Knowles dep., 5/3/04, 162:19 - 163:6.

        2.7.11.4 The most important job in selecting software is to license a product that meets the organization's requirements, and a vendor whose product did not satisfy the customer's requirements probably would not be kept in play during the final negotiations just for leverage. Keating, 6/10/04, 897:2-21.

        2.7.11.5 Vendors cannot provide detailed pricing until after they understand the customer's specific requirements. P3036, Knowles dep., 12/3/03, 185:1-185:18.

        2.7.11.6 A customer will negotiate with two to three vendors on price. Knowles, 6/23/04, 2848:25-2849:6.

        2.7.11.7 The first price quoted to a prospective customer will be fairly close to list price. P3036, Knowles dep., 12/3/03, 190:9-191:2.

        2.7.11.8 Oracle is often told that there is another competitor's price that they must beat in order to win a deal. P3009, Block dep., 12/16/03, 216: 3-14.

      2.7.12 Maintenance & Enhancements

        2.7.12.1 Customers rely on vendors to support and maintain high function software over the life of the product.

          2.7.12.1.1 Customers cannot efficiently maintain their own software using internal resources. [REDACTED TEXT]

          2.7.12.1.2 The statement "software doesn't wear out" is misleading. Vendors of high function software must continually develop the software to meet the changing needs of customers and to ensure that the software is compliant with legal requirements. Knowles, 6/23/04, 2919:22-2920:16; see also P3150, at 1 (no Bates).

          2.7.12.1.3 Technology is not static and customers depend on upgrades. Cichanowicz, 6/14/04, 1122:05-24; access to upgrades in software is very important for customers. Cichanowicz, 6/14/04, 1075:12-21.

          2.7.12.1.4 Oracle sets maintenance fees as a percentage of the customer's net (discounted) license fee. P3192, Block dep., 1/05/04, 445:11-21.

          2.7.12.1.5 Oracle's fee for maintenance is 22% of the license price. P3349, Catz dep., 6/25/04, 3506:04-08.

          2.7.12.1.6 High-function software is usually purchased with a maintenance contract from the software vendor, often for 5 years for 15-20% of the purchase (license) price annually. Support, updates, and enhancements are included. Elzinga, 6/18/04, 2395:3-2397:9.

          2.7.12.1.7 Maintenance fees are typically calculated as a percentage of the list price of the new software product. The fees ranges between 15-25% depending upon what a company can negotiate. Maintenance contract typically need to be negotiated every 3-4 years. Gorriz, 06/15/04, 1385:6-24.

          2.7.12.1.8 It is extremely rare for a customer or a third party vendor to maintain high function software, for example SAP provides maintenance to well over 90% of its customers. P3037, Knowles dep., 5/3/04, 130:3-130:15.

          2.7.12.1.9 At the time of licensing a maintenance contract is also initiated. The maintenance fee is based on a percentage of the license fee. The maintenance contract provides the customer with three things: 'hot line support,' basically question answering on the application, updates to the application software, and enhancements or upgrades. Wilmington, 6/16/04, 1776:3-19.

          2.7.12.1.10 'Hot line support' is necessary when there is a problem such as a 'bug' in an application. A customer calls into the Peoplesoft support analyst who fixes the problem and sends the product fix to the customer. Wilmington, 6/16/04, 1776:20-1777:3.

          2.7.12.1.11 'Updates' are changes to an application that Peoplesoft sends to a customer and usually address changes in regulatory reporting, for example, changes in W-2 processing. Wilmington, 6/16/04, 1777:4-13.

          2.7.12.1.12 'Upgrades' or 'enhancements' are more significant changes where Peoplesoft provides a customer with increased functionality or changes in technology to a product. An example would be where Peoplesoft upgraded from release 7 to release 8 and changed the technology base from client server to internet enabled. Wilmington, 6/16/04, 1777:14-25.

          2.7.12.1.13 Maintenance contracts do not guarantee a certain level or amount of upgrades or updates but Peoplesoft is inventive and driven to provide a high level of service to meet customer demands. Wilmington, 6/16/04, 1778:1-7.

          2.7.12.1.14 Enhancements and upgrades are very important to customers. When a customer licenses an application Peoplesoft takes over responsibility for the application's development and enhancement, and for making sure it keeps pace with the needs of the industry in which the customer operates. This relationship drives the references a customer will give to other customers looking to license a Peoplesoft product. Wilmington, 6/16/04, 1778:8-1779:5.

          2.7.12.1.15 Virtually all High-function software customers take and maintain maintenance contracts. Wilmington, 6/16/04, 1779:8-1780:1.

      2.7.13 The investment in a new enterprise software application is significant. Bullock, 6/7/2004, 202:6-202:9. Typically in the range of multiple millions of dollars. Maxwell, 6/9/04, 676:10-13 ($1.9 million for New Mexico); Exhibit P4035, "Analysis of Vendor Proposals for New Financial Systems"($14.96 million for MetroNorth); Cichanowicz, 6/14/04, 1121:11-17 ($8 million for Nextel's license).

      2.7.14 Customers purchase High-function enterprise software because of the benefits that they receive from the software in supporting mission critical functions. High-function enterprise software is mission critical, failed or delayed implementation would have grave economic consequences. P3037, Knowles dep., 5/3/04, 120:3-121:16.

          2.7.14.1.1 Ford is a global manufacturer of cars and trucks, employing 350,000 persons and operating in 200 or so countries. Ford owns several large businesses, including a credit operation (Ford Credit) and a substantial car rental company (Hertz), as well as numerous automobile divisions responsible for specific brands (e.g., Ford Oval, Volvo, Land Rover, and Jaguar) within specific countries. P3041, Patel dep., 6/03/04, 8:12-20, 9:1-7.

          2.7.14.1.2 Ford's objective is to simplify its business processes by implementing PS HRMS software throughout the company as quickly as resources permit, and reducing the number of multiple systems Ford uses. P3041, Patel dep., 6/03/04, 54:25-55:5.

          2.7.14.1.3 KMG concluded that its new Human Resources Information System ("HRIS") would provide numerous company benefits and efficiencies. Its new system would enable it to effectively manage its global workforce, eliminate certain outsourcing, automate tasks that were previously performed manually, eliminate the need to run and support duplicate systems in multiple locations, assist employees and managers via self-service, and employ a vendor-supported system which would benefit from upgrades reflecting HR best practices. P0095R, at Kerr McGee013.

          2.7.14.1.4 KMG estimated that the complete implementation of its PS HRMS software would save it in excess of $4 million per year. P0095R, at Kerr McGee022.

          2.7.14.1.5 Enterprise customers want to standardize and consolidate their enterprise software applications in order to save costs. P3191, Block dep., 12/16/04, 98:8-15.

        2.7.14.2 Between 90 and 95% of Tier One, Fortune 2000, or Public Service organizations that begin the process of selecting a Tier One software solution will ultimately license software because they need to meet certain reporting requirements, they want to realize economies, or they want to employ management reporting to run their business. Keating, 6/10/04, 930:2-9.

        2.7.14.3 Oracle witness Harris testified that financial management software was mission critical to the companies where he worked as CIO (Gap, Nike, and divisions of Pepsico). Harris, 6/24/04, 3207:23-25, 3208:1-5.

        2.7.14.4 "[We could not operate the company without these kind of systems [HRM and FMS software]. Bradley, 6/9/04, 597:22-598:3.

        2.7.14.5 Oracle witness Harris testified that financial management software was mission critical to the companies where he worked as CIO (Gap, Nike, and divisions of Pepsico). Harris 6/24/04, 3207:23-25, 3208:1-5.

        2.7.14.6 [REDACTED TEXT]

      2.7.15 Total cost of ownership for High-function software consists of the license fee for the application, the maintenance fee, the cost to implement the software, and the cost to have ongoing support to run the application. Wilmington, 6/16/04, 1774:10-1775:1; P3050, Render dep., 5/11/04, 78:14:22.

        2.7.15.1 The license fee is only a small part of the total cost of ownership. Cichanowicz, 6/14/04, 1075:03-05, and along with the maintenance fee offer customers the fewest competitive options. Gorriz, 06/15/04, 1438:3-21.

          2.7.15.1.1 In exchange for the license fee, a customer receives a license to use the application and the application code and functional code to run the application. Wilmington, 6/16/04, 1775:2-22.

          2.7.15.1.2 For license and maintenance fees in the sale of high function software, PeopleSoft competes with Oracle and SAP. Wilmington, 6/16/04, 1781:22-25.

        2.7.15.2 Implementation includes business consulting to match software and business processes, installation, set-up, and configuration. P3191, Block dep., 12/16/03, 18:9-19:7; Wilmington, 6/16/04, 1780:6-18.

          2.7.15.2.1 The implementation of high function software is a lengthy and complex process. P3037, Knowles dep., 5/3/04, 120:3-121:16.

          2.7.15.2.2 In the up-market, implementing the product can take from six to 12 month or longer. There are thousands of 'tables' or menus of choices in the software that an implementation team must choose from to configure the software to a customer's business processes. Implementation is much more detailed than just flipping switches on and off. Wilmington, 6/16/04, 1780:2-1781:18.

          2.7.15.2.3 Implementation services are available from the software vendors or third-parties such as Accenture. Harris, 6/24/04, 3223:8-25, 3224:1-25, 3225:1-9:

          2.7.15.2.4 PeopleSoft, Oracle and SAP can implement their own software but do not compete to implement each others software. Competition for implementation comes from systems integrators such as IBM Global Services, Accenture, Bearing Point, Capgenimi Ernst & Young. Wilmington, 6/16/04, 1782:1-1783:2.

          2.7.15.2.5 When Peoplesoft sells software to the up-market it also offers technical training courses for the people who will be supporting the infrastructure of the application. Oracle and SAP do not compete to offer this training on Peoplesoft's products because they lack the expertise. Wilmington, 6/16/04, 1783:3-1784:1.

          2.7.15.2.6 The transaction for implementations services is a separate and distinct transaction from the license purchase and there are a different set of vendors who offer those services. Peters, 06/14.04, 1283:6-17.

          2.7.15.2.7 Mr. Peters also testified that there was nothing in Emerson's license agreement that "has any linkage to the prices" Oracle charges for implementation services. Peters, 06/14/04, 1283:6-17.

        2.7.15.3 The total cost of ownership can also include the cost of acquiring database software and additional computer hardware to run the application software. Wilmington, 6/16/04, 1784:2-16.

          2.7.15.3.1 PeopleSoft does not sell database software products. Wilmington, 6/16/06, 1784:2-12.

          2.7.15.3.2 Oracle, PeopleSoft, and SAP do not sell the computer hardware necessary to run their application products. Wilmington, 6/16/04, 1784:13-21.

          2.7.15.3.3

3 Product Market.

    3.1 Definitions

      3.1.1 The licensing and maintenance of high-function HRM software constitutes a relevant product market. Elzinga, 6/18/04, 2149:9-2149:23.

      3.1.2 The licensing and maintenance of high-function FMS software constitutes a relevant product market. Elzinga, 6/18/04, 2149:9-2149:23.

      3.1.3 High function software describes a software product that is capable of executing a wide array of business processes at a superior level of performance. Iansiti, 6/17/04, at 2035:1-17.

        3.1.3.1 High function software is scalable. Burgum, 6/23/04, 3011:19-3013:3; Bergquist, 6/8/04, 289:22-290:7.

        3.1.3.2 High function software is highly configurable (e.g., it must enable unlimited levels of organization). Johnsen, 6/16/04, 1753:9-1754:8; Iansiti, 6/17/04, 2046:19-2048:5, 2036:7-20; Bergquist, 6/8/04, 284:17-285:6, 287:9-17.

        3.1.3.3 High function software is able to perform a variety of related transactions seamlessly and with a high degree of sophistication. Wesson, 6/14/04, 1135:16-1136:13; Iansiti, 6/17/04, 2039:12-2040:7.

        3.1.3.4 High function software is able to provide robust functionality that allows organizations to go beyond the basics. Iansiti, 6/17/04, 2039:2-2040:7.

        3.1.3.5 High function software has the capability to handle international aspect of a multi-national organizations, such as multiple currencies, multiple languages, and multiple legal regimes. P2208, at ORCL-EDOC-00244092; Bergquist, 6/9/04, 290:15-23.

    3.2 Customers Consistently Identify Oracle, PeopleSoft, and SAP as the Only Vendors of HRM or FMS Software That Can Automate Complex Human Resource and Financial Management Processes.

      3.2.1 Organizations that have acquired software to automate complex human resource processes identified only three commercial software alternatives that could meet their requirements: Oracle, PeopleSoft and SAP. Bradley, 6/9/04, 598:4-599:11; Gorriz, 6/15/04, 1428:14-18; Maxwell, 6/9/04, 664:25-665:3, 685:21-686:1; Cichanowicz, 6/14/04, 1066:11-17; Hatfield, 6/7/04, 96:7-12, 114:22-115:10; [REDACTED TEXT] P3041, Patel dep., 6/3/04, 81:24-82:4.

      3.2.2 Organizations that have acquired software to automate complex financial management processes identified only three commercial software alternatives that could meet their requirements: Oracle, PeopleSoft and SAP. Bradley, 6/9/04, 598:4-599:11; Gorriz, 06/15/04, 1428:14-18; Maxwell, 6/9/04, 664:25-665:3, 685:21-686:1; Cichanowicz, 6/14/04, 1066:11-17; Hatfield, 6/7/04, 96:7-12, 114:22-115:10; D8107, Anderson dep., 6/3/04, 189:15-190:6; P3041, Patel dep., 6/3/04, 90:11-16.

      3.2.3 Organizations with complex requirements currently evaluating HRM and FMS software alternatives identify Oracle, PeopleSoft and SAP as the only vendors capable of meeting their requirements. Johnsen, 6/16/04, 1739:10-14.

      3.2.4 Complex organizations identify Oracle, PeopleSoft and SAP as alternatives based on a number of selection criteria including functionality, industry specific experience, price and vendor viability. Maxwell, 6/9/04, 669:23-670:8; P3171, Ellison dep., 1/20/04, 45:2-46:8..

        3.2.4.1 Organizations seeking international functionality have three options, Oracle, PeopleSoft, and SAP. P3155 at 2.

        3.2.4.2 [REDACTED TEXT]

        3.2.4.3 Large, complex companies only purchase software from vendors that can provide a reference from a company in the same industry. Gorriz, 6/15/04, 1375: 9-15; Bradley, 6/9/04, 600:9-602:3; P3171, Ellison dep., 1/20/04, 286:14-287:8.

        3.2.4.4 High function software vendors must be able to demonstrate a level of research and development investment and spending that will ensure that the software is constantly upgraded and maintained to make certain that software improves as technology improves. Iansiti, 6/17/04, 2037:1-2038:6.

        3.2.4.5 High function software vendors must have a dedicated sales force committed to understanding the requirements of large and complex organizations in different industries and can work to match the software as much as possible to the needs of the enterprise. Iansiti, 6/17/04, 2037:1-2038:6; Knowles, 6/23/04, 2903:5-2903:9.

        3.2.4.6 High function software vendors must have strong support organizations. Iansiti, 6/17/04, 2037:1-2038:6.

        3.2.4.7 Large, complex organizations purchase high function software from vendors with a solid financial background because they need the vendor to be around to support and enhance the product over time. Gorriz, 06/15/04, 1374:25-1375:8; Bradley, 6/9/04, 598:9-599:5.

      3.2.5 Customers expect high-function software vendors to maintain and support the software over the life of the product.

        3.2.5.1 Customers recognize that it is more costly to maintain and upgrade these software systems themselves than it is to pay the vendor to do so. Gorriz, 06/15/04, 1383:22-1384:9.

        3.2.5.2 Upgrading high-function software enables organizations to implement new business strategies. D8107, Anderson dep., 6/3/04, 110:15-113:14; P3189R at DOJ-DC-NOV-000020.

        3.2.5.3 Through continued research and development and ongoing demands from clients, high function software is continually evolving to better meet changing customer needs. Knowles, 6/23/04, 2919:22-2920:16; P3150.

        3.2.5.4 Most customers purchase a maintenance contract from the software vendor. Knowles, 6/23/04, 2918:6-7; Catz, 6/25/04, 3488:21-3490:2, at 3489:6-11.

      3.2.6 Accenture, IBM Global Services, BearingPoint, Deloitte, and CGEY ("The Big 5 consulting firms") recommend Oracle, PeopleSoft, and/or SAP as alternatives for clients seeking to automate complex human resource or financial management processes. Elzinga, 6/18/04, 2184:5-2188:9.

        3.2.6.1 Fortune 2000 and Public Services 500 organizations have a set of requirements that are best met by Oracle, SAP and PeopleSoft. Keating, 6/10/04, 890:9-16; P3183, [REDACTED TEXT].

        3.2.6.2 SAP, Oracle and PeopleSoft "are the only ones that provide a product that will be acceptable to a large company in terms of product capabilities, vendor reliability, scalability, cost, handling multiple currencies, multiple calendars, and multiple charts of accounts." P0203 at 11; Keating, 6/10/04, 912:15-916:7; P3183, [REDACTED TEXT].

        3.2.6.3 Oracle, PeopleSoft, and SAP are the only vendors with global functionality that can serve organizations with worldwide operations seeking to generate reports in multiple languages and multiple currencies and satisfy international requirements. Thomas, 6/9/04 474:13-475:5; P3183,.[REDACTED TEXT]; Keating, 6/10/04, 867:15-868:6.

        3.2.6.4 Organizations with 200 to 2,000 plus legal entities seeking to support the reporting requirements of those entities using a single software solution have three alternatives: Oracle, Peoplesoft, and SAP. Thomas, 6/9/04 475:18-476:3; Keating, 6/10/04, 869:8-870:5.

        3.2.6.5 Organizations with multiple lines of business seeking to support those businesses with a single software solution have three alternatives: Oracle, Peoplesoft, and SAP. Thomas, 6/9/04, 476:10-15.

      3.2.7 Third party consulting firms cite the product breadth and depth, marketing prowess, and geographic coverage of Oracle, PeopleSoft and SAP in characterizing those three firms as the market leaders in financial management and human resource management software. Elzinga, 6/18/04, 2183:21-2184:4.

        3.2.7.1 Forrester Research's 2004 survey of a number of HRM vendors concluded "that the three large ERP vendors - SAP, PeopleSoft, and Oracle - dominate the high end of the market with sophisticated HRMS offerings." P3324, at p.4.

        3.2.7.2 Forrester Research's 2004 survey noted that multinational organizations are limited to Oracle, PeopleSoft and SAP because the "global features and localizations for HRMS tend to be far stronger among the big three ERP vendors." P3324, at p.8.

        3.2.7.3 The META Group released a 2003 study that stated Oracle, PeopleSoft and SAP "demonstrate an ability to both define a vision/strategy for ERP [defined in the study as "a suite of enterprise business applications and at least one of the following manufacturing, supply chain, or customer relationship management"] and execute against is globally across a wide range of industries." P4352 at PS-C076279.

        3.2.7.4 The Gartner Group distinguishes Oracle, PeopleSoft, and SAP HRM software from other vendors based on several criteria. Elzinga, 6/18/04, 2181:24-2183:4.

      3.2.8 Customers would not have considered other vendors if Oracle, PeopleSoft, and SAP were all to increase their prices for software by 10 percent while the prices charged by other vendors were held constant. Gorriz, 06/15/04, 1381:11-1381:19; Maxwell, 6/9/04, 669:23-670:8, 686:19-24.

        3.2.8.1 Organizations with complex requirements must invest in costly customizations to fill gaps in functionality if it implements a mid-market software solution. Hatfield, 6/7/04, 189:8-189:18; Wesson, 6/14/04, 1133:12-1135:15.

          3.2.8.1.1 Customizing software makes upgrades more difficult and costly. Hatfield, 6/7/04, 157:20-158:8.

        3.2.8.2 Customers believe a hypothetical merger of Oracle, PeopleSoft, and SAP would result in a loss of negotiating leverage, higher software license and maintenance fees and a loss in innovation. D8107, Anderson dep., 6/3/04, 148:5-22.

    3.3 Oracle, Peoplesoft, and SAP Recognize That They Are Competing in a Three Firm Market to Supply High-Function Software to Organizations with Complex Requirements.

      3.3.1 Organizations with complex requirements seeking a centralized HRM or financial management software system have three options: PeopleSoft, Oracle, and SAP. Bergquist, 6/8/04, 398:24-399:3; 399:12-399:17; Burgum, 6/23/04, 3006:3-3006:9; [REDACTED TEXT].

        3.3.1.1 PeopleSoft, Oracle, and SAP are the only vendors of HRM and FMS software applications with the functionality and capability to serve complex organizations. Bergquist, 6/8/04, 277:10-15.

        3.3.1.2 PeopleSoft, Oracle, and SAP continue to compete after the initial software license because otherwise they risk losing the customer to one of the other two vendors. Wilmington, 6/17/04, 1948:16-1949:6.

          3.3.1.2.1 All three firms compete to offer the best service for existing customers. Wilmington, 6/17/04, 1948:16-1949:6.

          3.3.1.2.2 All three firms work to lower the total cost of ownership. Wilmington, 6/17/04, 1948:16-1949:6.

          3.3.1.2.3 All three firms work to make sure the customer would be willing to be a reference for other clients. Wilmington, 6/17/04, 1948:16-1949:6.

        3.3.1.3 Oracle, PeopleSoft and SAP respond to one another's product announcements and compete to bring new products to the market. P4933.

      3.3.2 Oracle's executives recognize they compete with SAP and PeopleSoft to sell software to organizations with complex human resource and financial management requirements.

        3.3.2.1 Oracle's Co-President Charles Phillips stated that "the back office applications market for global companies is dominated by an oligopoly comprised of SAP, PeopleSoft, and Oracle. The market is down to three viable suppliers who will help re-automate the back office business processes for global enterprises for years to come" in a report written or issued under his name in April 2002 when he was working as an analyst for Morgan Stanley. P3068, Phillips dep., 6/3/04, 157:7-22; P2290 at MS 00914.

        3.3.2.2 Oracle's Chief Executive Officer Larry Ellison identifies SAP and PeopleSoft as Oracle's biggest competitors for ERP application software. P3171, Ellison dep., 1/20/04, 164:8-165:7.

          3.3.2.2.1 Mr. Ellison identifies SAP and PeopleSoft as its two biggest competitors for financial management software opportunities in the United States and worldwide in terms of revenue. P3171, Ellison dep., 1/20/04, 183:3-11.

          3.3.2.2.2 Mr. Ellison identifies PeopleSoft and SAP as its most frequent competitors for human resources software opportunities in the United States and worldwide. P3171, Ellison dep., 1/20/04, 184:11-16.

        3.3.2.3 Oracle's primary competitors for selling HRM software to federal agencies are PeopleSoft and SAP. P3059, Johnson dep., 1/5/04, 254:19-255:4.

          3.3.2.3.1 Oracle, PeopleSoft, and SAP are the only commercially available HRM software implemented at federal agencies. P3059, Johnson dep., 1/5/04, 142:8-142:15.

      3.3.3 Oracle's annual reports filed prior to its offer for PeopleSoft consistently identify PeopleSoft and SAP as Oracle's primary competitors in the ERP applications software market. P2048 at 9, P2049 at 9, P2050 at 9.

        3.3.3.1 The annual reports for fiscal years 2003 and 2004, filed after the tender offer, signaled a shift in the description of the competitive landscape, as Oracle began describing the applications market as "highly fragmented" and identifying a number of additional "competitors." P2051 at 10; D7147 at 11.

        3.3.3.2 Mr. Ellison does not believe the ERP applications market suddenly fragmented between the time Oracle filed its fiscal year 2002 annual report and the time it filed its fiscal year 2003 annual report. P3171, Ellison dep., 01/20/04, 247:1-248:18.

      3.3.4 Oracle win-loss customer survey data consistently identify PeopleSoft as Oracle's most frequent competitor in larger accounts. Elzinga, 6/18/04, 2188:21-2189:18; [RED.]; P2093 at ORCL-EDOC-00042675; P2132 at ORCL-EDOC-00054306; P3192, Block dep., 1/5/04, 509:11-510:8; P2090 at ORCL-EDOC-00038647.

      3.3.5 Oracle's Discount Approval Forms reveal that Oracle discounts most aggressively when competing against PeopleSoft and SAP. Elzinga, 6/18/04, 2176:9-2177:2; 2179:5-14; P3175; P4015A; P4015B.

        3.3.5.1 Oracle does not want to lose on price to SAP or "especially" PeopleSoft. P2137 at ORCL-EDOC-00043877.

        3.3.5.2 "Oracle is in a very competitive competition with PeopleSoft (& SAP) for the Crane Global HR Project. Past history competing with SAP & PeopleSoft indicates the need to discount at the levels above to win the business." Cran Company Discount Approval Form. P1114 at ORCL-EDOC-00511095.

      3.3.6 IBM's Senior Vice President of Software Steve Mills believes large or complex organizations will have three choices to automate their complex human resource or financial management processes: Oracle, PeopleSoft and SAP.

        3.3.6.1 "Our perspective of [the acquisition] was that customers would have a far more limited choice where they're trying to get to a consolidated ERP environment, that they're going to be choosing between two vendors instead of three, choosing between SAP and Oracle instead of Oracle, PeopleSoft, and SAP, that SAP would, in fact, likely be a beneficiary in some instances where customers were not comfortable doing business with Oracle." P3193, Mills dep., 5/27/04, 110:25-111:19; 124:5-19.

        3.3.6.2 "And one area that we've seen a lot of activity around for trying to choose from a limited number of choices has been in the ERP space, accounting, ledger, basic bookkeeping, human resources tracking, common core applications.... Again, as we were talking about, the larger-size businesses are looking for sophisticated packages that have a lot of function that can match scalability. If you have thousands or even tens of thousands employees, any old accounting package is not going to do. It has to match the nature of your business, and the three dominant providers of that technology in the market today are SAP, PeopleSoft and Oracle." P3193, Mills dep., 5/27/04, 128:5-129:25.

    3.4 Organizations with Complex Human Resource or Financial Management Requirements Do Not View "Mid-Market" HRM or FMS Software Applications as a Substitute for the High-Function Software Sold by Oracle, PeopleSoft, and SAP.

      3.4.1 Oracle's expert concedes that high-function software is a different product from the software products designed and sold to small and mid-market organizations. Hausman, 6/29/04 4057:3-4063:11.

        3.4.1.1 "Mid-market" software designed and developed to meet the requirements of "mid-market" organizations and small businesses is not included in the relevant product market. Elzinga, 6/18/04, 2184:16-23.

        3.4.1.2 Large or complex organizations tend to require a more sophisticated and complicated business application software than do smaller organizations. Hausman, 6/29/04 4057:11-4059:4; Wilmington, 6/17/04, 1849:16-19.

        3.4.1.3 "Mid-market" software does not offer one or more of the following dimensions: breadth and depth of functionality; global functionality; industry specific expertise; scalability and reliability; and technology. Bergquist, 6/8/04 277:22-278:3.

        3.4.1.4 The software sold by Oracle to large organizations is not an "out-of-the-box" product. P3009, Block dep., 12/16/03, 125:5-125:13.

        3.4.1.5 Oracle's documents distinguish high-function software from mid-market software. Elzinga 6/18/04, 2190:2-12.

        3.4.1.6 Large organizations may license and use mid-market software solutions for use at the division level. Wilmington, 6/17/04, 1972:20-1973:1.

          3.4.1.6.1 Small divisions of a large organizations may find a mid-market solution viable. Wilmington, 6/17/04, 1972:20-1973:1.

          3.4.1.6.2 Divisions operated as a stand-alone profit center, might find a mid-market solution viable. Wilmington, 6/17/04, 1973:2-1973:5.

      3.4.2 There are number of commercially available HRM and FMS software solutions for small and medium sized organizations.

        3.4.2.1 "Mid-market" software describes less-flexible software developed for midsized enterprises or less complex enterprises. Bergquist, 6/8/04 275:25-276:5.

        3.4.2.2 "Off-the-shelf business software describes HRM or FMS software designed to be used as delivered, without any business transformation or specialized design or tailoring of the software to the client's business requirements. [REDACTED TEXT].

        3.4.2.3 "Out-of-the-box" software is quicker and cheaper to implement than high function software because the system is pre-configured and the customer does not have the ability to make as many choices. P3009, Block dep., 12/16/03, 126:9-127:14.

      3.4.3 "Mid-market" organizations do not have the same characteristics or functional needs as complex organizations. P2061 at ORCL-BLOK-002-001091; P3032, Henley dep., 5/4/04, 100:24-101:20.

        3.4.3.1 Mid-market organizations need a product with basic functionality and a streamlined implementation process. Burgum, 06/23/04, 2994:20 - 2995:02; Wilmington, 6/17/04, 1971:18 - 1972:4.

        3.4.3.2 Mid-market organizations typically process fewer transactions, have fewer employees, smaller IT budgets and typically have less sophisticated requirements than large organizations. P3032, Henley dep., 5/4/04, 112:25-114:05; P4492 at SAP-DOJ002509.

        3.4.3.3 Mid-market organizations do not need necessarily need the depth and breadth of functionality or the global functionality available in high-function software. Wilmington, 6/17/04, 1971:18-1972:4; P3036, Knowles dep. 12/3/03, 46:7-47:15.

          3.4.3.3.1 Large organizations tend to be more complex because they have silo-ed departments with specialized work that create a more complex business process. P4954 at PS-E00423392; P4955 at PS-E0042623.

          3.4.3.3.2 Some mid-market organizations may have complex requirements that demand high-function software. P4954 at PS-E00423392; P4955 at PS-E0042623.

        3.4.3.4 An organization's need for a "mid-market" or a high-function product cannot be decided by looking at a single factor or dimension, rather the distinction is made by looking at a number of dimensions: whether a customer seeks a basic system or a highly functional one; the number and variety of its business units; the reliability and scalability needs that flow from those factors; and the number of countries involved. Bergquist, 6/8/04, 395:11-396:12.

      3.4.4 "Mid-market" vendors design and develop enterprise software applications around the simpler requirements of mid-market organizations and small businesses. P3191, Block dep., 12/16/03, 102:18-103:1.

        3.4.4.1 Vendors targeting larger enterprises must design and develop more sophisticated business software applications. Hausman, 6/29/04, 4053:16-4054:11.

          3.4.4.1.1 Vendors must invest more in research and development if they are targeting the large organizations than they would if they were targeting smaller organizations. Hausman, 6/29/04, 4053:16 - 4054:11.

          3.4.4.1.2 Sophisticated software applications used by larger organizations must beable to greater numbers of concurrent users and transactions. Hausman, 6/29/04, 4053:16 - 4054:11.

        3.4.4.2 Vendors targeting only small and medium sized businesses do not have to invest in developing all the capabilities demanded by large or complex organizations. Hausman, 6/29/04, 4060:20-4061:16.

        3.4.4.3 J.D. Edwards, now marketed as PeopleSoft EnterpriseOne, successfully developed and sold EnterpriseOne to mid-market organizations. Bergquist, 6/8/04 274:18-23; 275:12-15; Wilmington, 6/16/04, 1766:21-1767:14.

          3.4.4.3.1 EnterpriseOne lacks the functionality to meet the needs of large, complex enterprises. Bergquist, 6/8/04 275:16-24.

          3.4.4.3.2 EnterpriseOne lacks the scalability necessary for large, complex enterprises. Bergquist, 6/8/04 275:16-24.

          3.4.4.3.3 EnterpriseOne lacks real global functionality demanded by multi-national organizations. Bergquist, 6/8/04 275:16-24.

          3.4.4.3.4 Mr. Ellison acknowledged that different vendors compete in different market segments. Ellison, 6/30/04, 4340:15 - 4341:16.

        3.4.4.4 Oracle documents describe the positioning of different mid-market software vendors based on research Oracle accumulated from various sources, including analysts' reports and Oracle's own observations in the market. See e.g., P0923 at ORLITP0058141; P924 at ORCL-EDOC-01156927; P947 at ORLITF0069746; P927, at ORLITF0035587; P3070, Prestipino Dep., 40:7-40:24;.42:2 - 43:3; 44:5-13; 46:20 - 47:9.

      3.4.5 Market participants recognize that small, medium, and large organizations each have different business requirements and selection criteria for enterprise application software. P2061 at ORCL-BLOK-002-001091; P3032, Henley dep., 5/4/04, 100:24-101:20.

        3.4.5.1 In the past, Oracle's segmented the market between "enterprise" customers and "general business" customers, drawing the line at $1 billion in annual revenues. P2050 at 8.

        3.4.5.2 Organizations with less than $500 million in annual revenues do not need high function software because they have simpler requirements than larger organizations. P3191, Block dep., 12/16/03, 86:3-86:20; P3191, Block dep., 12/16/03, 88:22-89:8.

        3.4.5.3 Oracle internal marketing documents focused on the mid-market, recognize lower mid-market ($20-$99 million, 10-100 employees), middle mid-market ($100 million-$249 million, 100-500 employees), and upper mid-market segments ($250million-$l billion, 500-1,000 employees). See e.g., P0923, at ORLITP0058141; P925, at ORCL-EDOC-01156796; P927, at ORLITF0035587; P2061, at ORCL-BLOK-002-001091.

        3.4.5.4 Oracle's mid-market marketing group observed that "Oracle has a strong representation with large enterprise customers, but is relatively unknown as providing technology and applications to mid sized companies." P926, at ORCL-EDOC-00222088; P3070, Prestipino Dep., 104:3 -105:11; P2060, ORCL-EDOC-00046832 - ORCL-EDOC-00046855.

        3.4.5.5 PeopleSoft defines "mid-market" as organizations with fewer employees, smaller annual revenues and simpler requirements than larger or multi-national organizations. Wilmington, 6/16/04, 1765:8-1765:22.

        3.4.5.6 Microsoft Business Solutions defines three broad categories: (1) small business (0-49 employees); (2) mid-market (50-1000 employees); and (3) enterprise (over 1000 employees) which is further divided into Corporate Account Space (1000-5000 employees) and Global Major & Strategic Accounts (over 5000 employees). Burgum, 06/23/04, 2981:3 - 2982:21; [REDACTED TEXT]

          3.4.5.6.1 Microsoft defines different market segments by using metrics such as the number of PCs, number of employees and amount of IT spending. Burgum, 06/23/04, 2978:9-20; [REDACTED TEXT].

          3.4.5.6.2 The demonstrative exhibit used during Oracle's opening statement to segment the market based on number of employees is not a taxonomy used by Microsoft. Burgum, 06/23/04, 2985:11-19. The skills associated with selling to a large, complex organizations is much different than the skills needed to sell to many mid-market businesses. P3032, Henley dep., 5/4/04, 117:5-119:1.

      3.4.6 Customers' actions and words demonstrate that they do not regard "mid-market"software as a substitute for high-function software.

        3.4.6.1 Metro-North Railroad eliminated GEAC once it reviewed GEAC's written responses to its initial Request for Proposal. P3061, Desimone dep., 131:2-131:19, P4032 "RFP No. 201121 Integrated Financial System (IFS) October 28, 2003," at NY-JP 043255.

        3.4.6.2 Greyhound eliminated mid-market vendors Lawson Software and Ultimatem Software based on those firms' functionality deficiencies. Glover, 6/15/04, 1470:22-25; 1483:24-1484:05; P4038R, at 54-55.

        3.4.6.3 Novell did not consider J.D. Edward's FM software as viable for Novell because JD Edwards was perceived as a mid-tier vendor that would not meet the needs of a company of Novell's size. D8107, Anderson dep., 6/3/04, 170:11-170:17.

      3.4.7 Vendors of high-functions software do not price their software to compete with "mid- market" or "off-the-shelf software solutions.

        3.4.7.1 High function software vendors do not price their product to compete with off-the shelf product. P3037, Knowles dep., 5/3/04, 78:9 - 78:15.

        3.4.7.2 Oracle's discount approval forms demonstrate that mid-market software vendors do not constrain Oracle's pricing. Elzinga, 6/18/04, 2179:5-14; P3175; P4015A; P4015B.

    3.5 SAP, PeopleSoft, and Oracle Also Compete in the Market for Small Business and Mid-Market Accounts.

      3.5.1 PeopleSoft, Oracle and SAP have all identified "mid-market" organizations as an opportunity for new sales. Knowles, 6/23/04, 2815:4-2816:5.

        3.5.1.1 Frank Prestipino, Oracle's Vice President for marketing Oracle's "mid- market" products views "mid-market" and small businesses as a growth opportunity for its technology and application product portfolio. P3070, Prestipino dep., 5/18/04, 5:23-7:06; 12:10-12:19.

        3.5.1.2 Oracle's mid-market marketing group observed in June 2003 that "[t]he suite vendors (SAP, PeopleSoft) that traditionally targeted the Fortune 1000 are moving down into the mid-market to sustain growth." P926, at ORCL-EDOC-00222071; P3070, Prestipino Dep., 86:5 - 90:19.

      3.5.2 Oracle, SAP and PeopleSoft all have dedicated salespeople selling solely to mid-market organizations.

        3.5.2.1 PeopleSoft established a separate sales force for customers with under billion in annual revenues. Wilmington, 6/16/04, 1765:23 - 1766:16.

        3.5.2.2 SAP has dedicated salespeople to sell to organizations with less than $1.5 billion in revenues. Knowles, 6/23/04, 2814:13 - 2814:21.

        3.5.2.3 Oracle has dedicated salespeople selling to "mid-market" organizations. P3033, Henley dep., 5/4/04, 116:12-117:4.

        3.5.2.4 BearingPoint has a dedicated mid-market business group focused on consulting opportunities at smaller organizations with annual revenues of less than $2 billion that are not as geographically dispersed. Keating, 6/10/04, 864:12-17; 863:11-18; 864:22 - 865:2.

        3.5.2.5 Selling to large sophisticated organizations requires the vendor to commit additional resources to the sales cycle. P3033, Henley dep., 5/4/04, 118:4 - 119:1.

      3.5.3 Oracle and the other vendors of high-function software may sell some of the same basic code to smaller, simpler organizations but with much of the functionality shut-off. Hausman, 6/29/04, 4057:3-4061:16.

      3.5.4 SAP's high function software has too much functionality and is too complex for organizations with less than $800 million in revenues. P3037, Knowles dep., 5/3/04, 114:7 - 114:20.

        3.5.4.1 SAP introduced software products with less functionality in order to effectively sell to smaller organizations. Knowles, 6/23/04, 2907:3 - 2907:7.

        3.5.4.2 SAP's All-in-One and Business One products are targeted at organizations with less than $750 million in revenues. P3036, Knowles dep., 12/3/03, 51:19 - 51:24.

          3.5.4.2.1 Technical limitations in the design and architecture of these products limit their use by larger, more complex organizations. P3036, Knowles dep., 12/3/03, 51:25 - 52:11.

          3.5.4.2.2 The number of transactions these products can process are limited. P3036, Knowles dep., 12/3/03, 51:25 - 52:11.

      3.5.5 The breadth and depth of the functionality of PeopleSoft's high function software was too complex for many mid-market organizations. Bergquist, 6/8/04, 298:11-25.

        3.5.5.1 PeopleSoft packaged PeopleSoft 8 for the mid-market by eliminating certain modules, pre-configuring the software and turning off certain features. Bergquist, 6/8/04 299:1-10.

        3.5.5.2 PeopleSoft acquired J.D. Edwards to complement Peoplesoft's business and enhance Peoplesoft's efforts in the mid-market. Wilmington, 6/16/04, 1766:18-1767:14.

        3.5.5.3 J.D. Edwards was viewed as an attractive acquisition target by all three high function software vendors because of its strength in the "mid-market."

          3.5.5.3.1 Oracle considered acquiring J.D. Edwards in April, 2003, in part because of its "strong mid-market presence." P0035 at 26; P3033, Henley dep., 5/4/04, 110:15-111:09; Catz, 6/25/04, 3497:25-3498:08.

        3.5.5.4 Peoplesoft's prices its mid-market products differently than it does the products it targets at the up-market. Wilmington, 6/17/04, 1972:5-19; see P4965, at PS-TE2251790, PS-TE2251794 (PS-TE2251788-99, listing separate price calculators for the up-market and mid-market and describing mid-market pricing).

      3.5.6 Oracle simplifies its high function software for use by smaller, simpler organizations because those firms do not need all the available functionality. Hausman, 6/29/04, 4057:3-4061:16.

      3.5.7 Oracle's mid-market products are targeted at organizations that do not need the breadth and depth of functionality available in Oracle's high-function software. P3191, Block dep. 12/16/04, 101:5-15; P3070, Prestipino dep. 5/18/04, 25:5-27:7.

        3.5.7.1 Oracle's large or complex customers get a different product than Oracle's smaller and mid-sized customers with simpler requirements. Hausman, 6/29/04 4057:3-4059:4.

      3.5.8 Oracle has committed resources to developing, selling and marketing simpler versions of its software to "mid-market" organizations and small businesses in North America. P3191, Block dep., 12/16/04, 113:16-114:10.

        3.5.8.1 Oracle's mid-market software and license are designed is such a way as to prevent post-sale deviations from the pre-configured software settings. P2060 at ORCL-EDOC-00046839.

        3.5.8.2 Oracle's mid-market products such as "Special Edition North America" are pre-configured for smaller businesses seeking a higher return on investment with their lower FT budgets and smaller and less sophisticated IT staffs. P3070, Prestipino dep., 5/18/04, 32:19-35:19.

        3.5.8.3 Oracle's Special Edition North America is a less complex solution for mid-20 market customers because it has fewer components than Oracle's E-business Suite. P3070, Prestipino Dep., 5/18/04, 36:25 - 37:23.

        3.5.8.4 Oracle pre-configures its mid-market software to make it easier and cheaper to implement. P3070, Prestipino Dep., 5/18/04, 37:24 - 38:12; 142:6-18; P0930, at ORLITF0069468.

      3.5.9 The release of Oracle's North American mid-market products has been delayed several times in the last eight months. P3070, Prestipino Dep., 51:17 - 52:20; 84:2 - 85:14.

        3.5.9.1 February, 2004 e-mail correspondence between Oracle managers suggests Oracle was concerned that the release of a mid-market product could hurt the company's efforts to acquire PeopleSoft by creating a perception that Oracle defined a mid-market and offered a mid-market product in the United States. P0934 at ORLITF0071161.

          3.5.9.1.1 Notwithstanding the clear implications of the e-mail correspondence, Mr. Prestipino, who received a copy, claimed he did not know about the issue concerning the Department of Justice. P3070. Prestipino Dep., 178:15 - 182:17.

        3.5.9.2 Another early February, 2004, e-mail exchange involving Chimene Stewart, public relations manager for Oracle's mid-market campaign, suggests a related effort by Oracle counsel to restrict Oracle marketing personnel's comments about defining the mid-market, pending a decision by the Department of Justice on whether Oracle's proposed acquisition of PeopleSoft would be challenged.

          3.5.9.2.1 Mr. Prestipino, after a break to confer with counsel, testified he did not respond to Ms. Stewart's request for guidance and was unaware of any such restrictions. P0935 at ORLITF0071197; P3070, Prestipino Dep., 101:20 - 102:6; 182:22 - 183:8; 183:13 - 186:9; 186:18 - 187:11; 188:3-17.

        3.5.9.3 Oracle's mid-market product is a pre-configured software package that has "5 fixed integrated business flows needed by any business," operates on low-end hardware, and can be used by a maximum of 25 users. P2060 at ORCL-EDOC-00046835.

    3.6 Customers Would Not Switch to a "Best of Breed" Solution (Or "Point Solution") in Response to a Small, but Significant, Nontransitory Price Increase in High Function FMS or HRM Software Because These Solutions Lack the Functionality Required by High-function Software and Are Costlier to Implement and Maintain.

      3.6.1 Customers do not perceive a best of breed package of different software solutions as a substitute for the high-function HRM or FMS software sold by SAP, Oracle and PeopleSoft. P3041, Patel dep., 6/03/04, 86:21-87:4, 92:2-16; Gorriz, 06/15/04, 1430:13-15.

        3.6.1.1 "[I]f best of breed would mean that we have to buy a couple of software pieces and glue them together and hope that it runs, I would say, no, that's not a true option for us." Gorriz, 06/15/04, 1429:22-1430:2.

        3.6.1.2 The functionality of best-of-breed point solutions does not provide enough added value to overcome integration and multi-vendor challenges. P2336, at KEANE 06210.

      3.6.2 The integrated suites of high function HRM and FMS software provide superior functionality and better performance than a package of best of breed solutions. Oracle's Answer, at Response to Allegation 16; Ellison, 6/30/04,4317:12-4317:14.

        3.6.2.1 Best of breed software does not provide core HRM or FMS functionality. Bergquist, 6/8/04, 310:18-311:2; 311:8-25; P3198, [REDACTED TEXT]; Cichanowicz, 6/14/04, 1110:15-1111:02 (stating that using point solutions was not a viable alternative for Nextel as it would not provide the core HR and FMS functionality).

        3.6.2.2 Best of breed software vendors do not have the depth of functionality provided by the high function HRM and FMS software sold by PeopleSoft, Oracle and SAP. Bergquist, 6/8/04, 310:18-311:2; 311:8-25; REDACTED TEXT] .

      3.6.3 Integrating point solutions with core software can be risky and costly because the integration layer can malfunction [REDACTED TEXT].

        3.6.3.1 Organizations using a best of breed approach may find it "staggeringly" difficult to access and store data using because business information is fragmented among many different databases. P3171, Ellison dep., 01/20/04, 139:9-140:4.

        3.6.3.2 By having information stored in a single instance with an integrated software solution, customers enjoy the benefits of easily accessing and sharing the information. P3033, Henley dep., 5/4/04, 39:25-42:5.

      3.6.4 Organizations pursuing a best of breed strategy must bear the costs of integrating the solutions which include dealing with different software vendors, different release schedules, and different interfacing strategies.

        3.6.4.1 Mr. Ellison describes best of breed as the "gift that keeps on giving" because organizations incur significant ongoing costs associated with upgrades and maintaining interfaces. P3172, Ellison dep., 01/20/04, 133:17-135:20; see also P3033, Henley dep., 5/4/04.

        3.6.4.2 Organizations can achieve significant costs savings by implementing integrated HRM or FMS software suites, up to five times as much over the life of the product, as compared to pursuing a best-of-breed approach. Ellison, 6/30/04, 4320:2- 4320:17; see also, P3172, Ellison dep., 01/20/04, 13715-139:8; Hatfield, 6/7/04, 135:3-10;D8107, Anderson dep, 6/3/04, 93:1-94:5; 102:20-103:9.

        3.6.4.3 A best of breed strategy requires a customer to balance different software vendors, different release schedules and different interfacing strategies. P3061, DeSimone dep., 5/19/04, 150:12 - 151:5. 151:20.

        3.6.4.4 The customer bears the burden of integrating a best of breed package falls on the company using it.

          3.6.4.4.1 A company can carry this burden "if you have a big solution and include some smaller packages into it, but if you have three or four core functionality which you have to fulfill with ... different software pieces, this is a nightmare." Gorriz, 06/15/04, 1430:3-12.

        3.6.4.5 Customers receive the most value with pre-integrated FMS and HRM software because it is purchase in one instance, is easier to pretest, runs on one technology and one hardware, and less costly. Hatfield, 6/7/04, 93:16-94:8; Wesson, 6/14/04, 1150:16-1151:20; P-3189-R at DOJ-DC-NOV-000025.

      3.6.5 Enterprises are seeking to standardize their operations using fewer vendors for human resource and financial management systems. See, e.g., Bullock, 6/7/2004, 246:6 - 247:7; P1130, at ORLITE0013367 (0013366-68) (KCI International seeking to replace its best of breed system).

        3.6.5.1 AIMCO did not seriously consider a "best of breed" solution to automate its complex HR processes because the total cost is relatively high. Wesson 6/14/04 1150:16-1151:20.

        3.6.5.2 Emerson is recognizing substantial savings by slowly standardizing on Oracle's E-business suite of applications after using 45 different ERP systems throughout the corporation's various divisions. Peters, 6/14/04, 1256:2-1260:12; 1262:3-9; D5170.

        3.6.5.3 Public sector customers are also recognizing the benefits of an integrated solutions and are "slightly more" sensitive to the costs and problems associated with integrating best of breed solutions. P3171, Ellison dep., 01/20/04, 299:12-16; P3370, Fitzgerald dep., 05/20/04, 132:14-18 (confirming that no federal agency goes to market requesting point solutions).

        3.6.5.4 Fleet Bank replaced its best of breed system with an integrated Oracle-based outsourcing system because the fragmented and complex architecture was costly to maintain and created user confusion. Mearns, 6/24/04, 3305:23-3308:4, 3313:24-3314:8.

      3.6.6 It is extremely rare for any customer to acquire core HR or FMS functions from different vendors, for example it would be a rare case for an organization to purchase its general ledger, account payable and accounts receivable systems from different vendors. P3044, Wohl dep., 4/29/04, at 233:11-22.

        3.6.6.1 Vendors of integrated software solution enjoy a competitive advantage over best of breed. Teece, 7/1/04, 4452:9-4452:15.

        3.6.6.2 Mr. Ellison predicts that best of breed vendors will no longer be competing for business, "the suite suppliers are going to win and the best of breed companies are going to lose." P3171, Ellison dep., 01/20/04, 135:21-137:14; P3171, Ellison dep., 01/20/04, 146:2-17; P3264, at ORCL-EDOC-01242172.

      3.6.7 Best of breed software vendors do not discipline Oracle's pricing behavior. Block dep., 12/16/03, 257:1-258:10.

        3.6.7.1 Customers would not consider adopting a best of breed approach even if the license cost of the high-function software solution sold by Oracle, PeopleSoft and SAP was 10 percent higher. Cichanowicz, 6/14/04, 1077:10-24; Hatfield, 6/7/04, 137:20-138:2.

    3.7 Customers Would Not Develop FMS or HRM Software in Response to a Small, but Significant, Nontransitory Price Increase in High Function FMS or HRM Software.

      3.7.1 The costs of developing in-house solutions are such that customers would not consider developing in-house software solutions in response to a small, but significant, non-transitory increase in price. Hatfield, 6/7/04, 137:13-19 ("[I]t would be many, many times more money spent"); Bradley, 6/9/04, 607:8-15; Maxwell, 6/9/04, 670:9-18, 686:25-687:9.

        3.7.1.1 Customer will not divert resources towards developing in-house ERP systems when it is not within the customer's core-competencies. Block dep., 1/05/04, 526:13-527:18.

        3.7.1.2 Customers will not develop in-house ERP systems because of the time, expense, and scale required to develop high function software equivalent to Oracle, PeopleSoft and SAP. P3192, Block dep., 1/05/04, 528:15 - 532:4; see also P3041, Patel dep., 6/03/04, 84:3-13, 90:25-91:18 (explaining how for core functionality it does not make sense given that better functionality is available commercially).

        3.7.1.3 It is not rationale for a customer to devote the time and expense to developing in-house software when adequate products are available for license. P3171, Ellison dep., 01/20/04, 253:19-254:19; Bradley, 6/9/04, 602:19-604:6.

      3.7.2 SAP, Oracle, and PeopleSoft can leverage significant economies of scale in developing high-function HRM and FMS software that cannot be matched by customers pursuing in-house development of enterprise software. P3192, Block dep., 1/05/04, 531:13-19.

        3.7.2.1 Oracle, PeopleSoft, and SAP can spread the development costs associated with high function software over their customer bases, whereas a customer developing similar software in-house would need to fund the entire development by itself. Maxwell, 6/9/04, 660:19-661:14.

        3.7.2.2 "If you consider the thousands of man-years that companies like Oracle have put into creating this suite, and you look at the amount of money I spent, I could only invest a handful of person-years if I did it for myself. I could never recreate something as rich as what I can get in a suite, if I wrote it myself." Hatfield, 6/7/04, 135:11-136:9.

        3.7.2.3 Nextel would not develop HRM or FMS software to automate its complex business processes "[b]ecause of the tremendous cost of doing custom development. You already have with a vendor costs that have been incurred that could be spread across many companies as they make sales. If you do it yourself, you've got to reinvent that wheel and you've got to incur all the cost yourself." Cichanowicz, 6/14/04, 1078:18-1079:03.

      3.7.3 Customers generally decide whether they will build a "homegrown" solution before conducting a competitive bid procurement for software. D7166, Morea, 5/7/04, 40:10-24; see also P3370, Fitzgerald dep., 05/20/04, 262:02 - 263:06. (confirming that Oracle would generally not be competing with in house solutions at the price competition stage).

    3.8 Customers' Threats of Using Their Incumbent HRM or FMS Software Would Not Prevent a Significant Price Increase in High Function FMS or HRM Software.

      3.8.1 The possibility of remaining on an incumbent solution is not a constraint on prices for high function FMS and HRMS software. Elzinga, 6/18/04, Tr. 2393:9-18; 2192:21-2197:11.

        3.8.1.1 [REDACTED TEXT]

        3.8.1.2 Mr. Wilmington testified that while some customers might delay a purchase, he does not "view [the do nothing option] as a long-term solution for that particular customer." Wilmington, 6/16/04, 1792:15 - 1793:1.

        3.8.1.3 Oracle's discount approval forms demonstrate that legacy systems are not "disciplining the Oracle pricing the way the two manufacturers of high- function FMS software and HRM software do, and that's SAP and PeopleSoft." Elzinga, 6/18/04, Tr. 232179:5-14; 2193:19 - 2194:22.

        3.8.1.4 Price discrimination allows Oracle to separate the elastic and inelastic customers, including customers who may choose to stay with a legacy system for some amount of time. Elzinga, 6/18/04, Tr. 2195:20-2197:11.

        3.8.1.5 Customers generally will not remain on their legacy systems in the event that vendors raise the cost of licensing or maintaining high function software 10%. See, e.g., Cichanowicz, 6/14/04, 1078:9-17 (noting that even if the cost of the PeopleSoft license were 10 percent higher, Nextel would not have integrated its existing software); [REDACTED TEXT]; Hatfield, 6/7/04, 138:3-19 (stating the same for Cox); Peters (Emerson), 6/14/04, 1269:24-1274:23.

      3.8.2 Large or complex organizations rarely decide to remain on an incumbent system once they have begun to evaluate high function FMS and/or HRM software alternatives.

        3.8.2.1 Generally when a customer issues a request for proposal, the customer intends to purchase software, not remain on its legacy system. P3370, Fitzgerald dep., 05/20/04, 105:6 - 106:2.

        3.8.2.2 Less than 5% of the time will a prospective client decide to do nothing at the end of a sales cycle and continue to use an incumbent system. Keating, 6/10/04, 930:2-19; Knowles, 6/23/04, 2914:4 -2914:21.

          3.8.2.2.1 Organizations have "reporting requirements they have to meet, there's economies that they want to realize in their business, there's management reporting that they want to be able to use to run their business. They're not about to -- once they've gotten the capital approval to go buy that, they very rarely stop in the middle of that and go, no, I'm not going to do it." Keating, 6/10/04, 930:2-19.

        3.8.2.3 Customers rarely remain on incumbent systems because they incur significant costs during the sales process associated with understanding the business processes it is seeking to automate and evaluating alternatives. Wilmington, 6/16/04, 1791:5-1793:1 ("[a]lmost never do I see a company that's invested in an evaluation like this do nothing.").

        3.8.2.4 In those rare instances where a customer begins the procurement process but decides to "do nothing," it is typically due to a problem in funding, not a lack of intent to procure or because then decides that its legacy system is a viable alternative. See, e.g., Glover, 6/15/04, 1490:6-22 (explaining that Greyhound fully intended to purchase software, but needed to hold off only because a new CEO cut the project's budget).

      3.8.3 Organizations running older version of software or using legacy systems face increasing maintenance and support costs over time. Peters, 6/14/04, 1211:19-1212:5; P3061, DeSimone dep., 5/19/04, 251:16 - 253:11.

        3.8.3.1 Organizations using legacy software must invest in costly customizations to their existing systems as they grow in size and scope. [REDACTED TEXT].

        3.8.3.2 It would have been risky and costly for AIMCO to maintain its legacy systems because AIMCO would have to "maintain the product ourselves ... -- there's no support for it outside the walls of our company, and for a strategic application for a company of our size, that's pretty risky." Wesson 6/14/04 1142:10-22.

      3.8.4 Customers running legacy systems would not delay a decision to acquire high- function software. See, e.g., Bradley, 6/9/04, 607:16-608:8.

        3.8.4.1 Customers may face some urgency in acquiring and implementing a new FMS system because of the real possibility of a system failure. [REDACTED TEXT]

        3.8.4.2 Organizations must constantly adapt to changing business and legal requirements which requires updates in the software to meet the organization's changing needs. Elzinga, 6/18/04, 2194:23 - 2195:19, 4015A; Bradley, 6/9/04, 627:2-22, 634:3-634:8; P3192, Block dep., 1/05/04, 532:19-533:15.

          3.8.4.2.1 Organizations would have to incur significant costs if they had to maintain and support its legacy systems. Gorriz, 06/15/04, 1433:13-1434:1

          3.8.4.2.2 Many legacy systems are written in programming languages that are outdated and cannot be supported, thus requiring customers to replace such systems. P3192, Block dep., 1/05/04, 534:11-535:14; Glover, 6/15/04, 1466:17-1467:05, P4038-R, at GDD 005157-005158; P3061, DeSimone dep., 5/19/04, 252:17 - 253:7, 254:10-19.

        3.8.4.3 Many times the original vendor of a customer's system is no longer marketing the software, thus leaving no upgrade path. See, e.g., P3062, Elliott dep., 5/20/04, 86:22-87:16.

          3.8.4.3.1 Neiman Marcus would have had to invest significant money and resources to upgrade and maintain its legacy financial software if it had not licensed Oracle's high function FMS software solution. Maxwell, 6/9/04, 655:25-656:20, 660:19-661:7, P2316-R, at DOJ.KY.0003.

          3.8.4.3.2 Ford's multiple legacy HR systems cost it from $10-50 million per year to maintain and support. This is a significant cost for Ford. P3041, Patel dep., 6/03/04, 57:20-58:7.

      3.8.5 The functionality of legacy systems does not match the depth and breadth of functionality available in the high-function software sold by Oracle, PeopleSoft and SAP. Wolfe, 06/16/03, 1536:3 - 1537:04; Maxwell, 6/9/04, 655:25-656:20, 660:19-661:7, P2316-R, at DOJ.KY.0003.

        3.8.5.1 North Dakota replaced its incumbent systems because they lacked functionality and the costs of maintaining and supporting the systems was steadily increasing. Wolfe, 06/16/03, 1592:14 - 1593:10.

        3.8.5.2 [REDACTED TEXT]

        3.8.5.3 Emerson is standardizing on Oracle's E-business suite in part because many of its important divisions had software that was becoming obsolete and placing "abnormal risk" on its information systems platforms. Peters, 6/14/04, 1267:4-22.

    3.9 Large and Complex Customers Would Not Outsource in Response to a Small, but Significant, Non-Transitory Price Increase in High Function HRM Software.

      3.9.1 Outsourcing is primarily an HRM phenomenon, few, if any, organizations of any size outsource FMS functions. Elzinga 6/18/04 Tr. 2197:12 - 2198:11.

        3.9.1.1 Sarbanes-Oxley and other recent developments associated with financial reporting makes is more important than ever for Ford to be in control of its own financial processes and have good checks in place. P3041, Patel dep., 6/03/04, 85:17-86:8, 91:19-92:1.

        3.9.1.2 Novell would not seriously consider outsourcing FMS functions because the financial management system impacted processes that were a "core part of being successful as company" and outsourcing it would be "too risky." D8107, Anderson dep, 6/3/04, 94:6-95:9.

      3.9.2 The choice to outsource a particular function is a corporate strategy decision and is not driven by the prices of high function software licenses or maintenance. See P3171, Ellison dep., 01/20/04, 185:21-186:20.

        3.9.2.1 Organizations make the decision to pursue an outsourcing solution before entering into a sales process for HRM or FMS software.

          3.9.2.1.1 Mr. Ellison explained that "if someone's considering a service, an online service, rather than software, early if they decide to buy a service and they never even consider us, we're never really in the deal. We lost at the conceptual level very, very early on. And they had a beg off between ADP and Fidelity rather than between us and our competitors." P3171, Ellison dep., 01/20/04, 184:19-185:20 (emphasis added); D7166, Morea, 5/7/04, 39:10-25, 133:23 - 134:21.

          3.9.2.1.2 Mr. Johnson of Oracle also testified that a customer will specifically state whether a customer demands an outsourcing service or a software solution, which dictates the competitive proposals that a customer receives. P3059, Johnson dep., 1/13/04,90:20 - 91:11.

        3.9.2.2 Ford wanted to maintain control over its core HR areas involving employee records and basic employee demographics. P3041, Patel dep., 6/03/04, 85:17-86:8, 91:19-92:1.

        3.9.2.3 Cox Communications did not pursue an HRM strategy because implementing the software generates advantages from internal integration of HRM and its customer management system that cannot be leveraged as efficiently through outsourcing. Hatfield, 6/7/04, 97:15-98:9.

        3.9.2.4 Service oriented organizations rely on the skills and abilities of their employees and these organizations are uncomfortable with entrusting a third party to maintain and track employee data. Bullock, 6/7/2004, 211:9-21.

        3.9.2.5 Novell would not consider outsourcing its core HR functions because, among other reasons, it is too risky. Anderson dep, 6/3/04, 100:5-101:6.

      3.9.3 Customers seeking to automate complex core HRM and FMS processes and functions do not consider outsourcing a viable alternative to the high-function software sold by Oracle, PeopleSoft and SAP.

        3.9.3.1 The Big 5 consulting firms have few customers that seek outsourcing for the entire core HR functionality. [REDACTED TEXT] (testifying that he is not aware of any of his clients that have outsourced their HR business process functionality); Keating, 6/10/04, 962:4 - 963:15 (testifying that outsourcers like ADP and Ceridian provide only some HR functionality, and that he is not aware of any BearingPoint clients who have outsourced the entire HR functionality).

        3.9.3.2 Verizon could not justify outsourcing the entire HR functionality "in part, it's because the risks of our business are fairly high.... We've outsourced what we can that's simple, that is fairly generic, and we have not outsourced some of the darker corners, such as payroll, that are not simple and not generic." Bradley, 6/9/04, 604:7 - 605:16.

        3.9.3.3 AIMCO did not was not an option because "[o]utsourcers generally are not very quick to respond to the last-minute changes. And that might change the way that - we have to make decisions inside of our company. I guess the other thing, one - some of the things I referred to earlier, work flow, business rules, integration with the financial system, those things really would -- would be much diminished under an outsource." Wesson, 6/14/04, 1150:01-15.

      3.9.4 Customers lose flexibility to configure the process to meet their unique requirements when they choose to outsource a function.

        3.9.4.1 Outsourcers do not have the flexibility to customize the process for each customer, "the outsourcer tries to have a uniform process and benefit from economy of scale, so the reason that the outsourcers are in this business is they can - because they're going to have highly specialized labor, benefit by economies of scale by processing thousands of payrolls rather than just one. And if every company insisted on their own processes, then that economy of scale would be lost and their ability to deliver a high quality, low cost service would evaporate." P3171, Ellison dep., 01/20/04, 186:21-187:13.

        3.9.4.2 ADP limits customers' ability to customize its Enterprise payroll product to drive efficiency in staffing. P3180, Rising dep., 5/28/2004, 55:10-56:3; 58:2-10.

        3.9.4.3 Mr. Gorriz of DaimlerChrysler explained "[u]sually you can outsource if you have standard processes which other people have as well. And then you get scale in going with an outsourcer." But, if a company's needs are complex, "there is no economic benefit to outsource the processes." Gorriz, 06/15/04, 1382:11-1383:1; 1377:16-1378:2.

      3.9.5 HR outsourcing vendors target organizations with simple HR requirements, for example, [REDACTED TEXT]

        3.9.5.1 Fidelity is not targeting their outsourcing service to larger organizations with complex business process requirements. Fidelity has only ten customers with more than 10,000 employees. Sternklar, 6/23/04, 3169:11-21.

        3.9.5.2 Fidelity's business plan calls for it to sign up one additional one client with more than 10,000 employees in 2004 and as few as three in 2005. Sternklar, 6/23/04, 3169:11-21.

        3.9.5.3 [REDACTED TEXT]

        3.9.5.4 [REDACTED TEXT]

      3.9.6 Outsourcing vendors targeting larger organizations will support those customers using an HRM software platform licensed from Oracle, PeopleSoft, and SAP, which makes them a complement of high-function HRM software, not a substitute for the relevant product. Elzinga, 6/18/04, Tr. 2199:12-20.

        3.9.6.1 ADP's recent announcement that it was partnering with SAP was implicit acknowledgment that ADP's proprietary software could not meet the requirements of multi-national organizations. P3181R, Brand dep., 5/18/04, 15:23-16:6.

        3.9.6.2 [REDACTED TEXT]

      3.9.7 Oracle, PeopleSoft and SAP compete infrequently with outsourcing vendors such as ADP for HRM opportunities at large organizations.

        3.9.7.1 PeopleSoft rarely competes with ADP for large or complex accounts. Wilmington, 6/17/04, 1923:2-12, 1924:25 -1925:7, 1926:19-23.

        3.9.7.2 SAP does not compete with ADP in the enterprise application software market "[b]ecause they sell services and we sell software." P3037, Knowles dep. 5/3/04, 272:9 - 272:19; 272:20 - 273:11.

      3.9.8 Outsourcing vendors license high function software from Oracle, PeopleSoft, or SAP to provide functionality lacking in their proprietary software.

        3.9.8.1 ADP recently signed a license agreement to use SAP as its global HR platform. P3181R, Brand dep. 5/18/04, 15:23-16:6.

        3.9.8.2 Most of the HR outsourcing vendors (Hewitt, Exult, Mellon, IBM, Fidelity, and Accenture) have some type of licensing arrangement with PeopleSoft, Oracle, and/or SAP. Sternklar, 6/23/04, 3158:8-3159:20; Sternklar, 6/23/04, 3162:4-16.

        3.9.8.3 [REDACTED TEXT]

      3.9.9 Outsourcers who base their services on Oracle, SAP, and/or PeopleSoft software cannot constrain pricing in the high function HRM market because outsourcing is a complement to high function software and not a substitute. Elzinga, 6/18/04, 2199:12-20 (describing outsourcing as "essentially a value-added service in the HR pillar").

        3.9.9.1 Oracle's Discount Approval Forms demonstrate that outsourcing vendors are not a significant pricing constraint on Oracle. Elzinga, 6/18/04, 2197:12-2199:11.

        3.9.9.2 Oracle does not present alternative bids for using its outsourcing services or licensing its software because they are not substitutes, they are complements. P3191, Block dep., 12/16/03, 263:5-15.

      3.9.10 Customers would not switch significantly to outsourcing in the event high function software prices rise 10%-15%

        3.9.10.1 Neiman Marcus would not outsource its core HR functionality in response to a 10% to 15% increase in HRMS software from PeopleSoft or Oracle because Neiman Marcus believes that it needs to manage and control core HR functionality internally. Maxwell, 6/9/04, 687:10-16.

        3.9.10.2 Mr. Cichanowicz testified that if the price of licensing ERP software rose 10%, Nextel would not have considered outsourcing. Cichanowicz, 6/14/04, 1079:04-18.

        3.9.10.3 In response to a small but significant increase in price, Cox Communications would not consider switching to outsourced solutions because of the comfort in keeping systems internal. Hatfield, 6/7/04, 138:20-139:8.

        3.9.10.4 A small change in the license price of the ERP system would not have affected Fleet Bank's decision to switch from a commercial ERP system to outsourcing. Mearns, 6/24/04, 3312:13-3313:3

          3.9.10.4.1 Fleet Bank chose to switch its PeopleSoft-based platform for an Oracle based outsourcing platform. Mearns, 6/24/04, Tr. 3313:25-3314:8.

          3.9.10.4.2 Fleet Bank did not consider ADP because ADP does not have adequate functionality for Fleet Bank's HR needs. Mearns, 6/24/04, Tr. 3313:6-20.

          3.9.10.4.3 Fleet Bank's experience demonstrates that outsourcers attempt to minmize their clients' need for variation from the standard business process model. Mearns, 6/24/04, Tr. 3302:20-3303:6.

    3.10 Outsourcing Vendors Using Their Own Proprietary Software Do Not Compete with Oracle, PeopleSoft, and SAP for Organizations with Complex HR Requirements.

      3.10.1 The software acquired or developed by outsourcing vendors is typically developed for the mid-market, thus not competitive with the high function software sold by Oracle, PeopleSoft, and SAP. Iansiti, 6/17/04, 2070:9-25.

        3.10.1.1 Outsourcing vendors with their own proprietary software do not offer the breadth or depth of functionality provided as the high function software sold by PeopleSoft, Oracle, and SAP.

        3.10.1.2 Outsourcing vendors using software they developed or acquired lack global functionality, flexibility (via customization), and scalability and technical efficiency provided by high function software.

      3.10.2 Oracle, PeopleSoft, and SAP recognize that they do not compete with the proprietary software marketed by the outsourcing vendors. Wilmington, 6/16/04, 1813:3-17; P3037, Knowles dep. 5/3/04, 272:9-273:11 (observing that he has "never seen ... a loss" to an outsourcer against SAP's HR suite).

      3.10.3 Outsourcing vendors using proprietary software recognize that their software is not equivalent to the software sold by Oracle, PeopleSoft and SAP.

        3.10.3.1 ADP recognizes that no other HR and payroll software package could offer the breath and depth of functionality of Oracle, PeopleSoft, and SAP. P3181R, Brand dep. 5/18/04, 39:19-42:3.

        3.10.3.2 ADP identified only three global HR and payroll platform choices on which to service its global clients: Oracle, PeopleSoft, or SAP. No other HR and payroll software package could offer the breath and depth of functionality of the three participants in the high-function markets. P3181R, Brand dep. 5/18/04, 39:19-42:3.

      3.10.4 Fidelity has no plans to develop the HR Access product it acquired from IBM to be a global HR product. Sternklar, 6/24/04, 3186:21-3187:16.

        3.10.4.1 Fidelity is in the process of re-writing the code for HR Access into a proprietary software and using it as the platform for Fidelity's outsourcing service.

        3.10.4.2 Fidelity's plans for HR Access will not be completed until 2009. Sternklar, 6/23/04, 3180:20-3181:11.

        3.10.4.3 Fidelity's re-write of HR Access has been plagued with problems of delay, caused in part by concerns with robustness and third party contributions, that has impacted the productivity and timeliness of the project. Sternklar, 6/23/04, 3184:3-17; 3186:8-13.

        3.10.4.4 [REDACTED TEXT]

      3.10.5 ADP markets its proprietary HR software, ADP Enterprise, primarily to mid- market organizations.

        3.10.5.1 ADP has approximately eight to twelve customers with more than 10,000 employees using its proprietary Enterprise HR product. P3180, Rising dep., 5/28/04, 72:2-72:7.

        3.10.5.2 Jay Rising, president of ADP's national accounts division, testified that ADP views payroll as its primary business and only sells its Enterprise HR product coupled with payroll as an incremental option to augment the core payroll business. P3180, Rising dep., 5/28/2004, 13:8-14:20; 45:14-46:1.

        3.10.5.3 ADP cannot provide multi-national clients with global functionality for payroll and HR services that provided consistent record keeping, reporting and functionality among worldwide operations. P3181R, Brand dep. 5/18/04, 17:5-19:15; 26:13-27:19.

          3.10.5.3.1 ADP cannot offer a single payroll engine for global companies in multiple countries. P3180, Rising dep., 5/28/04, 80:8-14; 96:6-13.

          3.10.5.3.2 ADP's existing products lacked global integrated functionality. P3181R, Brand dep. 5/18/04, 42:4-45:20.

        3.10.5.4 ADP's existing platform must be sold separately in different domestic components of multi-national corporations. ADP will offer the Project Gloria service, available in two years, to allow U.S. corporations to perform one function for the entire global corporation. P3180, Rising dep., 5/28/04, 92:8-93-2; 104:7-13.

    3.11 All the Theoretically Available Options (Listed Above) to High Function Software, Taken Together, Would Not Make a SSNIP in High Function Software Unprofitable.

      3.11.1 A hypothetical monopolist or cartel in the sale of high function FMS or HRM software (including license fees and maintenance costs) could profitably raise prices above the prevailing level.

        3.11.1.1 The price discrimination mechanism through which the relevant products are sold means that the sale to the marginal customer will not constrain the sale to any other customer. Elzinga, 6/18/04, 2234:18-2235:21; 2222:24-2225:9; 2336:1-17.

        3.11.1.2 Critical loss analysis is inapplicable to markets characterized by price discrimination. Elzinga 6/18/04, 2289:24 - 2291:16.

        3.11.1.3 Demand is inelastic because license and maintenance fees are only a small proportion of TCO. Elzinga 6/18/04, 2382:21-2383:22, at 2382:21-2382:7..

        3.11.1.4 Demand is also inelastic because of the mission-critical nature of the software. Elzinga 6/18/04, 2382:21-2382:22, at 2382:8-22.

        3.11.1.5 High margins indicate a narrow market: if broader alternatives were available to consumers, margins would be competed down. Elzinga 6/18/04, 2336:1-17.

      3.11.2 In the event a hypothetical monopolist made a small, but significant, nontransitory price increase in the cost of the high function software, customers would not switch to many of the "alternatives" discussed in these proposed findings.

          3.11.2.1.1 Cox Communications would not have considered Lawson, Microsoft's Great Plains, writing its own software, a best of breed approach, outsourcing, or upgrading its existing system if Oracle and PeopleSoft increased the license and maintenance price of their FMS software 10%. Hatfield (Cox), 6/7/04, 136:10-18, 137:1-139:8.

          3.11.2.1.2 Daimler would not have considered additional vendors beyond Oracle, PeopleSoft, and SAP, even if these vendors offered licensing and maintenance fees that were 10% higher for their HR software. Gorriz (Daimler), 6/15/04, 1381:11-19.

          3.11.2.1.3 North Dakota would not have considered Microsoft, Lawson, outsourcing, SCT, developing its own software, or remaining on its current system if the best and final offers from Oracle and PeopleSoft were 10% higher. Wolfe, 6/16/04, 1569:15-1570:16.

          3.11.2.1.4 Nextel would not have considered a best of breed solution, outsourcing, or remaining on its legacy system even if the license cost of high function software were 10% higher. Cichanowicz (Nextel), 6/14/04, 1077:10-24, 1078:9-17, 1079:4-18.

          3.11.2.1.5 Customers would not develop an in-house software solution in response to a small, but significant, non-transitory increase in price. Bradley (Novell), 6/9/04, 607:8-15; Maxwell (Neiman Marcus), 6/9/04, 670:9-18, 686:25-687:9; Cichanowicz (Nextel), 6/14/04, 1078:18-1079:03.

          3.11.2.1.6 [REDACTED TEXT] and Oracle's witness, Mr. Peters of Emerson, also testified that their respective employers would not have stayed on their incumbent systems in the event that the vendors raised the cost of licensing and maintaining the software that was being offered 10%. P3020, [REDACTED TEXT]; Peters (Emerson), 6/14/04, 1269:24-1274:23.

          3.11.2.1.7 Mr. Maxwell of Neiman Marcus testified that Neiman Marcus would not outsource its core HR functionality in response to a 10% or 15% increase in HRMS software from PeopleSoft or Oracle. Maxwell, 6/9/04, 687:10-16.

4 Geographic Market. The United States is a relevant geographic market for high function HRM and FMS software.

    4.1 Selling High Function HRM and FMS Software in the United States Requires That a Vendor Have a Presence in the United States

      4.1.1 Software can be shipped worldwide at little cost. Elzinga, 6/18/04, 2154:1-2156:10, at 2155:1-8; Elzinga, 6/18/04, 2204:12-2206:19, at 2204:12-20.

      4.1.2 But it is not just the software code that is relevant to the geographic market analysis, because selling to and servicing United States customers requires a presence in the United States.

        4.1.2.1 High-function software, by its nature, requires an ongoing business relationship with the vendor. "What you buy when you buy this product is, you buy a relationship; that is, you don't just buy code, you buy [a] relationship that begins before you even make the - before you even cut the deal, before you even actually choose the vendor, you meet for many months in the the purchase process of trying to learn which vendors will meet your needs if you're a prospective customer. And then after you make the purchase, once, again, it's very different from just taking on code. That relationship continues as the FMS software or HRM software is installed, implemented, bugs are taken out. It's maintained. It's upgraded. And that's really where the competition takes place in that local relational sense." Elzinga, 6/18/04, 2152:21-2156:10, at 2155:13-25; 2203:22-2206:19, at 2205:12-13 ("what's important is how this product is marketed and supported"); 2280:24-2281:10.

        4.1.2.2 Recognizing the importance of "presence" in relation to a vendor's ability to sell high function HRM and FMS software, Oracle believes that PeopleSoft does better in major U.S. metropolitan areas where it has a presence, but not as well in "secondary cities" where PeopleSoft is "relatively thin." P2114, ORCL-EDOC-00037470.

      4.1.3 Customers need vendors to establish a local presence so the vendor can explain to clients the complexity of the products and help fit those products to the client's individual needs. Knowles, 6/23/04, 2903:5-2903:14.

        4.1.3.1 It is important to Nextel that the vendor that it deals with be one that Nextel has a "relationship" with and that has "a presence in the United States physically, where we could see and talk to them on a consistent basis for such a strategically important function as this." Cichanowicz, 6/14/04, 1080:16-25.

        4.1.3.2 DaimlerChrysler made separate purchases of HR software for its German and U.S. operations. Gorriz, 06/15/04, 1378:3-9. DaimlerChrysler would not buy HR software from a company that did not have a presence in the local market for its German operations because it needs a lot of consultancy in the presale phase and technical support for implementation and operation of the software. Gorriz, 6/15/04, 1378:14-25.

        4.1.3.3 SAP struggled to gain traction in the U.S. marketplace for a number of years. SAP established a significant U.S. presence because customers wanted to do business with a local entity. P3036, Knowles dep., 12/3/03, 15:10-22; Knowles, 6/23/04, 2901:20-2902:3; Knowles, 6/23/04, 2902:4-22.

        4.1.3.4 SAP's development organization, SAP Labs, has a significant U.S. presence. Knowles, 6/23/04, 2807:1-25.

    4.2 Nothing Links Together Customers of High Function HRM and FMS Software, Either Within the United States or Between the United States and Elsewhere

      4.2.1 "Arbitrage" means "the simultaneous purchase and sale of substantially identical assets in order to profit from a price difference between the two assets." http://dictionary.reference.com/search?q=arbitrage (citing David L. Scott, Wall Street Words: An A to Z Guide to Investment Terms for Today's Investor (Houghton Mifflin Co. 2003)).

      4.2.2 Arbitrage is "the economic force that frequently links different territories together into a common geographic market," Elzinga, 6/18/04, 2204:12-2206:19, at 2205:19-2206:2, and the high function HRM and FMS software markets are not subject to arbitrage.

      4.2.3 License restrictions preclude arbitrage through resale. Elzinga, 6/18/04, 2232:9-16, 2157:23-2158:21, at 2158:2-8.

      4.2.4 Arbitrage through customers seeking purchases outside the United States also does not occur. Elzinga, 6/18/04, 2205:11-2206:19, at 2206:3-16.

      4.2.5 The fact that the software is configured for a specific customer also precludes arbitrage because once the vendor "has spent literally months on the grounds of the [customer] getting that software - getting the functionality of that software, sometimes referred to as flipping the switches, getting all the switches flipped properly so it works at the [customer] and meets the specific configuration and capabilities that [the customer] has, it won't work at [other customers]." Elzinga, 6/18/04, 2156:11-2158:21, at 2158:15-21.

      4.2.6 This detailed iterative process results in pricing being set on a customer-by- customer basis. That is, vendors of high function FMS and HRM software engage in price discrimination. Elzinga, 6/18/04, 2222:9-2225:9.

      4.2.7 Oracle sells products in foreign countries that U.S. customers do not appear to purchase abroad and bring back to the United States. P2060, ORCL-EDOC-00046832 to ORCL-EDOC-00046849.

    4.3 Oracle and PeopleSoft Ensure That Their High Function HRM and FMS Software Contains Up-to-date Country-specific Functional Characteristics, Making Arbitrage More Difficult and Price Discrimination Possible.

      4.3.1 Oracle creates one master-version of its mid-market products for each country designed for that country's "specific statutory requirements." P2060, ORCL-EDOC-200046835 to ORCL-EDOC-00046836.

      4.3.2 PeopleSoft's price calculator charges according to how many localizations a customer uses: "Many of PeopleSoft's global products contain language translations and/or local functionality for many international countries. Because these localizations significantly increase the value of the software, we license our global products by 'Named Country.' The use of one Named Country is included in the base product price. Each additional country selected on the Named Country Tab will incur an uplift." P4965, PS-TE2251794.

      4.3.3 In updating its HRM product, Oracle conducts "statutory maintenance," through which it ensures that its high function HRM product is kept current with U.S. statutory requirements, including OPM (Office of Personnel Management) and FLSA (Fair Labor Standards Act) regulations. P2895 at ORLITF0091164 (OPM updates), at ORLITF0091184 ("US focus" performance enhancements), at ORLITF0091190 ("FLSA support"). This type of updating and support, focused specifically on U.S. customers, differentiates Oracle's product from that of SAP.

      4.3.4 PeopleSoft differentiates its HRM product from SAP's in part through its focus on U.S.-centric functional characteristics. P2542 at PS-C006434, PS-C006435.

      4.3.5 The fact of price discrimination and the impossibility of arbitrage means that sales in Europe or other parts of the world have no bearing on the prices charged to US consumers. Elzinga, 6/18/04, 2156:11-2157:1; 2284:16-2285:3.

    4.4 the Effects of the Proposed Acquisition on Customers Would Differ Significantly Both among Customers in the United States and Between Customers in the United States and Elsewhere

      4.4.1 Given the nature of the sales process for high function HRM and FMS software, see supra Section 2, which involves a detailed iterative process between the customer and each potential vendor, each customer will be effected differently by the lessening of competition that will result from Oracle's proposed takeover of PeopleSoft given the vendor's ability to discern each customer's different intensity of demand. Elzinga, 6/18/04, 2222:24-2225:9, at 2224:22-2225:9 (responding to question from the Court).

      4.4.2 United States market shares reflect the relative demand intensity for the products of the three market participants within the United States, compared with other geographies. Elzinga, 6/18/04, 2207:2-14; 2264:1-9.

      4.4.3 SAP's worldwide market shares are different than those for the United States. That difference reflects SAP's relative inability to penetrate certain industries in the United States, such as banking. Knowles, 6/23/04, 2940:10-2941:24.

        4.4.3.1 Vendors must design software to account for the fact that U.S. organizations often run their business operations differently than organizations in other parts of the world. For example, SAP's "software is not designed to compute average daily balances for whatever reason, because that's, you know, in Europe that's not the way they run their banking operations. We do need to remember that we did grow out of a European centric need development organization." P3036, Knowles dep., 12/3/03, 126:8-127:4, at 126:25-127:4.

        4.4.3.2 PeopleSoft believes that one of the reasons it has not "seen SAP very often" 15is because SAP "operates in different geographies." SAP only derives 23% of its revenues from the U.S. P4172 at PS-C006268.

      4.4.4 When selling in the United States, SAP has also had to contend with customer perceptions that its products are "costly and difficult to implement" (Knowles, 6/23/04, 2950:8-16), that SAP has been "slow to change and slow to meet customer needs" (Knowles, 6/23/04, 2950:17-20), that SAP has "had problems maintaining strategic relationships with partners" (Knowles, 6/23/04, 2150:21-23), that SAP's products have "a functionality" (Knowles, 6/23/04, 2950:24-2951:2), that SAP is "not as strong in the services or public sector industries in North America" (Knowles, 6/23/04, 2951:14-18) and that SAP's R3 product was "challenging to configure and modify and had a steep learning curve for users" (Knowles, 6/23/04, 2951:19-23).

      4.4.5 Oracle believes that SAP is a less potent competitor to Oracle in the United State than PeopleSoft. In Oracle's parlance, "SAP is not the enemy, as much as you'll want to spank them. PeopleSoft is the enemy. Bury them." P2841at ORLITF0050734.

        4.4.5.1 SAP faces hurdles in the U.S. that PeopleSoft does not face, including SAP's reputation for being large and unwieldly, difficult to implement and prone to massive cost-overruns. P3050, Kender dep., 5/11/04, 260:7-267:1, at 261:11- 263:11, 265:10-266:8; P2841.

        4.4.5.2 By contrast, PeopleSoft has sold HRM in the United States for 12 years, and had a large installed base and large "mindshare", which made PeopleSoft a more potent competitor to Oracle than SAP. P3050, Kender dep., 5/11/04, 260:7-267:1, at 266:10-267:1; P2841.

    4.5 Defining a United States Geographic Market Does Not Exclude Any Relevant Sellers on the Market's Supply Side

      4.5.1 All market participants capable of selling high function HRM and FMS software in the United States are accounted for by the geographic market of the United States. Elzinga, 6/18/04, 2273:11-20.

      4.5.2 For purposes of analyzing the impact of Oracle's proposed acquisition of PeopleSoft, the only foreign vendor that supplies the relevant product in the United States is SAP. Elzinga, 6/18/04, 2153:14-25. Indeed, there are no other vendors of high function HRM and FMS software that a customer could turn to. See infra Sections 4.6.2 to 4.6.5; see also Iansiti, 6/17/04, 2025:2-13 (Dr. Iansiti examined 148 vendors and found only Oracle, PeopleSoft and SAP could meet high function requirements).

      4.5.3 SAP has a significant presence in the United States and its activities in the United States are counted as part of Dr. Elzinga's analysis, notwithstanding the fact that SAP is a foreign vendor.

        4.5.3.1 SAP America has over 4000 employees selling, marketing and supporting the sale of SAP software in the United States. P3036 Knowles dep., 12/3/03, 15:23-16:11.

        4.5.3.2 SAP has established a separate marketing, finance and sales organization in the United States. Knowles, 6/23/04, 2902:23-2903:4

        4.5.3.3 SAP has a regional headquarters outside of Philadelphia and regional offices scattered throughout the United States. P3036, Knowles dep., 12/3/03, 15:23-16:11.

      4.5.4 Dr. Elzinga's analysis thus only excludes foreign vendors to the extent that they have no competitive impact in the United States. Elzinga, 6/18/04, 2153:15-25.

    4.6 Other Than SAP, There Are No Vendors Outside Of The United States That Would Be Positioned To Serve Customers' Needs Ffor High Function HRM and FMS Software in The Face Of A SSNIP

      4.6.1 Dr. Iansiti examined 148 vendors of enterprise application software and found that Oracle, PeopleSoft and SAP were the only of these vendors that sold high function HRM and FMS software. Iansiti, 6/17/04, 2025:2-13.

      4.6.2 Mr. Wesson of AIMCO does not know of any vendors that sell HR software of the type that AIMCO needs outside the United States but that are not currently selling inside the United States. Wesson, 6/14/04, 1151:21-1152:1.

      4.6.3 Cox Communications would be unable to license software from abroad to counter a domestic price increase. Hatfield, 6/7/04, 134:9-135:1. In response to a SSNIP, Cox Communications would not consider switching to software from firms without a presence in the United States. Hatfield, 6/7/04, 141:4-21.

      4.6.4 Mr. Cichanowicz of Nextel testified that he was not aware of any companies outside of the United States that are selling either HRM or FMS software that would meet Nextel's needs. Cichanowicz, 6/14/04, 1080:11-15.

      4.6.5 Mr. Cichanowicz also testified that Nextel would not consider a vendor that does not sell ERP software in the United States today to be a viable option for Nextel's HRM and FMS software needs. Cichanowicz, 6/14/04, 1080:11-25.

      4.6.6 The fact of price discrimination and the impossibility of arbitrage means that sales in Europe or other parts of the world have no bearing on the prices charged to US consumers. Elzinga, 6/18/04, 2156:11-2157:1; 2284:16-2285:3.

      4.6.7

      5 Firms in the Market. Oracle, PeopleSoft and SAP are the only three companies that make and sell high-function HRM and FMS software in the United States and accordingly, are the only three companies in the relevant markets.

    5.1 Oracle, PeopleSoft, and Sap Are in a Different Class from All Other Vendors of Enterprise Resource Planning Software in Their Ability to Meet the Needs of Large and Complex Enterprises.

    5.2 Expert Testimony Confirms That Oracle, PeopleSoft and Sap Are the Three Rivals That Determine Pricing in the Relevant Markets. Elzinga, 6/18/04, 2158:22-2159:9; McAffee, 6/21/04, 2450:25-2451:1-11; 2458:25-2460:1-18.

      5.2.1 Customers of high function software view Oracle, PeopleSoft and SAP as their only competitive alternatives.

      5.2.2 Nextel asked its consultants to identify what software Nextel's peer companies were using. The majority were using Oracle, PeopleSoft, or SAP. Cichanowicz, 6/14/04, 1068:18-1069:06.

      5.2.3 Mr. Wesson of AIMCO learned from a review of its human resources needs that the three firms that could meet its human resources software needs were PeopleSoft, Oracle, and SAP. Wesson, 6/14/04, 1132:4-16.

      5.2.4 Oracle, PeopleSoft and SAP were the three firms that Neiman Marcus felt could meet the FMS and HRM requirements of a company the size of Neiman Marcus. Maxwell, 6/9/04, 664:25-665:13, 685:21-686:1.

      5.2.5 DaimlerChrysler concluded that only SAP, PeopleSoft and Oracle could meet its HRM software needs because these were the only companies that had the financial track record, functionality and scalability to meet its needs. Gorriz, 06/15/04, 1376:7-17.

      5.2.6 If the prices of HRM software charged by Oracle, PeopleSoft and SAP were to increase by 10 percent while the prices charged by other vendors were held constant, Daimler Chrysler would not consider any other vendors for its HRM software needs "because only the three are potentially vendors that can satisfy all needs" of Daimler Chrysler. Gorriz, 6/15/04, 1381:11-19.

      5.2.7 [REDACTED TEXT]

      5.2.8 Ms. Bradley of Verizon testified that Oracle, PeopleSoft and SAP have superior technology, and that there are no feasible commercially available alternatives for Verizon that would meet its HRM and FMS functional needs. Bradley, 6/9/04, 598:4-599:11.

      5.2.9 Mr. Hatfield of Cox Communications testified that only Oracle, PeopleSoft and SAP could satisfy Cox's business case to replace its FMS software. Hatfield, 6/7/04, 114:22-115:10.

      5.2.10 The State of Ohio evaluated five ERP vendors against requirements for functionality, scalability, technical fit, R&D spending, and public sector spending. The State of Ohio short-listed Oracle, PeopleSoft, and SAP as most closely fitting its needs. AMS, J.D. Edwards and Lawson did not make the short list in the Ohio RFI. P1944, at ORLITF 0149246.

      5.2.11 Many large organizations do business in multiple countries and have software requirements for multiple languages, multiple currencies and multiple legal entities. Keating, 6/10/04, 00867:15 to 00868:6. As Keating (Bearing Point) testified, "The bigger the company, the more currencies, the more countries, the more legal entities, the more languages, the more Oracle, SAP, and PeopleSoft meet those requirements." Keating, 6/10/04, 868:3-6.

      5.2.12 The bulk of Bearing Point's Global Enterprise Solutions Group clients that require software capable of dealing with multiple legal entities use Oracle, SAP and PeopleSoft. Keating testified that, "Oracle, SAP and PeopleSoft are the three clear, you know, players in the marketplace." Keating, 6/10/04, 870:1-10.

      5.2.13 Bearing Point's Global Enterprise Solutions Group clients require software that allows them to maintain multiple charts of account so that they can track revenue by all their different products and services and combination thereof. No software vendor provides multiple charts of account functionality that is comparable to Oracle, SAP or PeopleSoft. Keating, 6/10/04, 871:3-20.

      5.2.14 Dortenzo (Deloitte) testified that Deloitte's clients turn to Oracle, PeopleSoft and SAP because the scope of the applications is broader and they offer more functionality than other software. Some of Deloitte's clients have global operations that cause them to look for software solutions that can manage their multinational organization requirements, multiple currency requirements, multiple language requirements, and multinational, country-specific regulatory and reporting requirements. Dortenzo dep., 5/5/04, 48:21-51:6.

    5.3 Oracle, Peoplesoft, and SAP Primarily View Each Other as Competitors for the Sale of High-function Software.

      5.3.1 Oracle contemplated and then ran a "replacement" campaign to convert customers from PeopleSoft and SAP enterprise software to its enterprise software, but did not operate similar campaigns against other software vendors. Oracle initiated the campaign to "cause some pain for these guys." P2116, at ORCL-EDOC-00039444; P2115, at ORCL-EDOC-00039432; Block dep., 1/05/04, 417:5-418:15.

      5.3.2 Oracle had a "Kill PeopleSoft Project" and "Kill SAP Project," but no similar program directed at Lawson or other competitors in the last few years. Block dep., 1/05/04, 375:2-378:8.

      5.3.3 PeopleSoft identifies its competitors in the up-market as Oracle and SAP. "They are the ones we run into all the time." Wilmington, 6/16/04, 1773:9 - 1774:5.

      5.3.4 In April, 2003 Oracle identified itself, PeopleSoft, and SAP as the only three "competitive" software vendors in the ERP applications market. P0035 at 3; P3033, Henley dep., 5/04/04, 92:13-94:03.

      5.3.5 In an April 26, 2002 Morgan Stanley report concerning PeopleSoft, Chuck Phillips, a current co-President of Oracle, wrote, "Stepping back a bit, the back-office applications market for global companies is dominated by an oligopoly comprised of SAP, PeopleSoft, and Oracle." P2290 at MS 00914.

    5.4 Other firms are not in the market.

      5.4.1 Lawson is a mid-market firm whose focus and capabilities are in delineated verticals- health care, retail and public sector for some purposes. The lack of high function features is a significant limiting factor, even within these industries.

        5.4.1.1 Lawson's CEO believes people would characterize Lawson as a mid-market competitor outside its verticals. Coughlan, 6/28/04, 3668:21-3669:6.

        5.4.1.2 In an April 2003 META Group document that placed Oracle, PeopleSoft and SAP as the leaders in Enterprise Resource Planning, META noted that Lawson "lacks breadth across numerous industry verticals, geographic coverage, and sufficient brand awareness" and that Lawson has "developed expertise in a smaller number of specific verticals." PX 4352 at PS-C076279.

        5.4.1.3 Lawson considers Oracle and PeopleSoft to be "horizontal players" because their focus is not restricted to a limited number of verticals. Coughlan, 6/28/04, 3661:25-3662:1-10.

        5.4.1.4 Lawson's sales and marketing efforts over the past several years have been in the vertical areas of health care, retail and public sector. Coughlan, 6/28/04, 3662:17-24.

        5.4.1.5 As part of its vertical strategy, Lawson has developed sopecific functionality targeted towards hospitals, integrated delivery networks, academic medical centers. Coughlan, 6/28/04,3662:25-3663:2; 3663:14-21.

        5.4.1.6 In its June 24, 2004 earnings call, Lawson stated that its vertical focus would not change, regardless of the outcome of the antitrust trial in San Francisco. Coughlan, 6/28/04, 3666:17-25-3667:1.

        5.4.1.7 In order for an industry to be delineated as a separate vertical within Lawson, it needs to produce enough revenue to equal five percent of Lawson's total revenue. Health care, retail, and public sector are the only three verticals that generate such revenues. Coughlan, 6/28/04, 3667:23-3668:1-13.

        5.4.1.8 Despite a four year effort, the financial services vertical has yet to generate at least five percent of Lawson's overall revenues. Coughlan, 6/28/04, 3668:14-20.

        5.4.1.9 Financial services does not get any separate research and development or marketing dollars because it is not a delineated vertical within Lawson. Coughlan, 6/28/04, 3677:18-3678:1-9.

        5.4.1.10 Lawson's ability to successfully expand beyond its delineated verticals is affected by its lack of industry knowledge. Coughlan, 6/28/04, 3673:10-20; [REDACTED TEXT].

        5.4.1.11 It would not be a good use of Lawson's resources to compete with Oracle, PeopleSoft and SAP in every vertical or industry, because Lawson is smaller in size and Oracle, PeopleSoft and SAP are embedded in a lot more industries. Coughlan, 6/28/04, 3679:9-15.

        5.4.1.12 Lawson does not believe it would be a good strategy to compete against Oracle, PeopleSoft and SAP, but rather, it should stay focused on its verticals. Coughlan, 6/28/04, 03679:16-22.

        5.4.1.13 Lawson's win ratios are highest in its delineated verticals. Outside its delineated verticals, it is most likely Lawson will not meet the functionality requirements of customers. Coughlan, 6/28/04, 3720:20-3721:12.

        5.4.1.14 Outside its verticals, Lawson believes it only has the functional capabilities to successfully compete against Oracle and PeopleSoft in the mid-market. Coughlan, 6/28/04, 3671:5-17. [REDACTED TEXT].

        5.4.1.15 Lawson is not a significant pricing constraint for Oracle's high-function software. Elzinga, 6/18/04, Tr. 2177:3-2178:13; P3175.

        5.4.1.16 The bulk of Lawson's customer base is in companies with revenues between $100 million and $1 billion; "the industries where they've been most successful tend to not have giant companies in it." P3171, Ellison dep., 1/20/04, 218:8-219:10.

        5.4.1.17 Lawson is "not a viable competitor for the up-market." PeopleSoft "very, very infrequently" sees Lawson in the up-market. PeopleSoft has "seen them a handful of times in the last couple of years." Wilmington, 6/16/04, 1802:25 -1803:12; 1806:15-17.

        5.4.1.18 Lawson does not have the functionality to compete in the up-market, they "play best in the mid-market." Wilmington, 6/16/04, 1803:13 -1805:12; D5890, at 30-33.

        5.4.1.19 PeopleSoft's CTO perceives Lawson is a mid-market application vendor. Bergquist, 6/8/04, 399:18-23; 400:2-15; P2356, at 3.

        5.4.1.20 Oracle's paid fact witness Ken Harris testified that during his tenure as CIO at major companies including Nike and Gap, he did not consider Lawson as a potential vendor in any detail. Harris, 6/24/04, 3270:10-12

      5.4.2 Customers and the Big 5 implementation firms view Lawson as a mid-market firm not capable of satisfying high-function upmarket requirements.

        5.4.2.1 Greyhound's evaluation showed that Lawson offered much less functionality than Oracle. Lawson could only meet 87% of Greyhound's requirements while Oracle was a 96% match. Lawson's software would have required expensive customization's to meet Greyhound's needs. Lawson lacked sufficient pre-defined work flows and user- defined fields. P4038-R, at GDD 005187; Glover, 6/15/04, 1475:01-11; 1479:21-23; 1483:24-1486:04; 1516:20-1517:02.

        5.4.2.2 When AIMCO evaluated Lawson HR software, they concluded that there was functionality lacking, i.e., that "there were functions that were not going to work for us inside of the current Lawson or the new Lawson system" and AIMCO "would have to write those ourselves, manage them, and maintain them ourselves." Wesson, 6/14/04 1132:17-1134:02.

        5.4.2.3 AIMCO's legacy Lawson HR system lacks important functionality. Wesson, 6/14/04 1154:19-1155:5; 1185:16-1186:6.

        5.4.2.4 AIMCO's evaluation of HR software found that Lawson's ability to handle AIMCO's business rules was inadequate: AIMCO would have had to "have] [its] IT staff building ... other software, to fit in between where there's gaps." Wesson, 6/14/04 1136:14-1137:14.

        5.4.2.5 AIMCO's evaluation of HR software found that AIMCO's needs to categorize employees who work at different entities within the company "tends to give other systems [than Oracle or PeopleSoft] a real run." Wesson 6/14/04 1140:11-23.

        5.4.2.6 AIMCO's evaluation of HR software found that AIMCO's needs to deal with employee training would have required purchase of third-party software with Lawson, in contrast to Oracle and PeopleSoft. Wesson 6/14/04 1141:12-1142:4.

        5.4.2.7 In reviewing analysts' reports and talking with others about potential software providers, Lawson and its software do not generally come up in the areas of concern to Ford, i.e., scalability, complexity, and appropriate functionalities. Lawson does not therefore feature when Ford is considering alternative solutions. P3041, Patel dep., 6/3/04, 83:15-18, 83:20-84:1.

        255.4.2.8 [REDACTED TEXT]

        5.4.2.9 CH2M Hill did not seriously consider Lawson an option. Bullock, 6/7/2004, 210:5 - 210:8.

        5.4.2.10 Lawson's inability to adequately scale and lack of successful implementations in the communications industry led Cox Communications to eliminate the firm from serious consideration for its FMS. Hatfield, 6/7/04, 117:2-6.

        5.4.2.11 In the competition for Carlson Companies the customer quickly dropped Lawson as a prospect because they realized Lawson's work flow was limited, its self-service functionality was insufficient, its payroll functionality did not meet Carlson's requirements, and it was behind in technology." Bergquist, 6/8/04, 400:20-401:19; P2356 at 28.

        5.4.2.12 The state of North Dakota ranked Lawson last out of six vendors due to lack of functionality in the Lawson HR and FMS products. Wolfe, 06/16/04, 1545:13-22.

        5.4.2.13 Following a review of vendor RFI responses and on-site demonstrations, KMG's vendor selection committee concluded that Lawson lacked the non-U.S. functionality and reporting capabilities that Oracle and PeopleSoft had, and eliminated Lawson from the competition. P3062, Elliott dep., 5/20/04, 108:8-16, 109:13-16, 112:6-12.

        5.4.2.14 [REDACTED TEXT]

        5.4.2.15 [REDACTED TEXT]

        5.4.2.16 [REDACTED TEXT]

        5.4.2.17 [REDACTED TEXT]

        5.4.2.18 [REDACTED TEXT]

        5.4.2.19 [REDACTED TEXT]

        5.4.2.20 Deloitte's Oracle practice never runs into Lawson as a competitor. Dortenzo Dep., 5/5/04, 53:15-54:5.

        5.4.2.21 BearingPoint's Enterprise Solution Group has about 300 staff who are dedicated to Oracle and between 800 and 1000 total staff who have Oracle experience and training. Keating, 6/10/04, 880:17-881:2. BearingPoint's Enterprise Solution Group has between 200 and 300 staff who are dedicated to SAP and additional staff who have SAP training. Keating, 6/10/04, 881:3-6. BearingPoint's Enterprise Solution Group's PeopleSoft staff is a little smaller than its SAP staff. Keating, 6/10/04, 881:3-8

        5.4.2.22 BearingPoint had only 3 Lawson projects in 2003. Keating, 6/10/04, 883:7-8.

        5.4.2.23 Keating testified that BearingPoint's "client base has found that Oracle, SAP and PeopleSoft software best meets their needs, and they have predominantly chosen those software packages than Lawson." Keating, 6/10/04, 884:3-9.

        5.4.2.24 Keating explained that BearingPoint's clients have selected Oracle, SAP and PeopleSoft over Lawson because of their requirements for functions such as multi- currency, multi-calendaring and multi-languages; because they like to do business with large vendors with significant revenue and a large number of staff capable of providing service; and because want to deal with global companies who can provide continuous 24 by 7 support. Keating, 6/10/04, 884:10-22.

        5.4.2.25 Keating said he would need to know the details about any Fortune 2000 firm's use of Lawson to know whether or not Lawson was a good choice. Keating, 6/10/04, 885:20 to 886:4. Keating explained that BearingPoint's "experience has been, it's the vast minority that would pick Lawson or one of the non-Tier One packages." Keating, 6/10/04, 886:2-4.

        5.4.2.26 Keating believes most publicly-traded large companies would select the Tier One vendors to comply with their regulatory reporting requirements as well as for functional needs, support needs and the desire to do business with a large company. Keating testified that he would be surprised if a lot of "companies would choose a Lawson to try and meet those needs. I think it would be tough to get it implemented, I think it would be tough to maintain, and I think it would stretch Lawson to its capabilities to try and get that to work that way." Keating, 6/10/04, 886:5-16.

        5.4.2.27 Keating testified that BearingPoint focuses its Lawson practice on health care. Keating, 6/10/04, 989:11-20. Keating also testified that, although some of BearingPoint's health care clients that use Lawson may be sufficiently large to qualify as Tier One clients, they would not have functional requirements in terms of number of currencies, number of legal entities, and other complexities on the same scale as BearingPoint's other Tier One clients. Keating, 6/10/04, 989:3-10.

      5.4.3 Oracle, PeopleSoft and SAP view Lawson as a mid-market firm.

        5.4.3.1 Ellison believes Lawson lacks the size and R&D to compete across the board in a lot of different industries. P3171, Ellison dep., 1/20/04, 236:8-21.

        5.4.3.2 Ellison also believes vendors need critical mass (in terms of reference customers), as well as the R&D capability and sales and marketing resources to compete in multiple industries, that there are only a few firms in this group and Lawson is not one of them. P3171, Ellison dep., 1/20/04, 236:22-238:13.

        5.4.3.3 Oracle considered acquiring Lawson in April, 2003, in part because of its focus on the mid-market. P0035, at 36; P3033, Henley dep., 5/04/04, 97:08-97:22.

        5.4.3.4 Within the past two years, Oracle was interested in acquiring Lawson because Lawson had a strong mid-market footprint that would add to Oracle's portfolio. Lawson's mid-market presence was specifically mentioned by Safra Catz during Oracle's discussions with Lawson. Coughlan, 6/28/04, 03752:17-3753:1-7.

        5.4.3.5 Oracle's Mr. Fitzgerald does not recall one discount approval form for a recent state FMS or HRM procurement where Lawson was identified in the justification. P3370, Fitzgerald dep., 05/20/04, 246:22-248:3.

        5.4.3.6 PeopleSoft's "Competitive Update - HRMS" from February 2002, notes that Lawson's "Solution Weaknesses" include the "Basic HRMS Suite," "Limited Analytics," "Unproven work flow," "No Upgrade assistance," and is "less configurable." For basic HR functionality, Lawson offers "3 modules only (HR, Ben Admin &payroll)." Other "Lawson weaknesses" include "targets up market with mid-market functionality." On global functionality, Lawson "cannot handle one global instance." Peoplesoft in contrast, has "localized product in 20 countries and support[s] 11 languages." D5890, at 32-32.

        5.4.3.7 SAP defines Lawson as a mid-market company because Lawson's products do not have deep product functionality, language support or currency support. Knowles, 6/23/04, 2924:22 - 2925:7.

        5.4.3.8 According to an internal SAP document, "Lawson is thought of as a mid- market vendor." P0026 at SAP01190.

    5.5 Lawson Lacks Functional and Vendor Characteristics to Serve as a Competitive Restraint in the High Function Software Market.

      5.5.1 Professor Iansiti studied the products and capabilities of Lawson and concluded they did not compare to those of Oracle, Peoplesoft or SAP. Iansiti, 6/17/04, 2045:2-15

      5.5.2 According to SAP, the scalability of Lawson Software's software architecture is limited in the number of simultaneous users its systems can support and the speed at which it can process transactions through the system. Knowles dep., 12/3/2003, 103:20 - 106:8. Lawson does not focus on business process improvement, rather it focuses simply on automation. In contrast, SAP's high function software provides business process changes that optimize software investments. P0026 at SAP01193.

      5.5.3 Lawson's HRM and FMS products are very limited in their ability to capture organizational complexity and cannot put data in the context that would make them effective for a large enterprise to manage its business. Iansiti, 6/17/04, 2045:17-2046:4.

      5.5.4 While Lawson's FMS application has a means of accounting for the disposition of an asset, as any firm would require, it doesn't have the capability to automatically feed the fully automate the process, and while that omission is not important for smaller firms where the disposition of assets is relatively rare, in large firms it is a common event and they are looking for software that can fully automate the process. Bergquist, 6/8/04 301:5-14.

      5.5.5 Lawson is not scalable because they use the Common Gateway Interface (CGI). This is a standard way for a Web server to pass a Web user's request to an application program and to receive data back to forward to the user. Every time someone requests a Lawson form, another CGI program is loaded into memory on the Web server. As a result, the system gets bogged down very quickly. P2356 at PS-E046504.

      5.5.6 Lawson has a fixed code block in their GL (General ledger) application, preventing it from handling a complex chart of accounts. Large companies like to have product information, business unit, etc. built into the chart of accounts. Lawson's limited functionality in this area eliminates it from large financial services deals. P2356 at PS-E046510

      5.5.7 Lawson does not offer daily rate pay functionality. It is a manually intensive process to perform retroactive pay adjustments in the Lawson HR system. P2356 at PS-E046513.

      5.5.8 Lawson provides a "business component integrator" integration tool to develop interface programs to produce and manage interfaced files that move from one system to another. However, the tool is inefficient and clients with large and complex files and transaction volume have been instructed not to use it. D 8103, Johnson dep., 5/28/04, 108:12-110:5

      5.5.9 Lawson software is only able to capture very limited levels in the organization. It can only enable five levels in the FMS application and three levels in the HRM application. This limitation would make it so that the model cannot capture the data at more levels of the organization and link it to the place where it is actually to be used. Iansiti, 6/17/04, 2046:19-2047:16.

      5.5.10 Lawson software does not have the same capability to deal with inter-organizational transfers as Oracle, PeopleSoft or SAP. There is very little flexibility in the software because the user can only configure a single company and cannot configure an organization so that one company looks like a manufacturing facility and another one looks like a sales organization. The processes in the software cannot be changed to reflect how the unit of business really is organized. As a consequence, there is a very limited amount of information that can be shared. Iansiti, 6/17/04, 2050:6-18.

      5.5.11 Lawson software is limited in that it can only track a maximum of 10,000 transactions between Lawson companies and after reaching that number of transactions the software cannot capture them anymore. Iansiti, 6/17/04, 2050:19-2051:5.

      5.5.12 Lawson software is very inflexible in its ability to change the organization over time and once the software is set up the user is unable to change the hierarchy that has been created. Iansiti, 6/17/04, 2051:7-13.

      5.5.13 Lawson software is unable to transfer employees between divisions or parts of the organization. To do so it is necessary to essentially fire them from one organization and hire them to another, and the data about the employee from the previous division or part of the organization is lost. Iansiti, 6/17/04, 2051:6-17.

      5.5.14 If an organization attempted to overcome Lawson's consolidation limitations by using a consolidation product, such as that offered by Hyperion, to aggregate different instances or installations of Lawson products, Hyperion would allow the data to be consolidated, but a manager could not drill back down on the data to understand where it came from. Iansiti, 6/17/04, 2052:5-2053:11.

      5.5.15 Lawson's product is also not web deployable. Additionally, Lawson, has not built strong relationships with systems integrators like Accenture, Bearing Point, and Cap Gemini. These integrators have not invested to build practices around Lawson like they have around Peoplesoft, Oracle and SAP. Wilmington, 6/17/04, 1980:10-1983:4.

      5.5.16 Lawson is very limited in its configuration capability, it has very little flexibility to adjust the software to the business process of different types of business units within an organization. Iansiti, 6/17/04, 2046:19-2047:2.

      5.5.17 The implementation of Lawson's software is not critical to the functionality of the software, because Lawson's software is more of an out-of-the-box solution. Coughlan, 6/28/04, 03660:25-03661:1-5.

      5.5.18 Lawson's FMS software is attractive to its vertical retail customers because it is an out-of-the-box solution. Coughlan, 6/28/04, 03664:16-25-03665:1-25-03666:1-16.

      5.5.19 One of the reasons Manulife selected Lawson's FMS product is because it was seeking an out-of-the-box solution. Coughlan, 6/28/04, 03731:2-16; D7052.

      5.5.20 If Lawson's products do not have specific functions to support a customer's unique business processes, Lawson may walk away from the deal. Coughlan, 6/28/04, 3731:17-24.

      5.5.21 If Lawson's products' do not meet the specific functional requirements of a customer to support its business processes, the customer's options are to spend money to customize the software or change its business processes. Coughlan, 6/28/04, 3732:9-15.

      5.5.22 Lawson has "limited capability awareness." Coughlan, 6/28/04, 3674:18-20; [REDACTED TEXT].

      5.5.23 According to SAP, Lawson lacks sufficient vendor credibility. P3037, Knowles dep. 5/3/04, 86:4 - 87:16.

      5.5.24 Few of Lawson's health care customers have international operations. Coughlan, 6/28/04, 3663:22-25.

      5.5.25 Lawson does not have an international sales force for its health care vertical. Coughlan, 6/28/04, 3664:13-15.

      5.5.26 A March 2003 internal Lawson document indicates that Hyatt did not believe Lawson "accommodates the international marketplace" but that vendors other than Lawson offer software that is configurable to meet the demands of the international marketplace. Coughlan, 6/28/04, 3733:16-3736:2; P3301.

      5.5.27 Lawson's products do not support any Asian languages. Coughlan, 6/28/04, 3736:6-7.

      5.5.28 Lawson's products do not support Unicode. Coughlan, 6/28/04, 3736:18-19.

      5.5.29 Lawson products only have payroll capabilities in the U.K., United States, and Canada. Coughlan, 6/28/04, 736:20-22.

      5.5.30 Because of international taxation issues, it would be difficult for Lawson to develop an international payroll product. Coughlan, 6/28/04, 3737:24-3738:10.

      5.5.31 Lawson's 10Q for the quarter ending February 26, 2004 cautions investors that Lawson does "not have substantial international operations and may not be able to develop out international operations successfully." Coughlan, 6/28/04, 3738:16-3739:1; P3292 at 34.

      5.5.32 Somewhere between five and eight percent of Lawson's total licensing revenues are derived from international operations. Coughlan, 6/28/04, 3740:20-25.

      5.5.33 Lawson competed for the Medical Center business alone at Stanford. Lawson has strength in healthcare providers, not the healthcare industry in general. Healthcare providers do not have global requirements, and many times don't have large scalability issues. Wilmington, 6/17/04, 1870:7-16.

      5.5.34 Lawson is currently trading at 50% below its initial IPO price from two to three years ago. Lawson's initial IPO price was $14/share, and Lawson is currently trading at $7/share. Coughlan, 6/28/04, 3657:11-3658:4.

      5.5.35 Lawson showed a net operating loss last fiscal year. Coughlan, 6/28/04, 3658:10-12.

      5.5.36 Lawson has had two separate sets of staff reductions in FY03. Coughlan, 6/28/04, 3672:24-673:9.

      5.5.37 Layoffs and employee turnover is a Lawson weakness. P2932 at ORLITF0092652.

      5.5.38 30% of Lawson's customers drive 80% of its revenues, and 30-40% of Lawson's revenues are generated by its health care vertical. Coughlan, 6/28/04, 3674:3-17.

      5.5.39 Lawson has only added 300 new customers in the last three to four years. Coughlan, 6/28/04, 3680:9-11.

      5.5.40 Lawson is perceived as takeover target. P2932 at ORLITF0092652.

      5.5.41 Lawson's FY2004 research and development budget is only $64,888,000. Coughlan, 6/28/04, 3689:3-8; P3291.

      5.5.42 Lawson competes on a limited basis for mid-market customers and for divisions of larger entities in financial services, healthcare, professional service organizations, public sector and retail. Peoplesoft does not view Lawson as a competitive threat. Wilmington, 6/17/04, 1875:18-1876:14.

      5.5.43 Lawson did not attract a single PeopleSoft customer in a marketing campaign targeted towards PeopleSoft customers and confusion surrounding Oracle's takeover attempt. Coughlan, 6/28/04, 3680:12-25-03681:1-4.

      5.5.44 Retail customers turning to Lawson do not have particularly complex business requirements in the areas of financial management and human capital management. Usually they are a single business line, located either in a region or certainly within the U.S. Wilmington, 6/17/04, 1975:14 - 1976:5.

      5.5.45 Lawson does not have a statewide FMS implementation in any state government. Coughlan, 6/28/04, 3727:5-7.

      5.5.46 A representative of the State of Michigan provided favorable reference customer information regarding Lawson's HR product to Arizona in September of 2001. D5040, ARIZ 01326; D5041, ARIZ 01341; D8101, Boncoskey dep., 5/21/04 41:18- 42:4, 43:21-44:11 However, this customer reference was provided before Michigan had implemented Lawson's employee self service functionalities and before the State had experienced its problems with growing batch cycle times. P , Spiekerman dep., 5/28/04,41:5-20.

      5.5.47 The State of Arizona HR system has not made the Lawson employee self service functions available to all state employees and does not expect to until some time after October of 2004. D8101, Boncoskey dep., 5/21/04, 49:24 - 50:9.

      5.5.48 The State of Arizona HR system does not have any unions on the system and lacks the complexity that would be created by the existence of multiple unions or bargaining groups. D8101, Boncoskey dep., 5/21/04, 58:18 - 59:4.

      5.5.49 Lawson was not considered as a supplier for McDonald's global requirements for integrated core financial and human capital management software. Lawson was only the incumbent in providing payroll processing to McDonald's North America corporate employees. The finalists for McDonald's global requirements were Oracle and Peoplesoft, and Oracle won. Wilmington. 6/16/04, 1806:18 -1809:2. P4951, at PS-C372114, PS-C372119; Wilmington, 6/17/04, 1883:2 -1884:20, Wilmington, 6/17/04, 1977:3-13.

    5.6 Lawson Counts Several Large Organizations as Customers but a Number of Those Organizations Are Moving off Lawson Having Made the Decision to Acquire High Function Software from Oracle, Peoplesoft and SAP.

      5.6.1 Large and complex customers of Lawson may initially purchase some of Lawson's products, but stop purchasing additional applications from Lawson. Johnson and Johnson is such a customer. According to an internal Lawson document, Lawson "should take these actions as well as the inaction of our key large clients as a reliable reading of how well our HR application is meeting the demands of world class global organizations. In my estimation, the HR application is not meeting the needs well enough and we are losing our foothold in these accounts." Coughlan, 6/28/04, 03707:8-24; P3297.

      5.6.2 Johnson & Johnson is standardizing its operations on SAP, rolling out SAP products as of March 2003. P3299 at 1.

      5.6.3 Johnson & Johnson is in the process of rolling out SAP's R/3 product which includes Human Resources and Financial Management Software. P3298 at 1, P3299 at 1. York International is rolling out SAP's R/3 which includes Human Resources and Financial Management Software to meet its global business needs for standardized processes and systems. P3155 at 1.

      5.6.4 McDonalds has signaled its intent to acquire high function software from Oracle or PeopleSoft to replace its incumbent systems one of which is Lawson Software. P0886 at ORCL-EDOC-01139146.

      5.6.5 McDonalds Corporation selected Oracle in August of 2002 to replace McDonalds' legacy systems for Finance, Human Resources and Supply Chain on a worldwide basis. P4892 at ORCL-EDOC-01083523.

      5.6.6 Oracle won the McDonalds' account from PeopleSoft. P4893 at ORCL-EDOC-01087342.

      5.6.7 Mastercard was an Oracle Financials customer. P3295 at ORCL-EDOC-00783629.

      5.6.8 Mastercard was continuing the process of standardizing onto Oracle's E-Business Suite. P3295 at ORCL-EDOC-00783629.

      5.6.9 Just because a Lawson customer receives product upgrade shipments does not mean that the customer has signed a new license with Lawson, plans on implementing the product, or in fact, is continuing to use Lawson's product. Coughlan, 6/28/04, 3704:9-16; 3709:18-20.

    5.7 Lawson products have problems when used by large or complex customers

      5.7.1 The City of Dallas encountered significant problems with its Lawson HR product in the calculation of overtime pay for fire and police officers and, at one point, withheld over one million dollars in payments to Lawson until the problem was rectified. Some firefighters did not receive paychecks from the Lawson system and others received miscalculated paychecks for as little as 47 cents. Coughlan, 6/28/04, 3699:15-23; [REDACTED TEXT]

      5.7.2 On September 12, 1997, the State of Michigan conducted a bid procurement and selected Lawson to provide an HR implementation for the State. At the time, Lawson was selected because they were perceived to be the best "off the shelf product with the lowest price. D8103, Johnson dep. 5/28/04, 38:8-20. D5314 at MI 0003645.

      5.7.3 Because of the multiple systems, unions and benefit plans administered by the State, Michigan's payroll administration is particularly complex and this complexity has led to longer batch processing times on the Lawson HR system. P3356, Spiekerman dep., 5/28/04, 13:6-14:4.

      5.7.4 In an attempt to reduce the batch cycle times experienced when running payroll in Michigan, the consulting firm of EDS was retained by the State and concluded that a "major limiting factor" in Michigan's attempt to rectify the problem was that the Lawson environment to database interface was "not fully customizable". P3356, Spiekerman dep., 5/28/04, 27:9-19; P4712 at MI 0016823.

      5.7.5 In benchmark data provided by Lawson during the State of Michigan's bid procurement evaluation process, Lawson represented benchmark payroll processing times of eight to twelve hours. P4711 at MI 0015108; P 3356 Spiekerman dep. 5/28/04, 10:11-11:10. Currently, it takes approximately 65 hours for the State to run its batch processing payroll calculation on the Lawson system. P3356, Spiekerman dep., 5/28/04, 11:10-13. For some payroll cycles, the time has exceeded 80 hours. P3356 Spiekerman dep., 5/28/04, 23:21-24:5.

      5.7.6 In order to preserve the integrity of the batch process payroll calculation, the State of Michigan has had to disable state employee access to the employee self service functions during the time that batch payroll calculations are being run on the Lawson system. P3356, Spiekerman dep., 5/28/04, 20:7-21:8. Coughlan, 6/28/04, 3727:15-3728:1.

      5.7.7 When payroll is being run every other week, State of Michigan employees are not allowed to make self service changes on the Lawson system for items like tax withholding deductions, insurance benefits, electronic direct deposits and similar items. P3356, Spiekerman dep., 5/28/04, 21:15-22:21.

      5.7.8 When State of Michigan employees attempt to log onto the Lawson system and make changes to their benefits information during batch payroll processing, they are taken to an online screen that states as follows: "Sorry, HRMN (the Lawson system) is currently unavailable for updating while your next pay check is being processed. Please try back on Friday. All non-update tasks are still available for inquiry/viewing purposes." P4706; P3356, Spiekerman dep. 5/28/04, 21:15-22:3.

      5.7.9 Separate from the batch payroll processing issues, the State of Michigan also experienced scalability issues with Lawson's employee self service products. The Lawson self service products provided to Michigan were a "demonstrable problem" and the earnings statement self service program was one of the State's poorest performing products. P3356, Spiekerman dep., 5/28/04, 35:1-19, 38:14-19.

      5.7.10 Ultimately, the State of Michigan abandoned the Lawson self service earnings statement process in favor of an alternative process that the State wrote on its own which contains none of the original Lawson software code. P3356, Spiekerman dep., 5/28/04, 37:22-38:11.

      5.7.11 MasterCard International, a customer identified in Oracle's Answer as a purported Lawson customer, experienced serious problems with Lawson's HRM product. According to an October 2003 internal Lawson document, the problems were having a "widespread effect at MasterCard" and "caused the CEO's paycheck to be calculated incorrectly." MasterCard also complained about Lawson's lack of responsiveness to this problem. Coughlan, 6/28/04, 3696:13-3697:25.

      5.7.12 McGraw Hill is another customer identified in Oracle's Answer as a purported Lawson customer, and it too expressed dissatisfaction with Lawson's HRM product. According to an internal November 2002 Lawson document, "MGH is not the first large client to begin to evaluate an HR application after using Lawson for several years. Ceridian is doing the same." Coughlan, 6/28/04, 3706:13-3707:6; P3297.

      5.7.13 In its evaluation of Lawson's HRM product as compared to Oracle and PeopleSoft, McGraw-Hill determined that Lawson's HRM product could not support its "current and future needs and should no longer be considered a contender." Coughlan, 6/28/04, 3708:11-3709:2; P3297 at LAW 5702.

    5.8 B. Ssa/baan Is a Mid-market Firm and Not a Seller of High- function Software.

      5.8.1 SSA/Baan does not meet the vendor credibility requirement for sales to large and complex enterprises, and maintains a lot of old, traditional mainframe legacy products. Iansiti, 6/17/04, 2064:10:-2065:7.

      5.8.2 SSA/Baan's product has stagnated as the company has failed to keep pace with the new technologies in the market or invest in research and development. P3037, Knowles dep. 5/3/04, 141:20-142:8.

      5.8.3 Investment bankers acquired SSA after the company emerged from bankruptcy in 2000. P4278 at ORCL-EDOC-00522074.

      5.8.4 SSA has acquired more than 15 different applications over the last few years making it a "modern day Frankenstein." P4278 at ORCL-EDOC-00522074.

      5.8.5 SSA's strategy is to milk maintenance fees as opposed to investing in new product development. P4278 at ORCL-EDOC-005 22075.

      5.8.6 [REDACTED TEXT]

    5.9 Microsoft is a seller of software for the mid-market (and smaller firms), not high function software for large and complex enterprises.

      5.9.1 Microsoft's products are targeted for small and mid-sized businesses. P0839, at MSOPSUB000000004419.

      5.9.2 Microsoft's Great Plains software is a mid-market product. P2060, ORCL-EDOC-00046837.

      5.9.3 Microsoft products are aimed at the small business to medium-size business segments. It does not have the capacity to model or manage complex organizational structures. Iansiti, 6/17/04, 2055:23-2057:4.

      5.9.4 Microsoft's products are pre-configured mid-market "out-of-the-box" solutions that Oracle is attempting to emulate as it enters the mid-market. P2061, ORCL-BLOK-002-001095.

      5.9.5 A June 2003 report concerning Oracle's proposed takeover bid for PeopleSoft (contributed to by META Group analysts that Oracle witness Mr. Kutnick described as "among the best in the industry," Kutnick, 6/30/04, 4224:3-4225:5; 4227:9-4228:12), recognized that Microsoft serves only the mid-market or "lower end" of the applications market and thus would be "mostly unaffected" by an Oracle acquisition of PeopleSoft. P3306 at (unnumbered) 5; Kutnick, 6/30/04, 4229:14-17.

      5.9.6 A December 2003 META Group analysis of Microsoft's MBS business and its business applications, titled "Microsoft Business Solutions, Strategic Next Step or Misstep?," concluded that "Microsoft's foray into business applications has been targeted specifically at small and medium businesses, (SMBs, businesses with less than $1 billion in annual revenue)." P3314. Kutnick, 6/30/04, 4197: 10-23.

      5.9.7

      5.9.8 The MBS products are primarily sold through channel partners, rather than a direct Microsoft sales force. [REDACTED TEXT]; see also Iansiti, 2053:25-2054:21.

      5.9.9 Unlike Oracle, SAP and PeopleSoft, all of whom use direct sales forces to target the complex customers in the GSM space, Microsoft relies on channel partners as its sales force for almost [RED] of its revenues. [REDACTED TEXT]

      5.9.10 [REDACTED TEXT]

      5.9.11 Without a direct sales force, it is difficult to serve large complex customers and to 1112understand their needs. Iansiti, 6/17/04, 2055:8-19.

      5.9.12 Implementation of Microsoft's applications products is usually handled by partners and Microsoft Business Solutions does implementation only on a very limited basis. MBS has 6,000 partners but only 100 individuals in the MBS consulting organization. Burgum, 06/23/04, 2992:19-2993:11.

      5.9.13 MBS does not have the implementation and consulting capability to serve a large, complex enterprise, and has no intention of building such capability. Burgum, 06/23/04, 2995:03-18.

      5.9.14 Under Microsoft's taxonomy of customers, the mid-market is defined as businesses with 50-1,000 employees; the Corporate Account market consists of businesses with up to 5,000 employees; and the Global, Major and Strategic space ("GSM") accounts are businesses with above 5,000 employees. [REDACTED TEXT] Microsoft's target market is the core small and mid-size segments and, depending on the fit with Microsoft's capabilities, divisions of large organizations. Moreover, even though some divisions of large organizations might be a good fit, not all are. [REDACTED TEXT]; P0839, at MSOPSUB000000004412.

      5.9.15

      5.9.16 Professor Iansiti evaluated the FMS and HRM functionality of Microsoft's MBS products, and concluded that Microsoft does not have the capability to provide high function software to serve the needs of large, complex enterprises. Iansiti, 6/17/04, 2053:12-24.

      5.9.17 Microsoft acquired Great Plains in 2001. Great Plains developed software primarily for mid-market customers. Burgum, 06/23/04, 2972:19-2793:9.

      5.9.18 At the time it was acquired by Microsoft, Great Plains had two products: Great Plains and Solomon. Burgum, 06/23/04, 2976:5-10.

      5.9.19 The Solomon product was sold to small and mid-market customers. Burgum, 06/23/04, 2997:25-2998:6

      5.9.20 The Great Plains product is primarily mid-market, with some customers that fit the Corporate Account Space (1000-5000 employees). Burgum, 06/23/04, 2999:11-19.

      5.9.21 MBS will continue to improve the Great Plains product, but the improvements will not expand the Great Plains product's ability to serve large enterprise customers because Microsoft is not targeting that segment. Burgum, 06/23/04, 3001:17-3002:1.

      5.9.22 Microsoft did not have any business applications before acquiring Great Plains in 2001. Burgum, 06/23/04, 2975:18-2976:1.

      5.9.23 In 2002, Microsoft acquired Navision, a mirror company to Great Plains in that it developed business applications for mid-market companies. Burgum, 06/23/04, 2972:14-2973:11.

      5.9.24 At the time it was acquired by Microsoft, Navision had two products: Navision and Axapta. Burgum, 06/23/04, 2976:11-21.

      5.9.25 The Navision product is targeted at the lower and mid-market segments. Burgum, 06/23/04, 2998:07-201.

      5.9.26 The Axapta product is targeted at the upper mid-market, Corporate Account Space (1000-5000 employees), and divisions or independent subdivisions of large organizations. Burgum, 06/23/04, 3002:02-21.

      5.9.27 Microsoft's goal was to have only 10% of its MBS products sold in the Corporate Account Space. This fiscal year, it will end up 15% - 30% short of that goal. Burgum, 06/23/04, 3008:22-3009:12.

      5.9.28 [REDACTED TEXT]

      5.9.29 Axapta's ability to serve divisions of large organizations does not mean that it can serve the overall needs of large organizations due to a large functionality gap. Burgum, 06/23/04, 3005:16-3006:02; D7173R, Burgum dep., 05/13/04, 101:2-102:10.

      5.9.30 As Microsoft continues to enhance and improve the Axapta product, it does not intend to increase the functionality of the product to meet the needs of large enterprises. Burgum, 06/23/04, 3007:11-16.

      5.9.31 Although the Axapta product does have some international functionality, because it was developed in Europe, its international capabilities are much different from Oracle, PeopleSoft or SAP. Iansiti, 6/17/04, 2055:23-2057:4.

      5.9.32 Microsoft has Axapta customers in some countries, some of which are served through the use of international English. D7173R, Burgum dep., 5/13/04, 25:11-26:3.

      5.9.33 [REDACTED TEXT]

      5.9.34 Axapta is sold through partners and those partners often try to close the functionality gap in Axapta relating to additional industry or country capabilities. D7173R, Burgum dep., 5/13/04, 34:4-20

      5.9.35 [REDACTED TEXT]

      5.9.36 [REDACTED TEXT]

      5.9.37 [REDACTED TEXT]

      5.9.38 Examples of financial functionality that Axapta is lacking that would be required to support corporate "hubs" includes treasury management, hedging; core intercompany transactions to do financial postings between different companies in an organization, sophisticated foreign currency transactions, investment strategies and cash flow projections. P3254R, Pollie dep., 150:15-151:1; 251:6-252:22.

      5.9.39 Examples of human resources functionality that Axapta is lacking that would be required to support corporate "hubs" includes full compliance with EEO reporting, an integrated payroll product, and compliance with U.S. regulatory HRM reporting. P3254R, Pollie dep., 252:23-253:14.

      5.9.40 The four product lines acquired by MBS represents an embarrassment of riches for mid-market customers. Burgum, 06/23/04, 2996:20 - 2997:1-14.

      5.9.41 Project Green -- a Microsoft R&D product meant to create a single solution to replace Microsoft's mid-market products would not be ready for market until 2008, at the earliest. Green Version 1 will have less functionality than Microsoft's current products. Burgum, 06/23/04, 3056:9 - 3059:17.

      5.9.42 Microsoft's Project Green is aimed at rationalizing Microsoft's four products onto a new code base. That release is dependent upon the release of Longhorn, the next version of the Microsoft operating system. Iansiti, 6/17/04, 2058:25-2059:24.

      5.9.43 Project Green applications will not be able to serve the needs of complex enterprises that Oracle, PeopleSoft and SAP can serve today. Project Green will serve mid-tier customers, perhaps up to $1 billion in revenues. Iansiti, 6/17/04, 2061:11-24.

      5.9.44 Green Version 2 will pursue functionality to try to catch up with Microsoft's current products and is not expected to be released until 2010 or 2011. Burgum, 06/23/04, 3061:7 - 3062:6.

      5.9.45 Microsoft developed a CRM product in-house. Burgum, 06/23/04, 2976:24 - 2976:3.

      5.9.46 The five product lines (Great Plains, Solomon, Navision, Axapta, CRM) are separate product lines that are not an integrated suite, except that CRM is integrated with some version of the ERP lines. Burgum, 06/23/04, 2977:4-12.

      5.9.47 [REDACTED TEXT]

      5.9.48 Microsoft Business Solutions currently is not profitable. Burgum, 06/23/04, 2977:25 - 2978:1.

      5.9.49 Microsoft has never considered building a product or changing its product to enter the large enterprise space, due to the gap in functionality, gap in a direct sales force, and gap in consulting services. Burgum, 06/23/04, 3023:17 - 3024:2.

      5.9.50 Microsoft's ability to pour lots of money into Project Green will not compensate for the company's lack of vertical experience, its lack of knowledge about the business practices of individual industries, its lack of relationships, its lack of a dedicated sales force. Building such experience takes time. Iansiti, 6/17/04, 2061:25-2063:15.

      5.9.51 Microsoft has had business failures before, for example Microsoft Money, which complete against Intuit. Iansiti, 6/17/04, 2126:9-20.

      5.9.52 Even with product development work and business development work and a lot of money it would take Microsoft more than six years to build a product to enter the large enterprise market; regardless, Microsoft is not planning on "pouring a lot of money trying to enter into new markets." Burgum, 06/23/04, 3024:23-3025:25.

      5.9.53 Microsoft would not attempt to enter the large enterprise space because it runs contrary to its strength, which is low prices and high volume packaged software sold to millions rather than tens of thousands of customers. Burgum, 06/23/04, 3026:11 - 3027:11.

      5.9.54 Absent the considered-and-rejected combination with SAP, Microsoft will not pursue an ERP product for use by Fortune 500 companies. P2640, at MS-OPSUB 000681.

      5.9.55 Microsoft rejected the possibility of a merger with SAP because of the complexity in completing the transaction and the complexity of post-transaction integration, which it had already experienced with the acquisition of Navision and Great Plains. Burgum, 06/23/04, 3047:22 - 3048:13.

    5.10 Oracle, PeopleSoft and SAP View Microsoft as Solely a Small to Mid-market Competitor.

      5.10.1 Microsoft appears in Oracle's Discount Approval Forms only once, and the sale represented in this Discount Approval Form was for a mid-market product. Elzinga, 6/18/04, 2215:23-2216:15.

      5.10.2 According to Oracle, Microsoft's products are not functionally equivalent to Oracle's products because Microsoft's products were designed for simpler companies. P2391 at ORLITF0017608.

      5.10.3 Oracle mid-market marketing documents describe where some of Oracle's competitors in the mid-market stand, based on research Oracle accumulated from various sources, including analysts' reports and Oracle's own observations in the market. A pyramid diagram shows Oracle's perception of its primary competitors for mid-market manufacturing customers, including the extent of Microsoft's marketing of its Great Plains and Navision products in the low range of the upper mid-market ($250 million - $1 billion). E.g., P0923, at ORLITP0058141; P0924, at ORCL-EDOC-01156927; P0947, at ORLITF0069746; P3070, Prestipino Dep., 5/18/04, 40:7-24;.42:2 - 43:3; 44:5-13.

      5.10.4 Oracle judged Microsoft's Great Plains mid-market product, compared to Oracle's Special Edition mid-market product, as not very strong in the functionality it was providing for manufacturers and as not competing well. The Oracle, Microsoft, and one of the SAP mid-market products evaluated offered limited or no HR functionality. P3070, Prestipino Dep., 5/18/04, 58:18 - 62:17; P924, at ORCL-EDOC-01156928.

      5.10.5 A March, 2004, Oracle document used for internal briefings on Oracle's mid-market initiative notes Microsoft's weaknesses in the mid-market include product fragmentation from its acquisition of Great Plains and Navision, which would take 3-4 years to rationalize. P0930, at ORLITF0069460; P3070, Prestipino Dep., 5/18/04, 129:24 - 130:25.

      5.10.6 An Oracle analysis of competitors in the mid-market, titled "Winning Against Key Competitors", shows sales of Microsoft's Axapta product ranging from the lower mid- market up to about the $750 million level of the upper mid-market, sales of its Navision product ranging into about the $175 million level of the mid-market, and sales of its Great Plains product ranging up to about the $75 million level of the lower mid-market. P0932, at ORLITF0039776; P3070, Prestipino dep., 155:3 - 158:13.

      5.10.7 According to an Oracle document and related testimony by Oracle's Vice President for Marketing, Mid-Market, Frank Prestipino, Microsoft's Axapta product is focused on the mid-market and has largely the same product offering that Oracle's mid-market Special Edition product has. P3070, Prestipino Dep., 5/18/02, 162:2 - 163:10; P932, at ORLITF0039776, ORLITF0039779.

      5.10.8 According to an Oracle document assessing key mid-market competitors, Microsoft's Navision product is a collection of loosely coupled acquisitions and third-party modules that raise integration issues, scalability issues, limited security and difficulties in reporting. P0932, at ORLITF0039793.

      5.10.9 PeopleSoft's Executive Vice President of Sales testified that Microsoft is not a competitor in the up-market, and that Microsoft acquired products that address the lower end of the mid-market. Wilmington, 6/16/04, 1811:8-22.

      5.10.10 According to PeopleSoft, it has taken Microsoft an incredible amount of time to roll out a product that it announced it was going to release for the mid-market in late 2005 or 2006. This shows how long it takes to enter any of these markets. Wilmington, 6/17/04, 1902:1-25; 1909:7-13.

      5.10.11 In a Competitive Selling Guide, related to Microsoft Business Solutions, PeopleSoft notes that Microsoft is focused on the small to mid-size market, or firms with less than $300 million in revenue and less than 500 employees. It goes on to describe the Axapta product as being directed at small and mid-size companies and suggests that MBS's current architecture and lack of scalability are real weaknesses. P2355, at PS-E0042444.

      5.10.12 According to PeopleSoft's CTO, Microsoft Great Plains lacks the functionality, scalability, reliability, global coverage and industry coverage necessary to compete for up-market customers. Bergquist, 6/8/04, 305:9-15.

      5.10.13 According to PeopleSoft's CTO, Microsoft's Axapta product lacks the functionality of PeopleSoft's Enterprise products. Bergquist, 6/8/04, 317:20-23.

      5.10.14 According to SAP, Microsoft's business model to market its product through channel partners restricts its market to small enterprise with less complex requirements. P4492, at SAP-DOJ002499.

      5.10.15 According to SAP, Microsoft's channel partners lack the required technical resources to provide the extent of customization and support necessary to service the more sophisticated mid-market enterprises. P4492 at SAP-DOJ002499.

      5.10.16 SAP has not seen Microsoft make any attempt to sell to organizations with greater than $350 million in revenues. Knowles, 6/23/04, 2899:23-2900:13

      5.10.17 In SAP's view, Microsoft Business Solutions' strengths are for organizations with annual revenues from $50 million to $200 million. P4492 at SAP-DOJ002499.

    5.11 Customers Do Not Believe Microsoft Provides High Function Software

      5.11.1 [REDACTED TEXT]

      5.11.2 [REDACTED TEXT]

      5.11.3 In response to a SSNIP, Cox Communications would not consider switching to Great Plains. Hatfield, 6/7/04, 137:1-12.

      5.11.4 Microsoft Great Plains' inability to adequately scale and lack of successful implementations in the communications industry led Cox Communications to eliminate the firm from serious consideration for its FMS. Hatfield, 6/7/04, 117:2-6.

      5.11.5 PepsiAmericas does not believe that it could turn to Microsoft's Great Plains or Navision product because there would be concerns about those products' capacity, performance, scalability and support for a company the size of PepsiAmericas and with the volume of transactions that PepsiAmericas processes on a daily basis. Johnsen, 6/16/04, 1739:15-19.

      5.11.6 Microsoft's installation to replace SAP at Esselte is not going terribly well, will not be completed until 2008 and will require them to a number of independent, isolated organizations. Iansiti, 6/17/04, 2066:3-2067:10.

      5.11.7 Esselte is not implementing a single instance of the Axapta product that will run across different countries. Every country will require a different Axapta installation. The Axapta product has very weak functionality in linking separate instances of the product, which means that it will be very difficult to ascertain what is happening across the entire organization. Iansiti, 6/17/04, 2124:10-2125:16.

      5.11.8 The state of North Dakota ranked Microsoft's Great Plains product fifth out of six vendors and eliminated Microsoft as a competing vendor due to lack of functionality in the Microsoft Great Plains HR and FMS products. Wolfe, 06/16/04, 1545:23 - 1546:04.

      5.11.9 There was no doubt at all that the Microsoft Great Plains product could not meet the functionality required by the state of North Dakota. Wolfe, 06/16/04, 1557:05-08.

      5.11.10 [REDACTED TEXT]

      5.11.11 [REDACTED TEXT]

      5.11.12 [REDACTED TEXT]

    5.12 Microsoft Does Not Provide Customers with High-function Software.

      5.12.1 While Microsoft has several large customers with brand-name recognition, these customers do not require high-function software and instead have mid-market needs.

        5.12.1.1 Ingersoil-Rand uses Navision for its accounting system at its BobCat dealerships. These are small five-to-six user type of operations. P3254R, Pollie dep., 23:9-25.

        5.12.1.2 Bowne Global uses Axapta as its accounting system at its headquarters, and serves its operations in several countries around the world, but its needs only require a 35 - 40 person installation. P3254R, Pollie dep., 24:19-26:3.

        5.12.1.3 Deloitte & Touche has a number of Navision installations in Europe for some of their very small firms. P3254R, Pollie dep., 43:19-24.

        5.12.1.4 [RED.], which is an $8-9 billion dollar holding company, uses Great Plains at their corporate headquarters to integrate core financials at a high level, with only 10 - 15 users. [ RED ] is also considering using Solomon at one of its subsidiaries. [RED.] is organized as 1400 independent agenices, each with only one to three users. P3254R, Pollie dep., 67:20-69:14; 70:13-19; 71:8-19; 72:97-15; 172:14-173:15; 178:8-179:15. (NAMES ARE ALL UNDER SEAL.)

        5.12.1.5 Helmeric & Payne uses Axapta for its remote oil rig operations. P3254R, Pollie dep., 179:16-22; 181:14-182:9.

        5.12.2 Microsoft has also had discussions with several other large customers, but has not secured the sales.

        5.12.2.1 An MBS partner is attempting to sell Axapta to [REDACTED TEXT], for use in Mexico, Australia, and Honduras. P3254R, Pollie dep., 120:18-125:1; 126:8-127:5. But these would be "spoke" sales for remote locations with simple requirements, with at most around 35 users at each of the three locations. P3254R, Pollie dep., 132:18-133:2; 134:24-135:11; 136:17-137:7. The MBS partner has not yet made any sales to [REDACTED TEXT]. P3254R, Pollie dep., 137:6-9.

        5.12.2.2 An MBS partner is also attempting to sell Axapta to [RED] and to a division of [RED.], but neither of those sales has occurred. P3254R, Pollie dep., 142:11-143:23; 156:17-157:23. " [RED.] is an example of a division of a large corporation ... whereas we get further down from the division level to the subdivision level to the spoke level present opportunities for a mid market solution that is easy to implement and very simple for some of the remote distribution and locations. . . ."

      5.12.3 One of Microsoft's largest customers is Esselte, which is not yet operational.

        5.12.3.1 Financial investors took Esselte private several years ago after the company struggled financially. D7159, Spund dep., 5/25/04, 8:3-12.

        5.12.3.2 Shortly before leaving Esselte, Lani Spund testified that he implementation of Axapta has been "very rough." D7159, Spund dep., 5/25/04, 68:7-69:3.

        5.12.3.3 Esselte's implementation of Microsoft's Axapta product is already behind schedule and over-budget. D7159, Spund dep., 5/25/04, 62:12-19, 207:22-208:2.

        5.12.3.4 Esselte's financial issues have contributed to the problems it has experienced with the implementation of the Axapta product. D7159, Spund dep., 5/25/04, 68:7-69:3.

        5.12.3.5 Esselte has yet to implement or use Axapta's products at any of its businesses. D7159, Spund dep., 5/25/04, 61:16-62:11.

        5.12.3.6 Esselte does not anticipate completing the implementation of Axapta prior to 2008. D7159, Spund dep., 5/25/04, 62:12-19.

    5.13 AMS is a supplier of FMS software solely tailored for the public sector.

      5.13.1 AMS sells only to public sector customers because it was built for government. D7166, Morea dep., 5/7/04, 127:25 - 128:5; 128:20 - 129:4 It is a niche vendor, not important in the commercial sector. Elzinga 6/18/04 Tr. 2215:5-14.

      5.13.2 In fact, AMS has licensed SAP's financial management and human resource software for its own business, and will use PeopleSoft at CGI, the company with which it recently merged. D7166, Morea dep., 5/7/04, 184:19 - 185:17.

      5.13.3 AMS does not offer HR software to federal public sector customers, nor does AMS have plans to start offering such. D7166, Morea dep., 5/7/04, 15:5-7; 130:20 -131:3.

      5.13.4 Since early 2001, AMS has placed only three bids for ERP software to new customers (i.e. customers who were not already using AMS's financial management or human resource software). D7166, Morea dep., 5/7/04,; 158:1 - 158:22.

      5.13.5 And of these three, two were for Departments of Transportation, for which AMS has developed unique functionality in its software. D7166, Morea dep., 5/7/04, 160:20 - 161:20.

      5.13.6 AMS limits sales to its existing customer base because AMS enjoys certain advantages when bidding for those customers. D7166, Morea dep., 5/7/04, 169:22 to 171:6

      5.13.7 AMS has some government accounts. Phil Wilmington of PeopleSoft testified he does not see AMS very frequently, only from time-to-time, generally when they are the incumbent, and when they have a pre-existing relationship with the customer. Wilmington, 6/17/03, 1994:22-1995:21.

      5.13.8 AMS is not winning new customers, it is only upgrading existing customers. It cannot win in an open competition against a Commercially Off The Shelf (COTS) vendor such as SAP, Oracle or PeopleSoft. P4491 at SAP-DOJ002716

      5.13.9 The Department of Justice's recent purchase of AMS was an initial test license for a financial management solution that was built specifically for federal customers. D7166, Morea dep., 5/7/04, 135:14-136:4; 137:10-20.

      5.13.10 AMS does not have the revenue to invest in research and development to improve its product. P4491 at SAP-DOJ002715.

      5.13.11 Mr. Fitzgerald cannot recall a discount approval form seeking a 70% discount for an FMS or HRM opportunity where AMS is listed in the justification. P3370, Fitzgerald dep., 05/20/04, 246:22-248:3.

      5.13.12 AMS's weaknesses in competing with PeopleSoft include the fact that it isn't web-based; that its solutions require extensive customization; and that its TCO is high. Bergquist, 6/8/04, 393:1-13; 393:18-395:10; see also D5935.

      5.13.13 Bearing Point testified that AMS has an encumbrance accounting system which can only be used by government entities and other organizations that use encumbrance accounting. Keating, 6/10/04, 989:25-990:9. Publicly traded companies do not use encumbrance accounting or AMS. Keating, 6/10/04, 990:10-12.

      5.13.14 AMS provides very limited HR processing functionality. It does not provide comprehensive HR, benefits administration and payroll processing and the AMS HR functionality does not compare favorably with the Oracle, PeopleSoft and SAP HR systems. Keating, 6/10/04, 990:13-21.

    5.14 Hyperion's Software Is Complementary, Not Competitive, to Oracle

      5.14.1 Bearing Point testified that Hyperion is complementary product that is used with and not instead of Oracle, SAP and PeopleSoft's core ERP products. Keating, 6/10/04, 888:1-16. Keating testified that companies will have core ERP products for day-to-day transaction processing and Hyperion for consolidation. Hyperion is not "a viable option to do the day-to-day type of transaction processing. It's not geared for that." Keating, 6/10/04, 888:1-17.

      5.14.2 Bearing Point testified that core ERP involves systems used to process transactions in the day-to-day running of the business, such as the quote to cash cycle, the purchase to payable cycle, the general ledger, HR, benefits, and payroll. Keating, 6/10/04, 887:1-25.

      5.14.3 In contrast, Hyperion is a set of specialized consolidation products that many Tier One companies use to roll up the operating results of different divisions for budget formulation and reporting. Keating, 6/10/04, 887:1-25.

    5.15 IBM Is Not a Participant in the Market

      5.15.1 IBM does not make business applications software. P3193, Mills dep., 5/27/04, 24:9-22.

6 Market Concentration. The markets are highly concentrated and the merger would significantly increase concentration, with the merging parties commanding very high combined shares.

    6.1 Computing market shares and concentration statistics within each relevant market is a standard feature of Merger Guidelines analysis that Professor Elzinga has employed. Elzinga, 6/18/04, 2207:23-2208:10.

      6.1.1 The economic logic of calculating shares and concentration statistics is that customers are protected in the marketplace when they have a number of alternatives. If a merger, which eliminates one alternative, occurs in an already concentrated market, then the competitive loss from the merger may be significant. Elzinga, 6/18/04, 2208:11-18.

      6.1.2 In assessing market concentration, Professor Elzinga employs a standard measure known as the Herfindahl-Hirschman Index (HHI), obtained by squaring each participant's market share and summing the results together. A higher HHI indicates a more concentrated market. In the extreme, a monopolist has 100% share and so has maximal HHI of 100 x 100 = 10,000. Elzinga, 6/18/04, 2213:9-2214:4.

        6.1.2.1 In markets where the HHI exceeds 1,800, a merger, which would further increase concentration, may raise significant competitive concerns. Elzinga, 6/18/04, 2214:13-16.

    6.2 In estimating market shares and concentration statistics, Professor Elzinga has relied on non-public sales data from Oracle, PeopleSoft and third party vendors, obtained through compulsory process. Elzinga, 6/18/04, 2208:19-2209:5; 2209:21-2210:1; P3177; P3178.

      6.2.1 To these sales data, Professor Elzinga has applied a threshold of $500,000 in minimum annual purchases by a customer, so as to filter out likely mid-market sales or other small transactions such as follow-up sales or point solutions. Elzinga, 6/18/04, 2210:2-16; 2211:21-2212:5; P3177; P3178.

      6.2.2 The market shares and concentration statistics Professor Elzinga has calculated are not very sensitive to his choice of $500,000 as the threshold demarcating likely high- function sales. He also tried thresholds of $400,000 and $600,000, and found the resulting shares and concentration figures to be similar. Elzinga, 6/18/04, 2212:6-16.

      6.2.3 It would be inappropriate to assign market share to incumbent versions of high function software. See III.H, supra.

    6.3 For the U.S. high-function FMS software market in 2003, Professor Elzinga calculates market shares of 39% for SAP, 31% for PeopleSoft, and 17% for Oracle. Oracle and PeopleSoft thus have a combined share of nearly 50%. Elzinga, 6/18/04, 2209:14-20; 2212:17-2213:2; P3177.

      6.3.1 While AMS, SSA, and Microsoft products are mid-market products that are not substitutes for Oracle PeopleSoft, and SAP high function FMS software, see V.C, supra (SSA); V.D, supra (Microsoft); V.E, supra (AMS), it was not possible to get a clean data set that did not include sales by those firms. In this way, Professor Elzinga's conclusions on market concentration are actually conservative, because they include some sales of mid- market products by AMS, with less than 10% share; SSA with just over 2%; Microsoft with less than 2%; and SCT with a half percent. Elzinga, 6/18/04, 2212:17-2213:2; P3177.

        6.3.1.1 AMS is limited to the government sector; it is not an important vendor in the commercial enterprise sector. Elzinga, 6/18/04, 2215:5-14.

        6.3.1.2 Microsoft appears in the data because of a single FMS sale captured in the data because it trips the $500,000 threshold. Elzinga, 6/18/04, 2215:15-22.

          6.3.1.2.1 Professor Elzinga does not consider this sale to Esselte to be high-function, given his understanding of the nature of the purchase and the functional capabilities of Microsoft's FMS product. Elzinga, 6/18/04, 2215:23-2216:15.

      6.3.2 Professor Elzinga finds the pre-merger HHI in the U.S. high-function FMS software market to be about 2,800, and finds that the merger of Oracle and PeopleSoft would raise the HHI by about 1,000 points, to 3,800. Elzinga, 6/18/04, 2214:17-25; P3177.

        6.3.2.1 The U.S. high-function FMS software market is already highly concentrated, and the merger of Oracle and PeopleSoft would raise the concentration level significantly. Elzinga, 6/18/04, 2214:22-2215:4.

      6.3.3 The merger of Oracle and PeopleSoft would combine two of the three largest vendors, for all practical purposes creating a duopoly in the high-function FMS software market in the U.S. Elzinga, 6/18/04, 2217:1-9.

    6.4 Professor Elzinga calculated shares and concentration statistics for the U.S. high- function HRM software market in similar fashion to his treatment of the FMS market. Elzinga, 6/18/04, 2217:16-2218:5; P3178.

      6.4.1 Professor Elzinga finds that PeopleSoft's share in the U.S. high-function HRM software market is a little over 50%, SAP is next with close to 30%, and Oracle has a bit more than 18%. Oracle and PeopleSoft have a combined share of close to 70%.

        6.4.1.1 Again, because of limitations in the available data, Lawson and SCT were included in the calculations. Even adding them had little effect, because each fall below 1% share. Elzinga, 6/18/04, 2218:14-2219:05; P3178.

      6.4.2 The pre-merger HHI in the U.S. high-function HRM software market is high at around 3,800, and the merger of Oracle and PeopleSoft would raise the HHI significantly by about 1,900 points, to 5,700 post-merger. P3178.

      6.4.3 Customers currently have three choices in high-function HRM software. The combination of Oracle and PeopleSoft would represent a three-to-two merger in the U.S. high-function HRM software market. Elzinga, 6/18/04, 2220:12-22; P3178.

    6.5 Even if the appropriate geographic market were worldwide, the merger of Oracle and PeopleSoft would result in increased concentration in an already highly concentrated market.

      6.5.1 There would still be only three vendors of high function HRM and FMS software: Oracle, PeopleSoft and SAP. Iansiti, 6/17/04, 2025:2-13; Wesson, 6/14/04, 1151:21-25-1152:1; Hatfield, 6/7/04, 134:9-135:01, 141:04-21; Cichanowicz, 6/14/04, 1080:11-25.

      6.5.2 Although the relative sizes of SAP, Oracle and PeopleSoft would change in a worldwide market, Elzinga, 6/18/04, 2302:09-25, the pre-merger HHI in such a market would be highly concentrated. As a matter of pure mathematics, the HHI would be in excess of 2,800. (The minimum HHI would be 33.3 squared times three, or 3,333).

      6.5.3 Oracle and PeopleSoft each have a substantial percentage of enterprise sales on a worldwide basis, with shares of high function HRM and FMS software almost certainly much higher. Elzinga. 6/18/04, 2303:23-2305:17, D5572 at Table 1 (IDC reports following shares of worldwide revenues from ERP applications and maintenance: SAP 20.0%, Oracle 5.5%, and PeopleSoft 7.0%). Accordingly, a merger of Oracle and PeopleSoft would significantly increase concentration and would be likely to create anticompetitive effects in a worldwide high function HRM and FMS software market.

7 Anticompetitive Effects - Concentration and Likely Consumer Harm. The Merger of Oracle and Peoplesoft May Tend to Lessen Competition Substantially.

    7.1 The Proposed Merger Will Eliminate Head-to-head Competition Between Oracle and PeopleSoft And, from the Vantage Point of Many Customers, Will Constitute a "Three-to-Two" Merger: Reducing the Number of Meaningful Competitors from Three Independent Centers of Initiative to Two. Elzinga, 6/18/04 2158:22-2159:19; 2245:4-2246:4.

      7.1.1 Oracle's Own Statements, Documents, and Actions Show That PeopleSoft Is its Most Significant Competitor in Applications Sales.

      7.1.1.1 Mr. Ellison concedes that PeopleSoft is a "significant competitor" to Oracle. Ellison, 6/30/04, 4269:13-15.

      7.1.1.2 An Oracle report, from the quarter preceding the announcement of Oracle's tender offer, states: "PeopleSoft is our #1 Competitor" and "SAP is our #2 competitor" in application sales. P2093 at ORCL-EDOC-00042674. Block Dep., 12/16/03, 233:17-234:18.

      7.1.1.3 Oracle' s 10-Ks - before it announced the tender offer for PeopleSoft - candidly acknowledged competing primarily with PeopleSoft and SAP for HRM and FMS. After the tender offer was announced, Oracle abruptly expanded the description of "competition" in its next 10-K.

        7.1.1.3.1 Oracle's 2000 10-K filing states, in the section labeled "Competition," that "[i]n the business applications software market, competitors include J.D. Edwards, PeopleSoft Inc., and SAP Aktiengeschellschaft. The Company continues to compete in these traditional markets as well as in some new, rapidly expanding markets like the CRM, procurement and supply chain marketplaces where the competition includes Siebel Systems, Ariba, Inc., Commerce One and 12 Technologies." P2048, at 9.

        7.1.1.3.2 Oracle's 2001 10-K filing states, in the section labeled "Competition," that "[i]n the ERP business applications software market, competitors include J.D. Edwards, PeopleSoft, Inc. and SAP Aktiengesellschaft." P2049, at 8.

        7.1.1.3.3 Oracle's 2002 10-K filing states, in the section labeled "Competition," that "[i]n the applications software market, our primary competitors include SAP Aktiengesellschaft, Siebel Systems, Inc. and PeopleSoft, Inc. In the consulting and outsourcing markets, we compete against International Business Machines Corporation Global Services, Electronic Data Systems and Accenture Ltd., as well as other service providers." P2050, at 9.

        7.1.1.3.4 Oracle's 2003 10-K filing, filed after the announcement of Oracle's hostile bid for PeopleSoft, states, in the section labeled "Competition," that "[i]n the highly fragmented applications market, we compete against Microsoft, PeopleSoft, Inc., SAP Aktiengesellschaft Systeme, Siebel Systems, Inc. and many other application providers, as well as outsourced and in-house solutions for customers." P2051, at 10.

        7.1.1.3.5 Oracle's surveys for large ($2 billion+) customers show PeopleSoft (50%) and SAP (29%) are the most frequent responses to the question "Other than Oracle, please indicate which of the following vendors were considered." Elzinga, 6/18/04, 2188:10 - 2189:25; P3176.

      7.1.1.4 Oracle Internal Documents: Win/Loss Reports and Data Show Head-to-Head Competition with PeopleSoft

        7.1.1.4.1 A 2003 Oracle e-mail captioned "psft results" states that Oracle competes against PeopleSoft and wants to "drive a stake through their heart...." According to the e-mail, Oracle desires to "apply all their best resources to bear on any psft [PeopleSoft] deal." P2123, ORCL-EDOC-00042524.

        7.1.1.4.2 A 2003 Oracle e-mail captioned "psft results" states that Oracle closely monitored competition against PeopleSoft and was pleased to find in April 2003 that PeopleSoft experienced a 40 percent year on year decline. Oracle believed the results against PeopleSoft in North America were particularly positive for Q3 and Q4. P2123,ORCL-EDOC-00042524.

        7.1.1.4.3 According to an Oracle document titled "2003 Win/Loss Analysis NAS FY03 Ql," its North American Sales Organization registered 25 losses in Ql FY03. Of the 25 losses, it lost more accounts to PeopleSoft than any other competitor, 16 losses. Moreover, its loss percentage was greater in head-to-head competition with PeopleSoft than any other competitor. Oracle lost 36.8% of the time against PeopleSoft, and 14.7% of the time against SAP. P2090, ORCL-EDOC-00038647.

        7.1.1.4.4 According to a 2003 Oracle e-mail, PeopleSoft was mentioned as a competitor against Oracle 51 times during the Q4 of 2003 in deals where Oracle discounted by over 50 percent, whereas SAP was mentioned 30 times. Oracle won 21 of 51 from PeopleSoft and 15 of 30 from SAP. Of the times when Oracle did not win, the figure includes losses and deals not yet completed. P2095, ORCL-EDOC-00055322-23; Block Dep., 12/16/03, 248:5-249:7.

        7.1.1.4.5 According to a 2003 Oracle document titled "PeopleSoft HRMS v7 Attack Sales Campaign," in the 21-month period preceding March 2003, Oracle won accounts against PeopleSoft 75% of the time. P2120, ORCL-EDOC-00051212-13.

        7.1.1.4.6 According to a 2003 Oracle e-mail, captioned "Re:why we lose to psft," Oracle and PeopleSoft compete fiercely. "Every HR deal is a street fight, though, and we start out behind." P2122, ORCL-EDOC-00051218.

        7.1.1.4.7 According to a 2003 Oracle e-mail, captioned "Re:why we lose to psft," Oracle must improve its HRM product because PeopleSoft is the historic leader in the HRM high-function market and Oracle must overcome this advantage. The advantage comes from having a larger installed base and more HRM high-function references in a larger number of industry verticals. P2122, ORCL-EDOC-00051218-20.

        7.1.1.4.8 Mr. Block's concern about the competitive threat from PeopleSoft is evident in an e-mail, captioned "why we lose to psft," which he wrote in 2003 as well as in his deposition testimony. He was concerned about losses to PeopleSoft and wanted to "improve execution on the sales side" against PeopleSoft "ASAP." P2121, ORCL-EDOC-00039736; Block Dep., 1/05/04, 421:12-422:8.

        7.1.1.4.9 A 2003 Oracle e-mail indicated that after only one month of Oracle's Attack PeopleSoft campaign, Oracle believed that it was showing initial promise because there were twelve potential customers that might switch from PeopleSoft high-function HRM to Oracle high-function HRM. P2124, ORCL-EDOC-00042559

        7.1.1.4.10 In a 2003 Oracle e-mail captioned "Win/Loss Survey & OSO," Oracle highlighted its win/loss statistics against PeopleSoft, SAP, J.D. Edwards, and Siebel for special attention. Oracle notes that it won $5.7 million and lost $16 million against PeopleSoft in Q3 FY2003. By comparison, it won $1.8 million and lost $8.8 million against SAP, and won $3.3 million and lost $5.6 million against J.D. Edwards. The report notes that Oracle wins about fifty percent of the time against PeopleSoft, but PeopleSoft wins 3:1 based on revenue. P2132, ORCL-EDOC-00054305-06.

        7.1.1.4.11 Siebel is not one of Oracle's primary competitors. Block Dep., 1/05/04, 393:2-393:19.

        7.1.1.4.12 A 2003 Oracle e-mail states that Oracle does not want to lose on price to SAP or "especially" PeopleSoft. P2137, ORCL-EDOC-00043877.

        7.1.1.4.13 According to a 2003 Oracle document titled "Q3 WinLoss Prelim Results," PeopleSoft is Oracle's "#1 competitor where we lost 19 of 35" and "SAP is our #2 Competitor where we lost 10 of 19." P2093, at ORCL-EDOC-00042674; Block Dep., 12/16/03, 233:17-234:18.

        7.1.1.4.14 According to a 2003 Oracle e-mail and the testimony of Mr. Block, in August 2003 there were approximately 53 accounts "in play" between PeopleSoft and Oracle. Competition was so fierce that Oracle's CEO, Larry Ellison, and PeopleSoft's CEO Craig Conway were personally calling potential customers. P2138, ORCL- EDOC-000575787-89; Block Dep., 1/05/04, 485:11-48619.

        7.1.1.4.15 Oracle often competes with PeopleSoft in accounts outside the 240 strategic accounts and as of October 2002, it was losing to PeopleSoft more often than it did to SAP. Block Dep., 1/05/04, 390:13-391:17.

        7.1.1.4.16 In 2003, Mr. Block reported to Mr. Ellison that Oracle had twenty wins against PeopleSoft in a particular quarter because he believed the results were "very, very good." Mr. Ellison responded that the results were "Awesome." P2127, ORCL-EDOC-00043465; Block Dep., 1/05/04, 442:5-443:1.

        7.1.1.4.17 In federal government accounts, PeopleSoft is the vendor that Oracle most frequently tries to replace. P3059, Johnson Dep., 01/13/04, 133:18 to 134:5.

        7.1.1.4.18 Many Oracle customers also use PeopleSoft software. Oracle salespeople proactively pursue opportunities to displace PeopleSoft from those accounts. P3033, Henley Dep., 5/04/04, 146:12-147:09; 147:11-149:13.

        7.1.1.4.19 A 2002 Oracle e-mail captioned "Apps Reps and SC's Learn how to ATTACK PeopleSoft," states that competition against PeopleSoft "can be a battle, especially in a non-mfg [non-manufacturing] deal." Oracle set up a series of meetings and web-based conferences, and web-based training sessions, to teach its sales reps about PeopleSoft's strengths and weaknesses and sales tactics. Oracle wanted to "screw up [PeopleSoft's] year end!" P2113, ORCL-EDOC-00078678-79.

        7.1.1.4.20 Oracle has an HRM specialist whose job it is to follow what PeopleSoft is doing in the market, do competitive analyses, analyze losses, and determine what new features Oracle should add to its product to be competitive. P3171, Ellison Dep., 01/20/04, 84:21-85:18.

        7.1.1.4.21 In a 2/13/03 email from Renee Lorton, head of PeopleSoft's FMS product, Lorton notes the competition that exists between Oracle and PeopleSoft for FMS, and given Oracle's investment in FMS, PeopleSoft needs to be on the "offensive" and FMS needs to be a priority for PeopleSoft's development, investments and marketing strategy. P4926, at PS-C067284.

        7.1.1.4.22 [REDACTED TEXT]

      7.1.1.5 Oracle conducted a "Kill PeopleSoft" Campaign.

        7.1.1.5.1 Oracle had a program called "Kill PeopleSoft." Block Dep., 1/05/04, 375:2-15.

        7.1.1.5.2 According to a 2002 Oracle e-mail captioned "RE: Kill PeopleSoft project," Oracle placed special competitive emphasis on winning accounts from PeopleSoft through the Kill PeopleSoft program. In particular, senior staff monitors: 1) accounts it has displaced PeopleSoft from; 2) accounts it has competed against PeopleSoft in and has won; 3) accounts it has competed against PeopleSoft in and has lost within the last 18 months; and 4) "current opportunities where PeopleSoft is a major competitor." P2109, ORCL-EDOC-00085825-26.

        7.1.1.5.1.3 Oracle initiated a campaign called "Kill PeopleSoft." The plan was to discount aggressively to stop or slow PeopleSoft's licenses of product to the financial management industry. Oracle targeted new customers as well as PeopleSoft's client base. Oracle, PeopleSoft, and SAP were all successful in converting some of each other's customers. PeopleSoft responded to this competition by discounting its price, driving for a high level of customer satisfaction, offering a high level of service and product quality, and investing heavily over the last year to lower its customers' total cost of ownership. PeopleSoft is motivated to get good references for potential new customers and to protect its maintenance revenue stream which is important to any software company. Wilmington, 6/17/04, 1989:24-1992:24.

        7.1.1.5.1.4 In a 2/26/03 email to Renee Lorton, head of PeopleSoft's FMS product, Lorton is advised that Oracle has put together a "Kill PeopleSoft" campaign in the area of FMS, and that Oracle "has not been lost on the fact that we [PeopleSoft] have overtaken them [Oracle] in the Financials area." The author of the email notes that the campaign directed to PeopleSoft is the "highest form of flattery" but it also indicates that Oracle "is beginning to invest in Financials again." P4920, at PS-C007540-40.

      7.1.1.6 Oracle maintains a "PeopleSoft War Room"

        7.1.1.6.1 Oracle has a "War Room" that generates weekly "War Room Reports" dedicated to monitoring competition between Oracle and PeopleSoft. Oracle does not have a similar apparatus dedicated to other competitors and does not generate weekly competitive reports on other competitors. P3192, Block Dep., 1/05/04, 546:8-549:9.

        7.1.1.6.2 Oracle's PeopleSoft War Room produces frequent competitive reports. In Q1 FY04 as of August 25, 2003, Oracle won three and lost eleven accounts against PeopleSoft and was currently competing in 59 other deals. Oracle believed that it was likely to win 23 of the deals, but that it was too early to tell on the other 36. P2139, ORCL-EDOC-00057999.

        7.1.1.6.3 Oracle tracks PeopleSoft's pricing policy in great detail. Ellison, 6/30/04, 4274:19-24; P3325, ORCL-EDOC-00181141-53, at 00181143-45.

      7.1.1.7 Oracle' s Discount Approval Forms show substantial head-to-head competition with PeopleSoft that leads to big discounts for customers.

        7.1.1.7.1 In general, Mr. Block's policy as recently as March 2003 was to compete head-to-head on price to win because "that [is] the game." Block Dep., 1/5/04, 451:4-13.

        7.1.1.7.2 An Oracle sales representative requested a 78% discount to compete with PeopleSoft at Neiman Marcus. P1278, at ORCL-EDOC-228413-15, at OO228415.An Oracle sales representative observed that PeopleSoft (the incumbent) and Oracle were the only vendors that Merrill Lynch was considering for its new HR Suite and that "PSFT has come in to protect their HR install at Merill and undercut our price ...." P1176, at ORCL-EDOC-00403219-27, at 00403221.

        7.1.1.7.3 Ingram Micro had Oracle Financials and PeopleSoft HR. An Oracle salesperson working on the Ingram Micro account wrote that Oracle was "vulnerable to a PeopleSoft takeover" for the Financials account. P1015, at ORLITE0418242-44, at 0418243.

        7.1.1.7.4 Overture had problems implementing Oracle HR applications and began to look at PeopleSoft HR as an alternative. An Oracle salesperson wrote that Overture had "started going down a PeopleSoft path" which was a "big blow internally for [Oracle]." P1024, at ORLITE01247221-26, at 0124722.

        7.1.1.7.5 Kelly Services had not recognized its expected return on investment from Oracle product and was considering replacing it with PeopleSoft HR. Oracle also was concerned that Kelly Services would look to PeopleSoft for financials. P1017, ORLITE0281593-94, at 0281593.

        7.1.1.7.6 Competition with PeopleSoft resulted in Oracle giving Buffet's, Inc., a 70% discount. Lawson was mentioned as competition but only PeopleSoft posed enough of a competitive threat to merit a discount. P1044, ORLITE0278978-79, at 0278979.

        7.1.1.7.7 Oracle and PeopleSoft both were installed at Ameritrade. To try to replace PeopleSoft, Oracle increased its discount to 40% over the usual 33% discount given to Ameritrade. P1058, ORCL-EDOC-00303021-26, at 00303022.

        7.1.1.7.8 An Oracle sales representative requested a 70% discount for the PepsiCo, procurement to compete with PeopleSoft. The salesperson noted that Oracle could "limit PeopleSoft to the North American business only and corner them into HR only for North America" by defeating PeopleSoft in this bid. P1098, at ORLITE0085427-32, at 0085428.

        7.1.1.7.9 Oracle sales personnel requested an 80% discount for Extendicare in order to be competitive with PeopleSoft references and functionality. PeopleSoft was "very aggressive" in its pricing. P1153, ORCL-EDOC-00029029-36, at 00029030.

        7.1.1.7.10 The sales representative who requested a discount for Hallmark noted that PeopleSoft was "dropping price to buy the [Hallmark] business" and that it would be a "huge loss" for PeopleSoft to lose to Oracle. P1164, ORCL-EDOC-00197732-37, at 00197733.

        7.1.1.7.11 Oracle and PeopleSoft both offered discounts between 70 and 80% for the Jacobs Engineering HR account. P1166, at ORCL-EDOC-00351098-100, at 0035110.

        7.1.1.7.12 An Oracle sales representative requested an 80% discount to compete against PeopleSoft in Qualcomm's procurement. P1188, at ORCL-EDOC-320921-26, at 00320923.

        7.1.1.7.13 After a "very competitive situation" at Reader's Digest, Oracle replaced the existing PeopleSoft HR software with its own. P1189, at ORCL-EDOC-351542-44, at 00351544.

        7.1.1.7.14 Competition between Oracle and PeopleSoft at TAC Worldwide Companies resulted in a request by Oracle for a 60% discount after Oracle was selected due to PeopleSoft's "extremely aggressive pricing to [try to] overturn the decision." P1028, at ORLITE0275053 (0275052-56).

        7.1.1.7.15 Competition between Oracle and PeopleSoft at The Longaberger Company resulted in a request for a 91.4% discount. P1031, ORLITE0268326-27, at ORLITE0268326.

        7.1.1.7.16 Competition between Oracle and PeopleSoft at Hanover Compression Company caused Oracle to request a 66% discount. PeopleSoft was "extremely aggressive" in its pricing and reduced its proposal by 50%. P1032, ORLTTE 0276587-90, at 0276590.

        7.1.1.7.17 Competition with PeopleSoft at Hudson Highland Group, Inc. caused an Oracle sales representative to request a 65% discount. P1053, at ORCL-EDOC-00367583-84, at 00367584.

        7.1.1.7.18 Oracle and PeopleSoft both discounted between 50% and 70% to compete for the Macromedia, Inc. account. Oracle described its win over PeopleSoft "as one of the most significant wins in the software vertical in 3 years." P1126, 0125524-26, at ORLITE0125525.

        7.1.1.7.19 In the State of Ohio Statewide ERP procurement, Oracle sales staff requested a 95% discount and other concessions in a competition against PeopleSoft, SAP and AMS. P1943, at ORLITF 0144992.

        7.1.1.7.20 Oracle sales staff requested a 93.8% discount to compete against PeopleSoft in the second and final best and final offer stage for the State of North Dakota statewide ERP procurement. The sales staff wrote that they wanted a low price to make PeopleSoft re-think their strategy of using low price to dominate statewide ERP procurements. P1939, at ORLITE 0209200-9201.-00173101-102.

        7.1.1.7.21 A tabulation of Oracle' s Discount Approval Forms shows that PeopleSoft is the competitor that Oracle cites most frequently as the basis for a discount request. SAP, while a significant competitor, appears less frequently than PeopleSoft. Elzinga 6/18/04, 2176:9 - 2179:17, at 2177:3-11; P3175.

      7.1.1.8 Other Oracle Documents show intense competition with PeopleSoft.

        7.1.1.8.1 PeopleSoft's "greatest competition on the high end 'up market'" deals are Oracle and SAP. P3047, at ORLITF0017254.

        7.1.1.8.2 Oracle believes that PeopleSoft prices less than PeopleSoft's competitors. P3047 at ORLITF0017253.

        7.1.1.8.3 An Oracle document identifies Oracle as the number one applications competitor to SAP and PeopleSoft. P0139 at ORLITF0057636.

        7.1.1.8.4 PeopleSoft and Oracle are "very close competitors in some of the service- related industries like financial services and utilities and telecommunications." PeopleSoft has an advantage over Oracle in the staffing and consulting and manufacturing industries. Wilmington, 6/16/04, 1815:23 - 1816:8.

        7.1.1.8.5 Oracle's NAS Apps group maintains a Forecast Summary Report of sales prospects, and in the report for September 9, 2003, lists 18 accounts in which PeopleSoft is the "Key Competitor," and includes a column for "win probability" in which values are reported ranging from 10% to 90%. P4999.

        7.1.1.8.6 An Oracle document states that "Oracle and SAP remain the greatest competition on the high end "UP-Market," with Microsoft, smaller companies, and new entrants on the low end." P3325, at ORCL-EDOC-00181146.

        7.1.1.8.7 P0021 is an Oracle competitive analysis of PeopleSoft. The analysis was part of Oracle's PeopleSoft Competitive Program. P0021 was carried out at Ronald Wohl's initiative to "collect information that can be exploited to neutralize PeopleSoft." P0021, at ORCL-EDOC-00902997-00903032, at 00902999. The project's purpose was to arm the (a) field sales forces with the information they need to neutralize PeopleSoft in our sales opportunities; (b) development organizations with information they need to plan product development; and ©) marketing organizations with the information they need to drive our product messages and share information with analysts. P0021, at ORCL-EDOC-00902999.

      7.1.1.9 Oracle executives confirm that they engage in intense competition with PeopleSoft.

        7.1.1.9.1 Oracle engages in "hot" competition against PeopleSoft which is used to justify software license price discounts. Catz, 5/26/04, 3512:10-25.

        7.1.1.9.2 An Oracle sales executive directed his subordinate to attack and "bury" PeopleSoft-"the enemy"-(and not SAP) in a HRM vendor "shoot-out" in Washington, D.C. in March 2002. P3050, Render Dep., 5/11/04, 266:10-267:1; P2841, at ORLITF0050734.

        7.1.1.9.3 In a 2003 e-mail, Oracle informed its sales representatives of PeopleSoft's quarterly schedule so that they could react accordingly and "be as disruptive as possible to PeopleSoft's quarter close." P2140, ORCL-EDOC-00059812.

        7.1.1.9.4 Oracle targeted PeopleSoft customers when PeopleSoft was converting customers from PeopleSoft version 7 to version 8. Block Dep., 1/05/04, 404:16-21; 406:9-407:8. Oracle perceived a competitive opportunity to get current PeopleSoft customers to switch away to Oracle. It launched what it called the "PeopleSoft 7 Replacement Campaign" and succeeded in taking a number of customers away from PeopleSoft. Wohl, 6/25/04, 3410:13-3411: 9.

        7.1.1.9.5 Oracle has identified 300 of the largest accounts where the customer has PeopleSoft HRM and Oracle FMS. Oracle developed a plan to replace PeopleSoft at those sites with Oracle software. Block Dep., 1/05/04, 408:1-410:2; 412:16-413:1.

        7.1.1.9.6 As of August, 2003, Oracle was competing with PeopleSoft in 59 deals. Catz, 5/26/04, 3569:06-3570:02; 3574:16-24.

        7.1.1.9.7 Competition with PeopleSoft for the Hallmark Cards account caused the sales representative to request a 70% discount. Hallmark slightly favored Oracle, which had made a roughly $2 million bid. PeopleSoft recognized that it was at a disadvantage. According to senior Oracle executive Paul Ciandrini, PeopleSoft did "what they normally do, which is lowball the price to nowhere." Mr. Ciandrini testified that he and his colleagues discussed steps to gather information about PeopleSoft's bid, "so that, you know, we didn't have to discount any more than we necessarily needed to get the deal and leverage the positive differentiation that we thought Hallmark saw in our product over the PeopleSoft product." Mr. Ciandrini further testified that he and his colleagues thought PeopleSoft's bid probably was between $800 million and $1 million, yet Oracle won the account with a $1.2 million bid. Ciandrini Dep., 1/16/04, 202:5-15; 212:13-213:6.

      7.1.1.8 Head-to-head competition with PeopleSoft was a justification for Oracle to award a discount when it might not do the same if another vendor made a bid. Block Dep., 12/16/03, 257:13-259:1.

        7.1.1.9.9 Oracle competes vigorously with PeopleSoft and is willing to grant substantial and aggressive discounts to gain market share. Block Dep., 1/05/04, 431:12-19.

      7.1.1.10 Oracle and PeopleSoft Have Equivalent Functionality.

        7.1.1.10.1 Oracle undertook a focused effort to "catch up" to PeopleSoft in HRM because PeopleSoft had more HRM functionality and more HRM customers. It took Oracle 1-2 years to catch up. P3171, Ellison Dep., 01/20/04, 109:3-110:18.

        7.1.1.10.2 According to a 2003 email from Mr. Block to Mr. Ellison, Oracle and PeopleSoft compete fiercely for HR sales. "We [Oracle] have caught up to them [PeopleSoft] in functionality for the most part. Each sales cycle is a street fight but my people expect this going in." P4162 at ORCL-EDOC-00387651.

        7.1.1.10.3 Until recently, Oracle was "behind" PeopleSoft in the HRM software market. Oracle "played catchup" over several years; it now offers enough HRM functionality that it is "very competitive" with PeopleSoft. Jeff Henley, Oracle's CEO, is "convinced" that Oracle currently is "very competitive [with PeopleSoft] across the board." Oracle's software is "equal" to or better than PeopleSoft's offering in most areas of HRM. P3033, Henley Dep., 5/04/04, 223:03-228:08; 228:11-229:23; P2965, at ORLITF0027597-ORLITF0027599.

        7.1.1.10.4 Oracle and PeopleSoft offer essentially the same product to software customers. Safra Catz, one of Oracle's two presidents, testified that there is "massive overlap" in the product lines of the two companies. Catz, 6/25/04, 3536:23-3537:07.

        7.1.1.10.5 Oracle and PeopleSoft now offer essentially the same functionality. According to Jeff Henley, Oracle's CFO, the products "overlap." For that reason, Oracle does not intend to integrate the PeopleSoft products into Oracle software or resell the PeopleSoft products separately if the proposed merger succeeds. P3033, Henley Dep., 5/05/04, 349:22-351:17; P0053, at ORCI-EDOC-00030694; P3033, Henley Dep., 5/04/04, 133:02-135:04; P0036, at ORCJ-EDOC-00033343.

        7.1.1.10.6 A PeopleSoft document called "Competitors: Architectural Evolution," compares PeopleSoft's and Oracle's architectural innovations and their strengths and weaknesses. SAP is not included in this comparison. P4923, at C038732-41. Another PeopleSoft document, titled, "PeopleSoft 8 Pure Internet Architecture," also compares PeopleSoft's and Oracle's internet architecture, and their strengths and weaknesses. Again, SAP is not included. P4922, at PS-C038721-31

      7.1.2 Peoplesoft's Own Statements, Documents, and Actions Show That Oracle Is its Most Significant Competitor in Applications Sales.

        7.1.2.1 [REDACTED TEXT]

        7.1.2.2 That PeopleSoft and Oracle target each other can be seen in a variety of PeopleSoft documents:

          7.1.2.2.1 P4941 (PS-C201620-35) - "Oracle Versus PeopleSoft: Why Oracle is Better, An Oracle White Paper" (a detailed, point by point comparison by Oracle of their respective products);

          7.1.2.2.2 P4943 (PS-C257365-443) -- "Oracle Sales Tactics" (a primer on how PeopleSoft should address Oracle's sales efforts against it);

          7.1.2.2.3 P2538 (PS-E0029863-74)- "Oracle Apps World" (a comparison of Oracle and PeopleSoft's FMS products, including general ledger, accounts receivable, accounts payable, asset management);

          7.1.2.2.4 P4971 (PS-TE2266598-99) - PeopleSoft's "Oracle Attack Program" (a program directed at companies where PeopleSoft and Oracle both have footprints or where the Oracle customers have not yet upgraded to Oracle's 11i release); and

          7.1.2.2.5 P4939 (PS-C199914-22) - an in depth, side by side comparison of PeopleSoft and Oracle HRM offerings (including care HRM, Self-service applications, HRM helpdesk, North American Payroll, Recruiting Solutions.

        7.1.2.3 Intense price competition between PeopleSoft and Oracle is evident in PeopleSoft's Approval Matrix Request Forms, which are used to request discounts above certain levels. In a number of opportunities where the sales representative indicated that Oracle was the competitor (or, in one instance, Oracle/SAP) discounts up to 85% off the already discounted price were requested - and approved. See P4092 (PS-TS2000040- 41); P4094 (PS-TS2000076-77); P4089 (PS-TS2000236); P4090 (PS-TS2000381-82); P4091 (PS-TS2000385-86); P4098 (PS-TS2000803-04); P4099 (PS-TS2000805-06); P4300 (PS-TS2001707-08); P4301 (PS-TS2001904-05); P4302 (PS-TS2002325-27); P4303 (PS-TS2002397-99); P4307 (PS-TS2004165-67); P4308 (PS-TS2004318-21); P4311 (PS-TS2004924-26); P4313 (PS-TS2005076-80).

      7.1.3 The Direct Competition Between Oracle and Peoplesoft Drives Both Vendors to Innovate and Develop New Features and Functionality.

        7.1.3.1 Competition drives innovation. Without competition, innovation is "wasted effort." Ellison, 6/30/04, 4313:25-4314:11. Continued innovation is guaranteed only by market dynamics, not by contract. Ellison, 6/30/04, 4312:17-4313:4.

        7.1.3.2 Oracle has been investing more each year in business applications in order to build more competitive functionality and become as competitive in the marketplace in as many segments of the application market as possible. P3044, Wohl Dep., 4/29/04, at 135:19 to 136:3.

          7.1.3.2.1 Oracle has been increasing the number of developers every year so that it can improve existing software features and create new capability and "improve the software at the fastest possible rate." P3044, Wohl Dep., 4/29/04, at 157:20 to 158:11.

          7.1.3.2.2 Oracle has been looking at building new localized HR and/or payroll functionality in more countries in order to become more competitive in new segments. P3044, Wohl Dep., 4/29/04, at 137:4-18.

          7.1.3.2.3 In addition to the new country HR and payroll localizations, Oracle is planning to add other new features to its HR product: "We are always adding many, many new features to these products at as rapid a rate as possible." P3044, Wohl Dep., 4/29/04, at 140:16-22.

          7.1.3.2.4 Oracle is also working on new financial management functionality "all the time." "There are many, many new features." Wohl Dep., 4/29/04, at 171:15-19; 171:21.

          7.1.3.2.5 Oracle is already planning significant new functionality for its next generation financial products, which will have a changed architecture and run on and take advantage of Oracle's newest database. P3044, Wohl Dep., 4/29/04, at 205:16 to 207:1.

        7.1.3.3 Oracle competes by improving its applications products not simply to be as good as its competitors, but to be better than or "leapfrog" them. One driver of Oracle's innovation is the need to "beat whoever is best." For a long time PeopleSoft was perceived to be the best vendor of human resources management software. Wohl, 6/25/04; 3389: 21-3390: 20.

        7.1.3.4 Oracle, PeopleSoft and SAP spend hundreds of millions of dollars on R&D every year to engage in such leapfrogging. Elzinga, 6/18/04, 2350:18-2351:5. PeopleSoft's R&D budget is roughly $420 million; SAP's is approximately $1 billion. Bergquist, 6/8/04, 337:2-4; 337:19; 338:5-6. According to its 2003 10K filing, Oracle's total R&D expenditure for all product lines in 2003 was $1.18 billion. P2051.

        7.1.3.5 Historically, Oracle first offered financial management software while PeopleSoft initially offered human resources software. Oracle then introduced its own HR applications; it took some time for those applications to catch up to or exceed PeopleSoft's HR products. Similarly, PeopleSoft introduced its own financial management applications which, over time, have come close to catching up to those of Oracle. Wohl, 6/25/04, 3391: 4-3392:1; P2288 at ORCL-EDOC-00034164-165 ("in recent years Oracle has made significant strides in catching up to PeopleSoft HRMS capabilities, and in some cases has surpassed PeopleSoft."); Wohl, 6/25/04, 3406: 10- 3407: 1.

        7.1.3.6 Accenture's Christy Bass testified that there has been a history of "leapfrogging" in functionality between Oracle, PeopleSoft, and SAP. Bass, 6/16/04, 1689:14-1691:02.

        7.1.3.7 Oracle and PeopleSoft have continued to leapfrog each other in innovation and new developments in their HR applications. P2288 at ORCL-EDOC-00034166 (""While Oracle has a very comprehensive core HR capability (the functions used by the HR professionals) PeopleSoft probably still has an edge here in selected features such as workforce organizational planning, onboarding for new employees, employee performance management (this is a recent catch-up. We were ahead until PeopleSoft delivered their new ePerformance module earlier this year; however, we are engaged in functional uplift as we speak."); Wohl, 6/25/04, 3408:15-34010: 5.

        7.1.3.8 After PeopleSoft introduced its PeopleSoft 8 release, Oracle launched an initiative called the "PeopleSoft Competitive Program" designed to help Oracle compete better and, in part, to help decide what development efforts it needed to do in its applications to keep up with PeopleSoft. Wohl, 6/25/04, 3392: 2-18; P0021.

        7.1.3.9 An Oracle document describing the PeopleSoft Competitive Program described it as an initiative "to collect information that can be exploited to neutralize PeopleSoft." The purposes of the initiative were stated as including "focused competitive effort in development" and to "[a]rm the Development organizations with information they need to plan product development." P0021 at ORCL-EDOC-00902999.

        7.1.3.10 Oracle developed various applications products at least partly in response to PeopleSoft's own development efforts. e.g., P3265 at ORCL-EDOC-00039297; Wohl, 6/25/04, 3401:3-18.

        7.1.3.11 During the Kerr McGee selection process, both Oracle and PeopleSoft indicated to Kerr McGee that each was a frequent competitor of the other. Both vendors further emphasized to Kerr McGee the significant amount of money that each spent on research and development. P3062, Elliott Dep., 5/20/04, 177:14-178:9.

      7.1.4 Innovation Benefits Customers by Providing Them with New Features and Functionality. Oracle and Peoplesoft, and SAP, Respond to Competition Form Each Other by Improving Their Products.

        7.1.4.1 "Software can't stay static, you've got to improve it over time because if you don't, there will be better choices on the market and sooner or later, if the better choices are better enough, there's no reason for a customer to stay with your product if yours has become aged and no longer as capable as a different offering on the market." P3044, Wohl Dep., 4/29/04, 170:23 to 171:14.

        7.1.4.2 Oracle has recently added localization capabilities for its financial system in a number of eastern European countries, including Poland and Hungary, and many of Oracle's large U.S. customers, including Alcoa, have been interested in Oracle's localizations for its financial and HR systems in eastern Europe. P3044, Wohl Dep., 4/2/04, at 171:22 -172:7; 173:12-25.

        7.1.4.3 Oracle is always working to improve the capabilities of its products in the industries the products already serve and to expand its product's coverage to new industries. P3044, Wohl Dep., 4/29/04, at 214:4-7.

        7.1.4.4 The Oracle Finance department has developed Oracle Internal Controls Manager to assist customers in the documentation and audit compliance responsibilities imposed by Section 404 of Sarbanes-Oxley. P3044, Wohl Dep., 4/30/04, at 256:1 to 258:22.

        7.1.4.5 [REDACTED TEXT]

          7.1.4.5.1 [REDACTED TEXT]

        7.1.4.6 Oracle was the first of the three firms to provide internet capable software. After Oracle introduced internet enabled software, PeopleSoft and SAP both developed internet capability, and all three firms continued to try to improve their products. Keating, 6/10/04, 995:17 to 996:11.

        7.1.4.7 DaimlerChrysler chose PeopleSoft HRM over SAP because PeopleSoft was the only company at that time that offered true web-based architecture. SAP is now in the process of developing a true web-based architecture in response to competition in the marketplace and PeopleSoft's current offering. Gorriz, 06/15/04, 1379:20-1380:1; 1380:8-17.

        7.1.4.8 PepsiAmericas purchased its PeopleSoft HRM and FMS software contemplating continuing enhancements, which it has received to date. However, if Oracle discontinues major enhancements to the PeopleSoft software, the cost and complexity of PepsiAmericas software will increase. Johnsen, 6/16/04, 1734:20-1736:16; 1737:1-24.

        7.1.4.9 PeopleSoft competes regularly with Oracle, and often goes into a leapfrog game of making sure that whatever better features and functions Oracle has, PeopleSoft matches or bests them. Bergquist, 6/8/04, 339:20-340:8.

        7.1.4.10 At one point, Mr. Wohl sponsored an initiative at Oracle to collect information about PeopleSoft and its offerings in order to help Oracle compete better and decide what development efforts Oracle needed to undertake in its software to keep up with PeopleSoft. Wohl, 6/25/04, 3392:2-8.

        7.1.4.11 There are a number of features that Oracle has added because PeopleSoft introduced a feature and PeopleSoft customers and prospective customers found it attractive. Wohl, 6/25/04, 3378:10-19.

        7.1.4.12 According to a PeopleSoft document entitled "Oracle Apps World, Comparative Analysis":

          7.1.4.12.1 In Financials - Cash Management: PeopleSoft has the leading edge over Oracle in Cash Positioning, but lags Oracle in Cash Forecasting. Oracle caught up and leads PeopleSoft slightly in Bank Reconciliation, while PeopleSoft has the competitive advantage over Oracle in Bank Communication and in Deals. P2538 at PS-E0029865.

          7.1.4.12.2 In Financials - Accounts Receivable: PeopleSoft is not as robust as Oracle in eBill Payment, but plans enhancements. PeopleSoft has no planned response to Oracle's "Intelligent" Credit Decision Support, but notes "We need to revisit this decision." P2538 at PS-E0029866.

          7.1.4.12.3 In Financials - Accounts Payable: Oracle is catching up to PeopleSoft in E-Settlements, but lags in Current Payables Functionality, while PeopleSoft has the competitive advantage over Oracle in Financial Gateway Release 8.8. P2538 at PS-E0029867.

          7.1.4.12.4 In Financials - Project Contracts: PeopleSoft is behind Oracle in Enterprise Contracts, but has a future release planned. P2538 at PS-E0029870.

          7.1.4.12.5 In Human Resources Management software: PeopleSoft sees Oracle's offering in Succession Planning as relatively weak. P4939 at PS-C199914. Oracle plans enhancement to its Succession Planning by Q3 04. P2895 at ORLITF0091182.

          7.1.4.12.6 In Human Resources Management software - For Monitoring Absence: PeopleSoft says "No FMLA rules are delivered with Oracle." P4939 at PS-C199915. Oracle has a cross-team initiative in Leave & Absence Management, Phase 1 July 2005, which includes "More seeded plans (e.g. FMLA)." P2895 at ORLITF0091171.

          7.1.4.12.7 In Human Resources Management software: PeopleSoft sees Oracle's functionality in Labor Relations / Grievances / Harassment Tracking as "Pretty basic." P4939 at PS-C199915. Oracle has an initiative in Grievance and Complaint Tracking deliverable Q2 05. P2895 at ORLITF0091183.

          7.1.4.12.8 In Human Resources Management software - For PeopleSoft finds Oracle's Performance Reviews to be "nothing like PSFT ePerformance." P4939 at PS- C199915. Oracle has Performance Enhancements set for July '04 / Dec. '04. P2895 at ORLITF0091184.

          7.1.4.12.9 In Human Resources Management software - For Garnishments: PeopleSoft describes Oracle's Garnishments as "difficult to navigate." P4939 at PS-C199916. Oracle has Garnishment consolidation (US) set for July '04. P2895 at ORLITF0091185.

          7.1.4.12.10 In Human Resources Management software - For Recruiting: PeopleSoft says Oracle's "conceptual search engine is not as powerful as needed" in Recruiting Solutions. P4939 at PS-C199917. Oracle has a Concept Engine for iRecruitment slated for Q3 '05. P2895 at ORLITF0091187.

      7.1.5 Customers That Have Not Yet Purchased High-Function Software Also Benefit from Innovation Because Vendors Produce Functionality That Satisfies the Customers' Requirements at Lower Cost than Their Existing Systems.

        7.1.5.1 Oracle's new software functionality would enable customers to run better businesses at lower costs. P3044, Wohl Dep., 4/29/04, at 211:25 to 212:9; 212:11-13.

        7.1.5.2 For example, in the HR arena, the development team worked closely with the sales management teams to determine what development areas it would be worthwhile for Oracle to invest in. Oracle's new HR country localizations resulted from an indication from the sales management teams that Oracle would be able to sell more HR software if it developed this functionality. P3044, Wohl Dep., 4/29/04, at 142:22 to 143:7.

      7.1.6 Peoplesoft Competes with Oracle in the Innovation of Applications Products and Other Features. Bergquist, 6/8/04, 341:9-15.

        7.1.6.1 In 2002, PeopleSoft introduced AppConnect, a product that consists of a portal, integration layer, and other features that provide for integration across applications product lines. Bergquist, 6/8/04, 341:17-22. AppConnect lowers application customers' Total Cost of Ownership (TCO) by reducing the need for extensive custom development. P4918 at 3.

        7.1.6.2 After PeopleSoft announced the introduction of AppConnect, Oracle began focusing more on integration, which they had not done in prior years. Bergquist, 6/8/04, 341:17-25. Six months after PeopleSoft announced AppConnect, SAP announced its introduction of NetWeaver, which includes a portal, integration technologies, analytics and master data management. Bergquist, 6/8/04, 342:1-9. Oracle has introduced functionality to compete with AppConnect and NetWeaver, but does not market that functionality under a single brand name. P4918 at 3.

        7.1.6.3 PeopleSoft won the Nextel deal from Oracle in part because of a technology play involving AppConnect, which gives PeopleSoft the ability to integrate easier and faster than Oracle to third-party applications. P2594 at PS-TE-2131342.

        7.1.6.4 The higher the complexity and flexibility of the software, the higher the customer's total cost of ownership. P4954 at PS-E0042404.

        7.1.6.5 In September 2003, PeopleSoft introduced an initiative termed Total Ownership Experience (TOE). TOE lowers software implementation costs by automating the installation, configuration and integration of applications. P4918 at 4-5. SAP has recently announced that it will be introducing "Lifestyle Management Solutions" to compete with PeopleSoft's TOE. Oracle has focused on a "Total Cost of Ownership" initiative. P4918 at 4-5.

        7.1.6.6 Having three firms vigorously engaging in R&D and the innovation process is more likely to result in a stream of innovations than having two firms or one. Elzinga, 6/18/04, 2380:4-10.

    7.2 Customers Benefit from Competition Between Oracle and PeopleSoft Because it Leads to Innovation, Quality Service, and Lower Prices. The Proposed Merger Will Eliminate That Competition and Thus Directly Harm Customers.

      7.2.1 The merger will eliminate one of three, and in many cases one of two, options that customers have for high-function software.

        7.2.1.1 Nextel Communications

          7.2.1.1.1 Cichanowicz testified that Nextel's negotiation was a two stage process: First, identifying three viable alternatives (Oracle, SAP, and PeopleSoft); Second, negotiating with the two best alternatives (Oracle and PeopleSoft). Cichanowicz, 6/14/04, 1074:01-11.

          7.2.1.1.2 Cichanowicz testified that Nextel benefitted from having three viable competitive alternatives (Oracle, PeopleSoft, and SAP). Cichanowicz, 1104:01-22.

            7.2.1.1.1.3 Nextel found that a majority of their peer companies used Oracle, PeopleSoft, or SAP. P4075R, at NEXTEL-000035.

        7.2.1.2 Verizon Communications

          7.2.1.2.1 The only 3 vendors that can meet Verizon's functional needs are Oracle, PeopleSoft, and SAP. Verizon uses PeopleSoft. The merger would leave Verizon in the "sorry state" of having to go to Oracle for support, and then having "no option" but turning to SAP. Bradley, 6/9/04, 627:4-628:15.

        7.2.1.3 Neiman Marcus customers get better products and pricing when there are three competitors as opposed to two. In Neiman Marcus's case, it eliminated SAP from consideration as a potential vendor of high function software because SAP was not strong in retail and had a high cost of implementation based on prior experience. Since Oracle and PeopleSoft were Neiman Marcus's best options, if one drops out, there is no competition. Maxwell, 6/9/04, 690:21-691:14.

        7.2.1.4 PepsiAmericas

          7.2.1.4.1 PepsiAmericas is currently running an unsupported version of PeopleSoft. Facing a substantial upgrade expense in the very near future, the company has discussed - both internally and with an outside technology research firm - whether this might be the best time to make a move to standardize on a single global ERP platform. If PeopleSoft ceases to exist as an independent company, PepsiAmericas will lose one of its two preferred options for its global platform and one of only three vendors that could meet its requirements. Johnsen, 6/16/04, 1733:11-1734:19; 1738:12-1739:14; 1758:1-24.

        7.2.1.5 Cox Communications

          7.2.1.5.1 Cox Communications is an example of a company that had two good options, PeopleSoft and Oracle, for FMS during a recent procurement. If PeopleSoft were not a separate entity, Cox would have had only one good option. Hatfield, 6/7/04, 88:22-89:4.

        7.2.1.6 [REDACTED TEXT]

        7.2.1.7 Ford Motor Company

          7.2.1.7.1 If the proposed merger takes place, Ford's three options for vendors of enterprise software would be reduced to two. Less competition in this area would create issues for Ford concerning the speed by which its software would be improved, the manner by which upgrades would be driven, and the price that a reduced number of vendors would charge. P3041, Patel Dep., 6/22/04, 113:24-114:3, 114:6-12.

          7.2.1.7.2 The harm to software innovation, upgrades, and prices caused by a reduction in competitors from three to two would occur regardless of whether Ford owned and used PeopleSoft, Oracle or SAP software. P3041, Patel Dep., 6/22/04, 216:5-23, 25.

          7.2.1.7.3 Ford was advised by the Oracle CFO that, after the merger, Oracle would not be developing a next generation PeopleSoft software product. The impact of this on Ford is significant. Ford's current PeopleSoft software will be of little use after a period of time, causing harm to the company's businesses, efficiency and effectiveness. In addition, Ford would have to devote substantial costs, time and resources to move from PeopleSoft to an alternative software product. P3041, Patel Dep., 6/22/04, 117:19-21, 24-25, 118:10-14, 118:17-119:3.

        7.2.1.8 Greyhound Lines

          7.2.1.8.1 Greyhound determined that Oracle and PeopleSoft were its two best options for human resources management software. Glover, 6/15/04, 1495:22-23.

        7.2.1.9 Kerr McGee

          7.2.1.9.1 According to Richard Elliott, HR portfolio manager for information management and technology at Kerr McGee, the proposed merger would reduce competition, which would lead to reduced price and innovation pressure. In addition, Oracle's decision to halt the marketing of PeopleSoft software would impose hardship on Kerr McGee based on its use of PeopleSoft software. P3062, Elliott Dep., 5/20/04, 150:11-19, 172:24-173:5.

          7.2.1.9.2 Even assuming that SAP could meet Kerr McGee's enterprise resource planning needs for HR, the proposed merger would reduce competition among potential vendors from three to two, which is a significant decrease in the competitive environment. P3062, Elliott Dep., 5/20/04, 173:9-174:4, 181:8-12, 181:15-19.

        7.2.1.10 Metro-North Railroad

          7.2.1.10.1 Metro-North Railroad held demonstrations for two software vendors, Oracle and PeopleSoft. P4032, at NY-TP-043258; P3061, DeSimone Dep., 5/19/04, 156:4-21.

          7.2.1.10.2 Mr. DeSimone testified that Metro-North Railroad would have held demonstrations from a third vendor if a third vendor capable of meeting their requirements had responded to their Request for Proposal. P3061, DeSimone Dep., 5/19/04, 156:22-157:9.

      7.2.2 Direct Competition Between Oracle and Peoplesoft Leads to Lower Prices for Customers.

        7.2.2.1 PeopleSoft and SAP have a significant effect on Oracle's pricing but a "regular competitor" will have a zero effect. Hausman, 6/29/04, 4004:24-4006:3.

        7.2.2.2 Nextel Communications

          7.2.2.2.1 Richard Cichanowicz of Nextel Communications testified that Oracle and PeopleSoft knew they were competing against each other, because the vendors would see each other at software demonstrations. Cichanowicz, 6/14/04, 1073:14-1074:18.

        7.2.2.3 Kerr McGee

          7.2.2.3.1 Oracle and PeopleSoft each responded to a request for quotes as part of the Kerr McGee HRM vendor selection process, and, following further negotiations with Kerr McGee, both reduced their prices. PS's price was roughly 45 percent off its list price. P3062, Elliott Dep., 5/20/04, 122:1-8, 120:19-121:1.

          7.2.2.3.2 Oracle and PeopleSoft competed on vendor viability, vendor service, functionality, technology and cost in the Kerr McGee HRM selection process. Kerr McGee rated PeopleSoft higher on vendor viability and service, while Oracle was favored on technology. Oracle and PeopleSoft were very close on functionality; Oracle lacked certain HRM modules and German functionality that PeopleSoft had. PeopleSoft also had a better vision of what HRM could be. P3062, Elliott Dep., 5/20/04, 123:19-124:16; P0095R, at Kerr McGeeO18.

        7.2.2.4 DaimlerChrysler

          7.2.2.4.1 Even though Oracle was not one of DaimlerChrysler's finalists for the selection of HRM software, "there was a benefit of having three independent competitors in the market." The three companies "publish price lists" and stay "in competition with each other." Gorriz, 06/15/04, 1381:2-10.

        7.2.2.5 North Dakota

          7.2.2.5.1 North Dakota's price to purchase enterprise software dropped due to head-to- head competition between Oracle and PeopleSoft. Wolfe, 06/16/04, 1570:17 - 1571:03.

        7.2.2.6 AIMCO

          7.2.2.6.1 Oracle told AIMCO that it would match the 70 percent discount that AIMCO was offered by PeopleSoft. Wesson, 6/14/04, 1145:01-06; 1169:02-14.

      7.2.3 Customers Predict That They Will Face Higher License Fees Once One of Their Two or Three Competitive Alternatives Is Eliminated.

        7.2.3.1 DaimlerChrysler

          7.2.3.1.1 DaimlerChrysler is concerned about the proposed combination of Oracle and PeopleSoft because it would result in competition being limited to two players which could service its needs, and it is concerned about the impact on price levels and innovation. Gorriz, 06/15/04, 1390:1-8. Although Oracle has stated they would maintain the PeopleSoft HR software in its present release, Mr. Gorriz stated, "I doubt that they will further develop the software for a long time, and therefore, I think this would not be beneficial to us." Gorriz, 06/15/04, 1390:9-14.

          7.2.3.1.2 Verizon Communications

            7.2.3.1.2.1 Despite nearly reaching completion of a full conversion to PeopleSoft, Verizon has licensed SAP software with the expectation that HR/FM license fees will rise post-merger. Bradley, 6/9/04, 609:6-612:19 [REDACTED TEXT]

          7.2.3.1.3 Ford Motor Company

            7.2.3.1.3.1 When there are only two software vendors as opposed to three, the two can apply greater pricing pressures on Ford, knowing Ford does not have that much choice. P3041, Patel Dep., 6/22/04, 116:22-117:4.

            7.2.3.1.3.2 After the merger, Ford would have no choice but to pay a 10 percent price increase imposed on the maintenance fees charged for its HRMS and FMS software. It would also be required to pay a 10 percent increase in the price of new modules offered by the post-merger company. P3041, Patel Dep., 6/22/04, 128:25-129:14.

          7.2.3.1.4 Greyhound Lines

            7.2.3.1.4.1 Greyhound believes that the merger of Oracle and PeopleSoft will make it more expensive to acquire new human resources management software. Threatening to buy software from Lawson probably would not be an effective negotiating tactic. Glover, 6/15/04, 1495:13-1496:06

      7.2.4 Existing Customers Benefit from Competition among Peoplesoft, Oracle, and SAP for New Customers Because it Spurs the Firms to Keep Current Customers Satisfied.

      7.2.4.1 [REDACTED TEXT]

      7.2.5 Customers Fear That Oracle Will Lose its Incentive to Keep its Customers Satisfied Once its Most Significant Competitor Is Removed from the Market.

        7.2.5.1 Mr. Cichanowicz testified that Nextel will have to migrate over to Oracle, at tremendous cost, if Oracle does not upgrade the PeopleSoft product. Cichanowicz, 6/14/04, 1081:01-1082:19.

        7.2.5.2 "Reputational effects" may force Oracle to provide some type of alternative to PeopleSoft customers, but will not force Oracle to act in those customers' best interests. Johnsen, 6/16/04, 1756:8-23.

        7.2.5.3 An SAP document states the following: "Bottom Line, Q: As a PeopleSoft Customer/Prospect is there any benefit for you in this acquisition [Oracle's offer to buy PeopleSoft]? A: NONE". P4482 at SAP62529. An SAP document states the following: "PeopleSoft customers are the one hurt the most, with Oracle abruptly devaluing users and their investment in PeopleSoft applications. Oracle plans to end future enhancements to products and will force customers to incur switching costs." P4482, at SAP62528.

        7.2.5.4 Today, if Oracle fails to keep its current customers satisfied by maintaining high service and quality standards, those customers could take advantage of competition and replace their Oracle applications. Wohl, 6/25/04, 3389:4-9.

        7.2.5.5 Oracle is likely to eliminate personnel with specialized knowledge of "what [a customer] implemented, how [the customer] implemented it, why [the customer] implemented it, the issues [the customer] experienced, the things [the customer] had to do to make it run well" - "tribal knowledge" of the customer's operations. The customer would experience "a period of less efficiency." Hatfield, 6/7/04, 147:5 - 148;13.

        7.2.5.6 It is not certain that Oracle will have adequate incentives to provide adequate support to PeopleSoft customers if the merger is approved. Software vendors do not always act in the best interest of their customers. [REDACTED TEXT]

      7.2.6 The Merger of Oracle and Peoplesoft Likely Will Reduce Innovation in High- Function HRM and FMS Software, to the Detriment of Customers.

        7.2.6.1 Mr. Ellison has stated that "if there's no competition, innovation would be wasted effort." Ellison, 6/30/04, 4314:10-11.

        7.2.6.2 Oracle must compete on both price and capability. Purchasers of enterprise software are "not just buying a product, they're buying into a company who's going to constantly improve their product, provide related services, provide related products." The relationship is long-term because the systems last a decade. P3171, Ellison Dep., 01/20/04, 45:9-46:8.

        7.2.6.3 Customers make purchasing decisions over the long term, so that while the current state of Oracle's product is important, at least as important to customers is Oracle's ability to keep investing, the fact that Oracle is going to be around as a vendor, responding to technology changes and adding features. P3171, Ellison Dep., 01/20/04, 110:8-22.

        7.2.6.4 Innovation is important to customers because it allows them to run their businesses more efficiently and to make better decisions. Ellison, 6/30/04, 4312:17-24. Oracle's product enhancements have to keep track of changing regulatory and legal requirements. Wohl, 6/25/04, 3417:9-15.

        7.2.6.5 While enterprise software does not physically wear out, tax changes, regulatory changes, and changes in business needs require the existing system to be updated. Elzinga, 6/18/04, 2194:23-2195:10.

        7.2.6.6 In addition to the one-time license fee, customers pay annual maintenance fees to receive a steady stream of innovation and improvement in the software product. Wohl, 6/25/04, 3380:2-9; 3386:13-23.

        7.2.6.7 Competition is one of the driving factors in Oracle's ongoing innovation. Wohl, 6/25/04, 3377:5-6.

        7.2.6.8 Customers are not guaranteed a certain level of innovation in their maintenance contracts; innovation is guaranteed by market dynamics. Ellison, 6/30/04, 4312:25-4313:4.

        7.2.6.9 Vendors look at what other companies are offering, and where there are deficiencies they try to catch up, close the gap and gain competitive advantage. They try to innovate in certain areas to create features no one else has. P3171, Ellison Dep., 01/20/04, 110:22-111:7. They "leapfrog" over their rival's technology in having a better product and so move from being technically disadvantaged to being technically advantaged. P3171, Ellison Dep., 01/20/04, 114:21-115:9.

        7.2.6.10 Customers have benefitted from competition between Oracle and PeopleSoft. See section 7.1.3, 7.1.4, 7.1.5 above.

      7.2.7 Competition among Three Firms, All Vigorously Engaging in Research and Development, Is More Likely to Result in a Stream of Innovations than Competition Between Two Firms or One Firm Operating Alone. Elzinga, 6/18/04, 2380:4-10.

        7.2.7.1 Mr. Ellison stated that if there's no competition, innovation would be wasted effort." Ellison, 6/30/04, 4314:10-11.

        7.2.7.2 Customers fear that the proposed merger could reduce the competitive pressure that motivates Oracle and PeopleSoft to innovate and improve their products. PS has been involved in a large effort to create new modules, new functionality and an improved ownership experience. P3062, Elliott Dep., 5/20/04, 176:1-24.

      7.2.8 Oracle Is Unlikely to Continue Innovation of the Peoplesoft Products Because it Does Not Intend to Market Them. Catz, 5/26/04, 3540:12-3542:17.

        7.2.8.1 In Kerr McGee's experience, software vendors that cease to market a product typically reduce or eliminate new development initiatives on it, and, instead, focus their efforts on developing products that continue to be marketed. These vendors also focus on switching customers of the discontinued product to another product. P3062, Elliott Dep., 5/20/04, 178:10-179:11. For example, once Integral stopped marketing Kerr McGee's lnPower software, it soon after stopped development of the software and then stopped support. Elliott Dep., 5/20/04, 178:10-179:3. Kerr McGee's experience with its legal software product, E-Tech, was similar-the software is no longer being marketed, and now there is no more product development occurring. Elliott Dep., 5/20/04, 178:10-22, 179:3-11.

          7.2.8.1.1 Oracle's public commitment to support PeopleSoft customers for 10 years does not lessen Kerr McGee's concerns about the proposed merger. Support for problems that occur in the existing software is quite different from software innovation and development, and adding new features or modules to the software. Elliott Dep., 5/20/04, 176:25-177:13.

          7.2.8.1.2 Reduced competition caused by the proposed merger will diminish innovation, the development of new modules and new functionality, and the overall customer service environment. These effects relate not just to Kerr McGee's use of PeopleSoft software, but to future development in the HRM space. P3062, Elliott Dep., 5/20/04, 174:12-175:5.

        7.2.8.2 Software vendors are driven to innovate by a strong customer base and by a strong competitive environment. PeopleSoft's elimination as a competitor by the proposed merger may reduce innovation by shrinking the number of large software company competitors from three to two. Maxwell, 6/9/04, 692:7-24, 693:5-11.

          7.2.8.2.1 Mr. Gorriz of Daimler-Chrysler explained that when he needs an innovation, what he typically does as a customer is to "go to the software vendor and try to convince him that what you need is what the rest of the market needs. So, there is ongoing discussion ... If you are lucky.... you implement your needs in the standard product." Gorriz, 6/15/04, 1383:22-1384:11.

          7.2.8.2.2 Oracle's North America Strategic Accounts (NASA) Apps reports as its mission "To grow revenue in a select group of industry leading accounts by forming deeper relationships through a better understanding of their industry and their business challenges." The stated reason why the group exists is that "[t]hese customers drive application industry direction" and "[t]hey drive Oracle product development." Of the 240 customers, among those mentioned are GE Capital, Alcoa and Agere. P2893 at ORLITF0074301-02.

        7.2.8.3 Oracle's commitment to provide basic support for the PeopleSoft product for ten years does not resolve PepsiAmerica's concerns regarding its current PeopleSoft products because discontinuing enhancements would cause the company to seek out other software solutions - adding costs and complexities to the IT environment. A migration to another ERP platform could cost the company $30 million. Johnsen, 6/16/04, 1737:1-1738:6.

        7.2.8.4 Cox Communications currently enjoys the benefits of PeopleSoft's innovations and enhancements for its HRM software which are driven by competition. Cox believes that there will be less innovation if Oracle acquires PeopleSoft because there will be less competition. Hatfield, 6/7/04, 145:9-146:1.

        7.2.8.5 DaimlerChrysler is concerned about the proposed combination of Oracle and PeopleSoft because it would result in competition being limited to two players which could service its needs, and it is concerned about the impact on price levels and innovation. Gorriz, 06/15/04, 1390:1-8. Although Oracle has stated they would maintain the PeopleSoft HR software in its present release, Mr. Gorriz stated, "I doubt that they will further develop the software for a long time, and therefore, I think this would not be beneficial to us." Gorriz, 06/15/04, 1390:9-14.

        7.2.8.6 While customer demands and technology evolution fuel some innovation, vendors' desires to differentiate themselves from a functional perspective push innovation forward. Existing customers benefit from improved products as firms continuously compete to woo new customers. Johnsen, 6/16/04, 1740:12-1741:24.

        7.2.8.7 Competition, together with judgments made by vendor management and customer demands, all determine the direction of HRM and FMS software development. If there are fewer vendors, there will be a lot less innovation, and the innovation cycle will be driven by those two vendors as opposed to economic forces and customer demand in part. P3041, Patel Dep., 6/22/04, 115:8-116:3.

        7.2.8.8 Upgrades that take advantage of new technology or are required by a customer for a specific country are more likely to be provided if there are three as opposed to two software vendors in the market. P3041, Patel Dep., 6/22/04, 116:4-21.

        7.2.8.9 If the proposed merger occurs, and Oracle does not develop a next generation PeopleSoft software, Oracle would have no incentive to upgrade the existing PeopleSoft software that Ford is using. P3041, Patel Dep., 6/22/04, 120:5-8, 18.

    7.3 The Merger Is Likely to Produce Harmful Unilateral Effects.

      7.3.1 The Expert Testimony of Plaintiffs' and Defendant's Experts Support (Or Concede) the Appropriateness of Using a Unilateral Effects Theory in this Case.

        7.3.1.1 A merged firm in this market will have pricing authority or pricing discretion that Oracle and PeopleSoft, acting independently, do not have today. Elzinga, 6/18/04, 2159:10-19.

        7.3.1.2 Plaintiff's economic experts explained how this industry is likely to have harmful unilateral effects post merger. McAfee, 6/21/04, 2626:6-18; Elzinga, 6/18/04, 2265:15-2267:12.

        7.3.1.3 Dr. Hausman acknowledged the applicability of the unilateral effects theory, specifically that a merged firm can significantly raise its prices if the competitive constraint imposed by the other firm has been removed. Hausman, 6/29/04, 4041:23-4042:2.

        7.3.1.4 None of the Defendant's other economic experts effectively explained why unilateral effects should not be applied to this merger. Hausman, 6/29/04, 4041:23-4045:9 at 4042:10-21.

        7.3.1.5 As the parties' economic experts explained, with differentiated products, a merger of even two small firms in a market can have anticompetitive effects. McAfee, 6/21/04, 2626:6-18; Hausman, 6/29/04, 4042:10-21.

        7.3.1.6 The combination of Oracle and Peoplesoft would "have pricing authority or pricing discretion that independently the two firms do not have." Elzinga, 6/18/04, 2158:22-2159:19.

      7.3.2 Conditions That Enable Unilateral Anticompetitive Effects Exist in the Markets for High-Function HRM and FMS Software.

        7.3.2.1 Vendors of high-function HRM and FMS software have highly heterogeneous products.

          7.3.2.1.1 Oracle, SAP, and PeopleSoft perform many of the same functions, but they have different heritages based on the historic strengths of their respective products; there are architectural differences between the products and the products operate differently. Keating, 6/10/04, 897:23-898:7.

          7.3.2.1.2 Oracle, SAP and PeopleSoft have all had different strengths historically and that Oracle and SAP had taken the major market share in those industries in which their respective strengths were important to the industry. For example, Oracle had very rich financial functionality and was dominant in industries in which financial functionality was most important. SAP had very rich supply chain and manufacturing capability and had won a majority of the market share in industries in which those characteristics were important. PeopleSoft had the best human resources functionality and had the dominant market share in the market for human resource software. Keating, 6/10/04, 898:11-899:3; Hausman, 6/29/04, 3857:20-3858:20; 4018:2-16.

          7.3.2.1.3 Oracle took a major market share in the telecommunications industry because many telecommunications companies have billing systems that must be interfaced with a strong general ledger. Oracle also took a major share of the financial services industry because those customers had other financial systems that had to interface with the general ledger and other core financials. In contrast, SAP had a dominant position in industries that required sophisticated supply chain functionality, including aerospace and oil and gas. Keating, 6/10/04, 899:9 to 900:19. Some of the financial systems that financial services customers need to integrate with the general ledger include brokerage and lending systems. Keating, 6/10/04, 947:10-21.

          7.3.2.1.4 Mr. Keating testified that the differences that continue to exist between SAP and Oracle and PeopleSoft continue to provide a competitive advantage for Oracle and PeopleSoft for some customers. Keating, 6/10/04, 994:3-6.

          7.3.2.1.5 North Dakota eliminated SAP during the first round of its procurement process because SAP's price was too high and because SAP did not provide the necessary functionality. Wolfe, 06/16/04, 1546:13-24.

          7.3.2.1.6 AIMCO excluded SAP from consideration because configuration of SAP's software is more complex from that of Oracle and PeopleSoft and thus more expensive. Wesson, 6/14/04, 1182:15-1183:05.

          7.3.2.1.7 SAP's implementation at the engineering and construction firm Fluor was very difficult and costly. Fluor continues to have trouble with SAP's software and recently suspended making maintenance payments. P3150.

          7.3.2.1.8 Customers perceive SAP to have a higher total cost of ownership than either Oracle or PeopleSoft. Knowles, 6/23/04, 2950:24-2951:2; Wilmington., 6/16/094, 1816:9 - 1817:7.

          7.3.2.1.9 The implementation of SAP's software can be a lengthy and costly process for an organization because the software is higher priced and more complex than other solutions. P30036, Knowles Dep., 12/3/03, 152:11-152:19.

          7.3.2.1.10 Oracle and PeopleSoft position themselves as quicker and cheaper to implement than SAP. Knowles, 6/23/04, 2950:8-2950:16; Knowles Dep., 12/3/03, 153:19-154:18.It was only after years of focus on solutions for manufacturers that SAP extended it's the functionality to create a broader FMS and HRM offering. Keating, 6/10/04, 898:11-899:3, 901:6-20, 946:18-20; P3200, [REDACTED TEXT]

          7.3.2.1.11 SAP has long been the least flexible of the three vendors in the way it has sold its HRM and FMS products. Keating, 6/10/04, 901:6-20, 946:18-20; P3200, [REDACTED TEXT].

          7.3.2.1.12 SAP is less likely to discount than Oracle and PeopleSoft. [REDACTED TEXT]

          7.3.2.1.13 SAP's software has a reputation of being complex and expensive. Bullock, 6/7/2004, 209:13-209:16.

          7.3.2.1.14 P0017 provide the results of two 2003 surveys on the relative costs of Oracle and SAP applications software, including an independent study by Meta Group. Both studies concluded that Oracle applications were significantly less costly to license and install than SAP applications. The Meta survey found that SAP software fees were about five times higher than Oracle's and that SAP's combined license fee and implementation costs were almost seven times higher than Oracle's. P0017, at ORCL-EDOC-00140867.

          7.3.2.1.15 The second study reported in P0017, conducted by the research firm Current Analysis, found that Oracle's license and implementation fees were about 48% less than SAP's, and that Oracle's license fees were about 34% less than SAP license fees. Oracle's maintenance costs were 14% less SAP's maintenance. The study also found that 71% of Oracle's customers are web enabled, compared to only 13% for SAP. P0017, at ORCL-EDOC-00140869.

          7.3.2.1.16 Even though SAP has tried to be more flexible and modular, it has not closed the gap with Oracle and PeopleSoft in terms of modularity and openness. Keating, 6/10/04, 993:16-994:2.

          7.3.2.1.17 Recognizable differences exist among Oracle, PeopleSoft and SAP applications, with the functionality of some products fitting better in some organizations than for others. [REDACTED TEXT]
          [REDACTED TEXT]

          7.3.2.1.18 PeopleSoft started off in HR, and has been the market leader in HR management solutions, while Oracle initially developed strong financial management products. [REDACTED TEXT]; Keating, 6/10/04, 898:11-899:3; P3200, [REDACTED TEXT] . In fact, SAP had to "play catch-up" to try to reach equal functionality with PeopleSoft for its HR offering. [REDACTED TEXT]. Consequently, companies that purchased ERP solutions in the 1990s would have been more likely to license the combination of Oracle financials and PeopleSoft HR. [REDACTED TEXT] . There is a trend among companies that have different software for HRM and FMS to reduce the number of vendors whose software they run. Keating, 6/10/04, 930:20-931:11, 997:19-999:6. Because of that trend, Bearing Point has identified more than 1,200 companies that now have an Oracle Financials and PeopleSoft HR footprint. Keating, 6/10/04, 931:12-932:3.

          7.3.2.1.19 Because of Oracle's and PeopleSoft's different historical focuses, there are many customers that have PeopleSoft HR applications and Oracle financials. That fact gives Oracle an advantage in trying to sell its HR applications to customers who already have its financial applications. Wohl, 6/25/04, 3411:22-3412:17.

          7.3.2.1.20 SAP's strength has been serving the needs of manufacturing and consumer product goods companies. P3037, Knowles Dep., 5/3/04, 67:21 - 68:7.

          7.3.2.1.21 SAP's strength is in the manufacturing space; PeopleSoft's is in HR. Bergquist, 6/8/04, 319:21-320:6.

        7.3.2.2 Head-to-head competition between vendors of high-function HRM and FMS results in substantial discounts.

          7.3.2.2.1 Mr. Ellison testified that in procurements where Oracle faces a genuine competitor, it's a bidding process - Oracle tries to figure out how low it can bid to win this deal and still make a profit. P3171, Ellison Dep., 01/20/04, 293:20-295:2.

          7.3.2.2.2 See sections 7.1.1 and 7.2.1, above.

          7.3.2.2.3 Gap, Inc. received a higher discount on HR modules than on other products because PeopleSoft is competing with an HR upgrade: "Additionally we have given an additional larger discount on HR Applications to seed them with the modules to encourage the Gap to move to Oracle HR rather than Upgrade to PeopleSoft 8.0. By giving them these products we have put the Gap on par with what they already own from PS and have left future revenue through iRecruitment, iLearning, and Training Admin." P1011, at ORCL-EDOC-00359871 (00359863-76).

          7.3.2.2.4 In a competitive bid at Merrill Lynch for HR, an Oracle sales representative observed that "[t]his is for a Peoplesoft competitive replacement for the HR Suite and complete ERP commitment to Oracle. The price of PSFT migration to V8.8 is seen as expensive and considered a new implementation for Merrill. We have verified that this price point puts us in a competitive position..... Oracle is the only other vendor they are considering." Later, the representative came back with a new discount because "[a]fter discussions with Tony Kender and License sales management, we need to request an additional 3% discount for this PSFT competitive HR replacement..... PSFT has come in to protect their HR install at Merill and undercut our price....." P1176, at ORCL-EDOC-00403221 (00403219-27).

          7.3.2.2.5 Oracle offered Extendicare an 80% discount in order to be competitive with PeopleSoft references and functionality. "This is a highly competitive opportunity against PeopleSoft who are well positioned in this account. The largest opportunity area, from the customer perspective, is around HR/Pay and this is PeopleSoft's traditional domain. Of the three largest companies in the Long - Term Care business (Beverly, Mariner, Extendicare), 2 are already PeopleSoft users (Beverly, Mariner)... In order to compete against PeopleSoft we need to be at the ~80% discount level for this product offering. PeopleSoft has been very aggressive on their pricing. According to the customer the PeopleSoft price is US$1.4M and is an unlimited user license. Even if we are in the lead from a user point of view, if we do not beat PeopleSoft's price, it will be very difficult to get beyond the senior executive bias towards PeopleSoft." P1153, at ORCL-EDOC-00029030 (00029029-36).

          7.3.2.2.6 At Qualcomm, an Oracle sales rep requested an 80% discount, justifying the request by writing: "This is a very visible competitive evaluation versus Peoplesoft. Up until two years ago, Qualcomm was PeopleSoft's flagship reference account. Given that Psft is the incumbent, Oracle must be priced competitively in order to justify the additional purchase of license with a new vendor. PeopleSoft's license costs are much less than Oracle's." P1188, at ORCL-EDOC-00320923 (320921-26).

          7.3.2.2.7 The leader of PeopleSoft's Application Sales Team, Mr. Wilmington, testified that in June 2003, Target Corporation was in the market to buy human capital management software. It had narrowed its choice to PeopleSoft and Oracle. PeopleSoft and Oracle competed aggressively to meet Target's functional needs on price by discounting their fees. PeopleSoft continued to discount until it won. P4964, Wilmington, 6/16/04, 1796:2-1798:2; PS-TE2251423.

          7.3.2.2.8 Mr. Wilmington also testified that July 2003, PeopleSoft successfully converted a large division of Cardinal Health from Oracle's human capital management software to PeopleSoft's by aggressive pricing and by convincing Cardinal Health that PeopleSoft would better serve the large division's needs. Wilmington, 6/16/04, 1798:3 - 1799:5; P4970, at PS-TE2254709.

          7.3.2.2.9 Around November 2002, PeopleSoft successfully competed to replace an Oracle application at Nextel Communications. Nextel was considering either spending to upgrade to the next release of Oracle or choosing a competitive alternative. PeopleSoft won and replaced Oracle at Nextel in this situation. P4962, Wilmington, 6/16/04, 1799:19 - 1800:16, at PS-TE2251098-PS-TE2251100.

          7.3.2.2.10 Peoplesoft has analyzed Oracle's FMS product offering in P2538 (PS- E0029863-74)- "Oracle Apps World" (a comparison of Oracle and PeopleSoft's FMS products, including general ledger, accounts receivable, accounts payable, asset management).

          7.3.2.2.11 Peoplesoft has analyzed Oracle's HRM product offering in P4939 (PS-C199914-22) - an in depth, side by side comparison of PeopleSoft and Oracle HR offerings (including care HR, Self-service applications, HR help-desk, North American Payroll, Recruiting Solutions.

          7.3.2.2.12 Oracle considers SAP and PeopleSoft to be genuine competitors, as well as AMS in the public sector. In bids against those competitors, Oracle bids "up to the point of losing money." P3171, Ellison Dep., 01/20/04, 295:3-16.

          7.3.2.2.13 In the public sector procurement process, the vendors go through multiple rounds of bidding, "as the government works to get the best possible deal leading up to what's called the best and final phase," in which more than one company, and sometimes two or three or more companies are asked to give a best and final offer. P3171, Ellison Dep., 01/20/04, 292:17-293:19.

        7.3.2.3 For many customers, the products of Oracle and PeopleSoft are much better suited to their needs than the products of SAP.

          7.3.2.3.1 Emerson eliminated Peoplesoft from its vendor selection for a global ERP platform, in part, because Emerson is in the "discreet manufacturing" industry, an industry where SAP and Oracle enjoy better reputations. Peters, 6/14/04, 1248:4-25; 1263:1-9; 1262:14-20.

          7.3.2.3.2 SAP has struggled to penetrate organizations in the services sector. P3037, Knowles Dep., 5/3/04, 67:21 - 68:7.

          7.3.2.3.3 Oracle is PeopleSoft's closest competitor in the services sector. Bergquist, 6/8/04, 319:6-16.

          7.3.2.3.4 SAP has been working to enter the financial services industry in the United States for some time without much success. Knowles, 6/23/04, 2877:24 - 2878:9.

          7.3.2.3.5 SAP has unsuccessfully attempted for years to sell its software to the North American banking industry. Keating, 6/10/04, 993:3-10. SAP's recent announcement about its product modification plans is SAP's most recent attempt to make its software attractive to North American financial services customers. Keating, 6/10/04, 992:23-993:7.

          7.3.2.3.6 Mr. Keating testified that SAP has been more successful at selling its software to financial institutions in Europe than in North America. Banks in the United States must satisfy more requirements than European banks. In addition, the U.S. market for Tier One core ERP software in financial services is much bigger than the European market. Keating, 6/10/04, 996:20-997:15.

          7.3.2.3.7 SAP has a presence in less than 5% of the United States financial services providers market. Knowles, 6/23/04, 2941:6 - 2941:18.

          7.3.2.3.8 SAP had a presence at only five banks in the United States. Knowles, 6/23/04, 2940:23 - 2941:5.

          7.3.2.3.9 Accenture's Christy Bass testified that SAP currently has "zero" penetration in the top twenty banks in the United States, and that most of them use Oracle for FMS. Bass, 6/16/04, 1698:1-8.

          7.3.2.3.10 For many consumers in the banking business, the software offered by PeopleSoft and Oracle is the best possible option, while SAP's software is a poor option. Thomas, 6/9/04, 496: 18 - 22

          7.3.2.3.11 For many consumers in the investment management business, the software offered by PeopleSoft and Oracle is the best possible option, while SAP's software is a poor option. Thomas, 6/9/04, 497: 17 - 498:1.

          7.3.2.3.12 For many consumers with insurance businesses, the software offered by PeopleSoft and Oracle is the best possible option, while SAP's software is a poor option. Thomas, 6/9/04, 496: 11-17

          7.3.2.3.13 For many consumers in the utilities business, the software offered by PeopleSoft and Oracle is the best possible option, while SAP's software is a poor option. Thomas, 6/9/04, 497:6 - 9.

          7.3.2.3.14 SAP's partnership with Accenture has not led to any new sales to financial services organization. Knowles, 6/23/04, 2947:10 - 2947:18.

          7.3.2.3.15 Accenture partnered with Oracle to convince one of SAP's few banking clients to switch to Oracle. Knowles, 6/23/04, 2948:22-2949:23; P3152.

          7.3.2.3.16 Accenture has questioned SAP's ability to sell into financial service organizations. P3152

          7.3.2.3.17 A representative of Nextel testified that even if the prices offered by PeopleSoft and Oracle were 10 percent higher, Nextel would not have considered SAP because of the cost and conversion issues related to SAP. Cichanowicz, 6/14/04, 1080:05-10. Nextel was a company with a split Oracle/PeopleSoft footprint and chose to consolidate its ERP system on a PeopleSoft platform. Cichanowicz, 6/14/04, 1058:6-11.

          7.3.2.3.18 Charles Peters of Emerson explained that SAP was not a viable option for Emerson. SAP "didn't seem to have an interest in that type of pricing approach or format" that Emerson was pursuing. Peters, 6/14/04, 1244:4-1245:5 at 1244:18-24.

        7.3.2.4 The advantage of the products of Oracle and PeopleSoft over SAP for certain customers will persist.

          7.3.2.4.1 Dr. Elzinga explained that where three products in a differentiated market are close in product space it is not certain that rankings of those three products will remain static over time. Elzinga, 6/18/04, 2265: 15-2268:23 at 2268:10-23.

          7.3.2.4.2 However, as Dr. Elzinga went on to explain, SAP has already spent enormous sums of money to improve its product but is nevertheless continues to be perceived as the most rigid, least flexible of the three products. Elzinga, 6/18/04, 2370: 15-2373:4 at 2372:22-2373:4.

          7.3.2.4.3 Moreover, Dr. Elzinga explained that it was very unlikely that SAP could manage to "reposition" itself in such a way that it could become an appealing option for those many, many consumers who have already established oracle and Peoplesoft within their HRM and FMS footprints. Elzinga, 6/18/04, 2372:5-21.

          7.3.2.4.4 Dr. McAfee explained that there had been testimony about the big differences in the prices being offered in today's markets for HRM and FMS software, which meant that those markets were not close to being perfectly competitive markets today. McAfee, 6/21/04, 2486: 17 -2489: 20 at 2486:17-2487-12.

          7.3.2.4.5 Dr. McAfee explained that the ability to use SAP to obtain better prices exists in today's markets, yet there is still evidence of dramatic differences in prices .McAfee, 6/21/04, 2488: 15-2489:9.

          7.3.2.4.6 As Dr. McAfee explained, the ability to use SAP as a threat is something that exists today, and there are no reasons to expect that the ability to use SAP as a threat post-merger would be greater when SAP has become one of only two available options. McAfee, 6/21/04, 2489:10-2490:5 at 2489:10-20.

        7.3.2.5 The vendors of high-function HRM and FMS software engage in price discrimination.

          7.3.2.5.1 Dr. Elzinga explained that the sellers in this market price discriminate. Elzinga, 6/18/04, 2224:22-2225:9.

          7.3.2.5.2 All the predicates for price discrimination are met in the relevant markets. Dr. Elzinga explained that there were different demand intensities for the products being offered, supporting a finding of price discrimination. Elzinga, 6/18/04, 2224:22-2225:9.

          7.3.2.5.3 Dr. Elzinga explained that due to the long sales process, vendors were able to identify the demand intensities for the products being offered, supporting a finding of price discrimination. Elzinga, 6/18/04, 2224:22-2225:9.

          7.3.2.5.4 Dr. Elzinga explained there is no possibility of arbitrage, supporting a finding of price discrimination. Elzinga, 6/18/04, 2224:22-2225:9; 2341:7-22.

          7.3.2.5.5 The market and sales process in which licenses are sold in the market involves discriminatory list prices and discriminatory discounts.

            7.3.2.5.5.1 The purchase of the license for HRM and FMS applications occurs at the end of a time-consuming sales cycle. P 4014-B.

            7.3.2.5.5.2 Price is typically not discussed until after a customer is satisfied a particular vendor can meet their minimum requirements and vendors do not provide detailed pricing until after they understand the customer's specific requirements. P3036, Knowles Dep., 12/3/03, 185:1 - 185:18.

            7.3.2.5.5.3 At that point in time, a customer will negotiate with two to three vendors on price. Knowles, 6/23/04, 2848:25 - 2849:6.

            7.3.2.5.5.4 There is substantial evidence that the sellers of high function HRM and FMS software price discriminate, that is, they base their prices not on their own costs, but on characteristics of the buyer. McAfee, 6/21/2004, 2460:4-11.

          7.3.2.5.6 The list prices the firms charge are undeniable evidence of price discrimination.

            7.3.2.5.6.1 Oracle's pricing is done by algorithm, based on the customer's characteristics and not the cost of the software. The algorithm typically starts with the number of computers that will have the software and be adjusted based on the number of employees. However, the pricing for an individual customer may be only based on the number of computers, or only based on the number of employees, or based on both. Block Dep., 12/16/03, 141:8-142:10.

            7.3.2.5.6.2 PeopleSoft uses two different pricing metrics that are not correlated to cost. PeopleSoft charges either based on the number of employees or the customer's annual revenue. P2166, ORCL-EDOC-00051149, PeopleSoft, 2/18/03.

            7.3.2.5.6.3 Because Oracle prices on a per seat basis, it is clear that they are price discriminating, conditioning pricing on characteristics of the buyer and not basing prices on their own costs. Elzinga, 6/18/04, 2221:18-2226:23 at 2221:18-2222:8 ; McAfee, 6/21/04, 2520:4-2522:11.

            7.3.2.5.6.4 Dr. Elzinga explained how Oracle's pricing scheme within these markets led to Oracle charging different price to customers, prices that were not based on Oracle's costs. Elzinga, 6/18/04, 2222:24-2225:9.

            7.3.2.5.6.5 Dr. McAfee explained how this per seat licensing led to Oracle charging different price to customers, prices that were not based on Oracle's costs. McAfee, 6/21/04, 2520:4-2522:11.

            7.3.2.5.6.6 Dr. Hausman was forced to admit that based on per-seat pricing, a larger customer would be paying more even though they are getting the same disks and the same code from Oracle. Hausman, 6/29/04, 4056:23-4059:17.

            7.3.2.5.6.7 Dr. Teece could not effectively explain how different customers were not paying different prices and "receiv[ing] physically the same thing" was not price discrimination. Teece, 7/01/04, 4495:10-4496:12.

            7.3.2.5.7 The discounts offered by the firms are additional evidence of price discrimination.

            7.3.2.5.7.1 Dr. Elzinga testified to discounts ranging from 20 to over 90 percent, and noted such discounts could not be explained by differences in cost. Elzinga, 6/18/04, 2222:9-19.

            7.3.2.5.7.2 Dr. McAfee also noted there was a broad range of different discounts offered to different customers by Oracle. McAfee, 6/21/04, 2486:20-2487:4 and McAfee, 6/21/04, 2489:10-20.

            7.3.2.5.7.3 [REDACTED TEXT]

            7.3.2.5.8 The end result is prices that are set individually and reflect a customer's ability to employ competition in the bidding.

              7.3.2.5.8.1 Oracle's own internal documents demonstrated that Oracle routinely based its pricing decisions on characteristics of the buyer, not on its own costs. P1014; P1013; P1006.

              7.3.2.5.8.2 Dr. McAfee explained that Oracle was basing its pricing on a variety of buyer characteristics, including the type of employees a customer might have (P1014, McAfee, 6/21/04, 2466:4-20); economic conditions in the customer's industry (P1013, McAfee, 6/21/04, 2467:14-25); and the customer's negative experiences with Oracle in the past (P1006, McAfee, 6/21/04, 2472:6-18).

              7.3.2.5.8.3 Price variation among customers is not related to the vendor's costs, since Oracle is selling essentially the same software. McAfee, 6/21/04 2520:4-2522:11, 2521:3-10; Hausman, 6/29/04, 4032:1-12.

              7.3.2.5.8.4 As Dr. McAfee's trial testimony demonstrated, Oracle routinely conditions its prices based on competition with PeopleSoft, not it's own costs:

                7.3.2.5.8.4.1 For the Ebusiness Suite, including HRM and FMS modules at GAF, lowering its prices to "counter" PeopleSoft's bids. P1009; McAfee, 6/21/04, 2459:24-2461:21.

                7.3.2.5.8.4.2 For HR at Hallmark, lowering its prices in "an extremely competitive situation against PeopleSoft." P1014; McAfee, 6/21/04, 2464:4-25.

                7.3.2.5.8.4.3 For HR at Greyhound, lowering its prices to "gain a third straight win" over PeopleSoft in a flagship area. P1013; McAfee, 6/21/04, 2465:21-2466:25.

                7.3.2.5.8.4.4 For HRM at CH2M Hill, lowering prices in another "head-to-head" competition. P1006; McAfee, 6/21/04, 2469:5-21.

                7.3.2.5.8.4.5 For Financials at Barnes & Noble, lowering prices to unseat "incumbent Peoplesoft head-to-head." P1004; McAfee, 6/21/04, 2471:23-2472:19.

                7.3.2.5.8.4.6 For FMS at Merrill Lynch, lowering its prices to compete where "Price is the number one key issue..." P1019; McAfee, 6/21/04, 2472:20-2473:22.

              7.3.2.5.8.5 Moreover, Oracle's business records showed, on their face, that PeopleSoft's presence as a competitor at GAF resulted in a 57% change in the price being paid for GAF. P1009; McAfee, 6/21/04, 2462:24-2463:24.

              7.3.2.5.8.6 Oracle's own Merrill Lynch business records demonstrated that PeopleSoft presence as a competitor led to a 20% change in the price being paid by Merrill Lynch. P1019; McAfee, 6/21/04, 2484:21-2485.

              7.3.2.5.8.7 Oracle's numerous other business records demonstrate that Oracle routinely discounts, that is, lowers the prices it offers to consumers, in response to competition from PeopleSoft. P4729; McAfee, 6/21/04, 2488:2-14.

              7.3.2.5.8.8 Drs. Hausman and Teece were not able to explain that Oracle was not engaged in price discrimination, that is, that Oracle was selling the same software to different buyers and charging different prices. Hausman 6/29/04, 4056:23 - 4057:24; Teece 7/01/04, 4496:1-4498:11.

          7.3.2.6 Oracle and PeopleSoft Compete in Head-to-Head Bidding Competitions.

            7.3.2.6.1 Following the sales process, the vendors negotiate prices and terms with the buyers. P4014-B.

            7.3.2.6.2 In the terminology of economics, the purchase and sale activity is best described as "bidding" because the presence of one company's "bid" is used to extract a better price out of the competing vendor. McAfee 6/21/04, 2509:9-2510:5 and McAfee 6/21/04, 2510:6-22.

            7.3.2.6.3 Here, there were numerous examples of Oracle and Peoplesoft bidding against one another in head-to-head competitions for the sale of HRM and FMS products. P1009; P1013; P1006; P1004; P1019; P1014.

            7.3.2.6.4 Dr. McAfee discussed these head-to-head bidding competitions between Oracle and Peoplesoft and how the impact of that competition will not be replaced after a merger. McAfee 6/21/04, 2488:2-2490:5; 2458:6-12; 2460:4-12; 2462:7-23;2464:15-25; 2469:14-21;2472:1-19.

            7.3.2.6.5 Numerous other examples of head-to-head competitions were entered into evidence, including, e.g., P1075; P1081; P1073; P1025; P1066. See section 7.2.1.

            7.3.2.6.6 Oracle sales executives testified to examples of Oracle's head-to- head competitions with Peoplesoft for the sale of HRM and FMS products. P3018, Ciandrini Dep., 1/16/04, 202:7-15.

            7.3.2.6.7 Mr. Ciandrini described a bidding and counterbidding scenario in which Oracle lowered its price to the customer in a series of steps in response to the head-to- head competition with Peoplesoft. P1014, Ciandrini Dep., 1/16/04, 205:4-14, P3018. He testified that Oracle used a "step-down process," for its Hallmark bids, meaning it took Oracle approximately two or three steps to reduce its bid from more than $2 million to approximately $1.2 million dollars. P3018, Ciandrini Dep., 1/16/04, 212:13-213:6. Mr. Ciandrini, who is responsible for sales of applications in North America, testified that an aggressive price from Peoplesoft as part of the competition for application sales was "what they [Peoplesoft] normally do". P3018, Ciandrini Dep., 1/16/04, 202:7-15, at line 9.

            7.3.2.6.8 Oracle discount forms directly reflect the multiple rounds of bidding that occur in Oracle's competitions with PeopleSoft. P1009; P1019; P1073; P1057.

            7.3.2.6.9 A series of discount forms relating to the sale of applications to Cooper Industries reflected how Oracle, due to competition with PeopleSoft over time, repeatedly increased its discount and lowered its price. P1141; see also P1142, P1143, P1144, P1145, P1146, P1147, P1148, P1149, P1150, P1151.

            7.3.2.6.10 Dr. McAfee explained how multiple rounds of bidding are demonstrated on Oracle's discount forms themselves. McAfee, 6/21/04, 2458:11-24; 2603:24-12.

            7.3.2.6.11 The manner in which Oracle approves discounts, with more senior Oracle executives having increasing amounts of authority reflects the fact of the market that Oracle will be engaging in multiple rounds of bidding requiring sequentially higher discounts. P3018, Ciandrini Dep., 1/16/04, 113:11-114:19; P3004, Block Dep., 12/16/03, 143:20-144:4.

          7.3.2.7 Sellers Have Reasonably Good Information on Customers and Rivals.

            7.3.2.7.1 Sellers in the market have good information about buyers, and obtain much of this information during the lengthy sales cycle. McAfee, 6/21/04, 2456:9-23.

            7.3.2.7.2 Oracle documents demonstrate that Oracle has reasonably good and detailed information about buyers, and at times references the source for this information at the buyer. P1013; P1006: P4729, at Gap, Inc; P4729.

            7.3.2.7.3 Dr. McAfee explained that Oracle records demonstrate how "champions" of sellers within the account can be an information source for Oracle. P1013; See McAfee, 6/21/04, 2467:1-2468:6.

            7.3.2.7.4 At Greyhound, Oracle gained information that PeopleSoft had gained recent executive penetration and some champions. P1013; See McAfee, 6/21/04, 2467:1-2468:6.

            7.3.2.7.5 At Constellation Energy Group, Oracle gained information that although former Oracle champions were no longer in a position to influence the buying decision, Oracle had worked to regain executive access and was aware of the specific options Constellations Energy Group was providing to them. P4729, at Constellation Energy Group.

            7.3.2.7.6 At GAP, Oracle gained information that Oracle was a preferred choice of executives, but those executives would not override the decision of the users. P4729, at Gap, Inc.

            7.3.2.7.7 At Maytag, Oracle's "coach"explained, in a scenario similar to the one Dr. McAfee explained at trial, that the Maytag CEO and the Maytag CIO had different priorities with respect to the possible selection of Oracle's e-business suite. P4729, at Maytag.

            7.3.2.7.8 At CH2M Hill, Oracle gained insight of the level of trust and confidence executives at that account had for Oracle. P1006; McAfee, 6/21/04, 2470:22-2471:12.

            7.3.2.7.9 Oracle Can Gain Information about the Customer's Return On Investment. Information Oracle gains through the sales cycle and during the bidding process allows Oracle to price discriminate on the strength of crucial customer information (i.e. the value of customers placed on Oracle's products).

              7.3.2.7.9.1 Oracle learns information about the customer's budget and their ROI . P4729, at Forsythe Technology Inc. P4729, at Family Dollar form P4729, at Church of Scientology.

              7.3.2.7.9.2 An "ROI" or Return on Investment is a calculation made by buyers of software licenses, and it expresses the value that the software holds for the buyer. P3004, Block Dep., 12/16/03, 177: 21- 178:6; Peters, 6/14/04, 1272:51274:23, at 1274:16-23.

              7.3.2.7.9.3 An ROI is a study or analysis that's done to determine what the benefits are of implementing the software, what the payback period is, measured against the costs and the risks. P3004 Block Dep., 12/16/03, 177: 21 - 178:6.

              7.3.2.7.9.4 According to Mr. Block, there are few if any instances in which an ROI will not be prepared, since even if the customer does not prepare the ROI, and the customer does not accept Oracle's offer to prepare an ROI, Oracle will sometimes go ahead and prepare the ROI on its own. P3004, Block Dep., 12/16/03, 179:1-12.

              7.3.2.7.9.5 According to Mr. Block, there are few if any instance in which Oracle will not have some understanding of what the customer's ROI would be, since even if all customers do not share ROI information with Oracle, and even if there are customers who do not want Oracle to do an ROI for them, Oracle will sometimes go ahead and prepare the ROI on its own. P3004, Block Dep., 12/16/03, 179:1- 20.

              7.3.2.7.9.6 Mr. Ciandrini, another Oracle sales executive also testified that Oracle encourages customers to do an ROI, offers to provide the ROI. P3018, Ciandrini Dep., 1/16/04, 92:10-22.

              7.3.2.7.9.7 According to Mr. Ciandrini, Oracle, in certain instances, works with the customer and provides a template and discusses with the customer how to populate the template for the ROI, and that the ROI process can be a "two-way thing". P3018, Ciandrini Dep., 1/16/04, 92:10-93:4.

            7.3.2.7.10 Oracle Also Gains Information about the Firm(s) it is Competing Against in Specific Competitions.

              7.3.2.7.10.1 For the sale of HRM and FMS software, Oracle often has information that it is competing with SAP and PeopleSoft. P4729, at AIG; P4729, Bay Area Rapid Transit; P4729, at City of Atlanta; P4729, at Dassault Systems.

              7.3.2.7.10.2 As presented by Dr. McAfee at trial, there are many instances in which Oracle understands it engaged in head-to-head competitions with PeopleSoft. P1019; P1009; P1004; P1013; P1006; P1014.

              7.3.2.7.10.3 At Barnes & Noble, the Oracle representative described the competition as a "head-to-head competition with PeopleSoft for the sale of HRMS licenses, specifically the core HRMS, Payroll, Benefits, and Time and Expense licenses. P1004, McAfee, 6/21/04, 2472:1-19.

              7.3.2.7.10.4 At Greyhound, the Oracle representative described the competition as a potential "3rd straight win over [PeopleSoft] in their flagship area." P1013; McAfee, 6/21/04, 2466:4-13.

          7.3.2.8 See sections 7.1.1.7, 7.1.1.9.7 and 7.1.1.10 above.

          7.3.2.9 Because the markets for HR and FMS high-function software are ones in which customers have heterogeneous demands, and a substantial number of customers have a preference for PeopleSoft and Oracle, and the markets are ones in which price discrimination can be practiced and the vendors have a substantial amount of information about customers' demands and the ability of other vendors to meet them, these markets are ones in which a merger of Oracle and PeopleSoft is likely to raise prices to customers because the competitive check of competition between the two of them is removed, and competition from SAP is unlikely to replace the effect of the lost competition. Elzinga, 6/18/04, 2519:13-2160:11; 2265:15-2267:19; 2371:18-2372:21; McAfee, 6/21/04, 2485:10-2490:5, at 2488:15-2489:9; 2532:15-2534:2.

        7.3.3 Dr. McAfee Performed an Analysis Using Regressions That Demonstrates the Significant Competitive Effect That Peoplesoft Is Having on Oracle's Prices.

          7.3.3.1 Dr. McAfee employed regression analysis to determine the effect of PeopleSoft's and other competitors on Oracle's discounts. McAfee, 6/21/04, 2495:18-2496:2.

          7.3.3.2 Through regression analysis, Dr. McAfee has studied whether the significant effect PeopleSoft has on Oracle's discounts holds more across a broader range of Oracle HRM and FMS license sales. McAfee, 6/21/04, 2497:7-14.

          7.3.3.3 Dr. McAfee uses two separate datasets in his regression analysis. One dataset involves E-Business Suite sales reported in two charts Mr. Ciandrini used to track discount trends (P874; P4887). McAfee, 6/21/04, 2496:3-17; 2504:21-2505:2.

          7.3.3.4 Dr. McAfee constructed the second dataset by matching Oracle discount approval forms to corresponding transactions in Oracle's sales rep survey reports. McAfee, 6/21/04, 2496:3-17; 2504:21-2505:2.

          7.3.3.5 The Percentage Discount Regression relates Oracle's discounts on E-Business Suite sales to the presence of PeopleSoft and other vendors as competitors, and to the size of the deal. McAfee, 6/21/04, 2497:15-2498:16; P4023B.

          7.3.3.6 Running the Percentage Discount Regression across all deal sizes, Dr. McAfee finds that competition from PeopleSoft on an E-Business Suite sale is associated with an Oracle discount of about 10 percentage points higher on average. McAfee, 6/21/04, 2498:17-2499:5; P4023B.

          7.3.3.7 Running the Percentage Discount Regression on the subset of deal sizes above $500,000, Dr. McAfee likewise finds that competition from PeopleSoft on an E-Business Suite sale is associated with an Oracle discount of about 10 percentage points higher on average. McAfee, 6/21/04, 2499:6-2500:3; P4023B.

          7.3.3.8 An increase in Oracle's discount at this level translates to a change in price for the purchase of the E-Bus suite license. McAfee, 6/21/04, 2499:6-2500:3; P4023B.

          7.3.3.9 Dr. Hausman criticized the specification of Dr. McAfee's E-Business Suite Discount Regression. After re-estimation using Dr. Hausman's specification, Dr. Hausman found that competition from PeopleSoft on an E-Business Suite sale is associated with an Oracle discount of 10 percentage points. Hausman, 6/29/04, 4004:7-4005:9.

          7.3.3.10 Dr. McAfee has also performed an Incremental Percentage Discount Regression, on a separate dataset he constructed by matching Oracle discount approval forms to corresponding transactions in Oracle's sales rep survey reports. McAfee, 6/21/04, 2503:25-2505:2; P4023A.

          7.3.3.11 This second dataset, consisting of ERP, and E-Business Suite sales, has increased information on the circumstances of each transaction, which potentially allows for better control of factors other than competition that may influence Oracle's discounts. McAfee, 6/21/04, 2503:25-2505:2; P4023A.

          7.3.3.12 The second dataset also has increased information on Oracle's discounts. It breaks out Oracle's total discount into two components: an e-biz discount that the customer obtains automatically based on deal size, and an incremental discount that typically requires additional justification and executive approval. McAfee, 6/21/04, 2502:15-2503:11; P4023A.

          7.3.3.13 The Incremental Percentage Discount Regression controls for customer size, whether the sale involves an E-Business Suite rather than distinct ERP modules, and whether the customer has previously purchased applications from Oracle. McAfee, 6/21/04, 2505:4-25; P4023A.

          7.3.3.14 The Incremental Percentage Discount Regression also controls more precisely for deal size, through a Net License Revenue variable. McAfee, 6/21/04, 2506:1-20; P4023A.

          7.3.3.15 Given that the Incremental Percentage Discount being explained is arithmetically related to Net License Revenue, Dr. McAfee was concerned about a possible feedback effect in this regression. McAfee, 6/21/04, 2506:1-20; P4023A.

          7.3.3.16 Using a standard econometric technique called "instrumental variables," Dr. McAfee determined that feedback was not a problem in this case. McAfee, 6/21/04, 2506:1-20; P4023A.

          7.3.3.17 As with the Percentage Discount Regressions, Dr. McAfee has run the Incremental Percentage Discount Regression two ways: across all deal sizes, and for the subset of deals larger than $500,000 in Net License Revenue. McAfee, 6/21/04, 2506:21-2507:10; P4023A.

          7.3.3.18 Across all deals, Dr. McAfee finds that competition from PeopleSoft is associated with an Oracle discount of about 8 percentage points higher on average. McAfee, 6/21/04, 2506:21-2507:10; P4023A.

          7.3.3.19 For the subset of large deals, the corresponding figure is about 14 percentage points. McAfee, 6/21/04, 2506:21-2507:10; P4023A.

          7.3.3.20 Dr. McAfee finds that these incremental discounts are about 8 and 14 percentage points off list price. McAfee, 6/21/04, 2507:11-20.

          7.3.3.21 An increase in the Oracle discount of this size translates to Oracle price reductions of about 13 and 25 percent, respectively, when PeopleSoft is a competitor for the deal. McAfee, 6/21/04, 2507:11-20.

          7.3.3.22 Dr. Hausman criticized the way in which Dr. McAfee calculated the incremental discounts off list price. Dr. Hausman argued that the incremental discounts are about 6 to 10 percentage points. Hausman, 6/29/04, 4014:5-11.

          7.3.3.23 Accepting Dr. Hausman's argument would result in the price customers would pay being significantly decreased due to PeopleSoft. Hausman, 6/29/04, 4014:5-11.

          7.3.3.24 Dr. McAfee's regressions demonstrate that the presence of Peoplesoft is correlated with a significantly higher Oracle discount rate. McAfee, 6/21/04, 2507:11-20; 2498:17-2499:5; P4023A, P4023B.

          7.3.3.25 The regressions are based on Oracle data which could reflect a number of factors that exist in these markets today. Whatever effect these other factors may have on competition, Dr. McAfee's regressions indicate that holding them constant, the presence or absence of PeopleSoft in an opportunity made a significant difference in the prices paid. McAfee, 6/21/04, 2611:57-2612:12; 2534:8-2537:10 at 2535:23-2536:18.

        7.3.4 Auction Theory. Economic Theory Predicts That this Merger Would Likely Have a Significant Competitive Effect.

          7.3.4.1 Economists routinely use theoretical models of competition to describe the behavior of industries, and an auction theory is the appropriate framework for high function HR and FM software. McAfee, 6/21/04, 2509:6-2510:5; 2589:15 - 21; 2589:22 - 2590:21.

          7.3.4.2 High function HRM and FMS software is ordinarily purchased through an involved procurement process, so an economic model of price determination in the relevant markets must account for individual transaction of a purchase of HRM and FMS software licenses. Exhibit P4014B; McAfee, 6/21/04, 2598:23-2599:8.

          7.3.4.3 As Dr. McAfee explained, given the prevalence and magnitude of price discrimination in these markets, other models were not appropriate. McAfee, 6/21/04, 2598:23-2599:8.

          7.3.4.4 Economists studying negotiations utilize two classes of models - bargaining models and auction models. McAfee, 6/21/04, 2510:6-22.

          7.3.4.5 The difference between bargaining models and auction models is that the latter includes any situation where one party is using competition among would-be participants on the other side of the transaction to effect better terms of trade. McAfee, 6/21/04, 2509:6-2510:5; 2589:22-21.

          7.3.4.6 Bargaining models do not include any dynamic in which the bid of one potential buyer or seller is used as part of the negotiation with other buyers or sellers. McAfee, 6/21/04, 2510:6-2510:22.

          7.3.4.7 The presence or absence of formal trappings of an auction (auctioneer, gavels, and paddles) is not relevant to an economist in determining whether an auction theory provides an appropriate insight into the way prices are determined. McAfee, 6/21/04, 2510:23-2511:8; 2586:21-2587:6.

          7.3.4.8 Since the pricing of high function HR and FM software generally involves the use of each potential supplier's bid against the other potential suppliers, an auction theory is the economic model that allows for such a competitive process and is the best way to understand the effect of eliminating competition. McAfee, 6/21/04, 2511:9-2512:12.

          7.3.4.9 Dr. McAfee used an auction theory to model the competition to sell high function HR and FM software. McAfee, 6/21/04, 2511:9-2512:12.

          7.3.4.10 Dr. McAfee simulated the likely effects on prices of high function HR and FM software using a normal model of auctions routinely employed by economists. McAfee, 6/21/04, 2509:6-8.

          7.3.4.11 The particular auction theory used by Dr. McAfee fits the facts of this case. It captures the multiple rounds of bidding, the head-to-head competition, and the reasonably good information competing buyers have about each other and about the customer. McAfee, 6/21/04, 2511: 9-2512:12; 2602: 18-2604:12.

          7.3.4.12 Head-to-head competition makes auction theory, in general, the appropriate class of economic theories for describing the relevant markets. McAfee, 6/21/04, 2511:9-2512:12 at 2512:3-12; 2458:6-12.

          7.3.4.13 Auctions that proceed with multiple rounds of bidding involve a particular mechanism by which the bids of one competitor are used against other competitors. Each round of bidding, the new bids respond to the bids of the last round. McAfee, 6/21/04, 2511:9-2512:12 at 18-22; 2459:12-18.

          7.3.4.14 Multiple rounds of bidding is not the only way an auction can work. But, Dr. McAfee observed that multiple rounds of bidding was very common in the relevant markets and selected an auction model with this mechanism. McAfee, 6/21/04, 2511:9-2512:12; 2512:21-2513:10.

          7.3.4.15 Auctions can proceed where bidders have relatively more or less information relating to their competitors and relating to the buyers in a transaction in a given market. McAfee, 6/21/04, 2518:24-2520:3.

          7.3.4.16 The amount of information bidders have about their competitors and from the buyers can affect the strategies employed by bidders. McAfee, 6/21/04, 2518:24-2520:3; 2588:16-2589:14.

          7.3.4.17 Dr. McAfee observed that the sellers of high function HR and FM software generally have good information about their competitors and come to have a good idea of the value of the various products in the market to a customer as the process goes along. McAfee, 6/21/04, 2589:3-14.

          7.3.4.18 Dr. McAfee selected an auction model without all of the complications of significant informational advantages to one side or another as the bidding behavior did not demonstrate the kinds of strategies that would follow from such an information structure. McAfee, 6/21/04, 2589:22-2590:21; 2589:3-14.

          7.3.4.19 Auctions can also proceed where every bidder is identical or where they offer a more or less differentiated choice. McAfee, 6/21/04, 2600:21-2601:9; 2601:17-2602:7.

          7.3.4.20 Dr. McAfee observed that the products in the relevant markets are differentiated and selected an auction theory which allowed for a variable degree of differentiation. McAfee, 6/21/04, 2615:15-2619:1 at 2617:19-2618:14.

          7.3.4.21 Dr. McAfee's choice of an English auction with complete information about randomly distributed valuations by each customer separately for each of the products available in the market fit the facts of the case. McAfee, 6/21/04, 2518:19-2520:3; 2517:4-2518:6; 2513:11-21.

          7.3.4.22 The economic model used by Dr. McAfee accounts for the essential facts about how competition occurs in this market, that is, that prices are lowered in order to increase a would-be supplier's chance of winning a license sale to any particular customer. McAfee, 6/21/04, 2511:9-2512:20.

          7.3.4.23 The degree to which this impetus to lower prices will be reduced (and prices will be higher) after a merger of Oracle and PeopleSoft depends on the likelihood that Oracle and PeopleSoft would have been the two best choices for any given customer and the degree to which products in the market tend to be differentiated. McAfee, 6/21/04, 2511:9-2512:12.

          7.3.4.24 Dr. McAfee calibrated these two parameters of the model (the probability distribution by which any two vendors are identified as the best two vendors for any particular customer and the degree of differentiation) by selecting the values for these parameters to predict how the market behaves today, absent a merger of Oracle and PeopleSoft. McAfee, 6/21/04, 2528:7-19.

          7.3.4.25 The observed market shares reveal the distribution of the best vendor for all of the customers today. McAfee, 6/21/04, 2522:14-2523:10; 2599:11-18.

          7.3.4.26 Dr. McAfee, accordingly, used Dr. Elzinga's market shares to calibrate this parameter. McAfee, 6/21/04, 2513:22-2514:11.

          7.3.4.27 Dr. McAfee also included all other options observed to have been selected by potential customers in Oracle's records. McAfee, 6/21/04, 2513:22-2514:11; 2598:12-17; 2523:11-14.

          7.3.4.28 The distribution of first-best vendors also gives the distribution of second-, third-, etc. best vendors unless there is a reason to include some bias in the rankings based on the identity of the higher ranked vendors. McAfee, 6/21/04, 2615:15-2619:1, at 2617:13-2618:3.

          7.3.4.29 Dr. McAfee did not include such bias in the values of customers for alternative products. McAfee, 6/21/04, 2615:15-2619:1 at 2618:4-24.

          7.3.4.30 Dr. McAfee allowed the degree of differentiation of the products in the market to vary. McAfee, 6/21/04, 2524:21-2525:12; 2529:22-2530:22.

          7.3.4.31 If a customer was willing to pay $1M for the most valuable product, Dr. McAfee allowed that on average the customer might be willing to pay $500K, $600K, $700K, $800K, or $900K for the next most valuable offering. McAfee, 6/21/04, 2528:7-2529:19; 2597:6-2598:4.

          7.3.4.32 Dr. McAfee offered several calculations demonstrating that this range covered all reasonable values of this parameter. In particular, the lowest estimate Dr. McAfee made was that the next best alternative might be worth $540K to the customer. McAfee, 6/21/04, 2516:15-2518:6 at 2518:3-6; 2529:22-2530:22; 2531:6-19.

          7.3.4.33 The appropriate economic theory as fit to the facts of this case predicts that the merger would give Oracle the ability and incentive to impose a significant price increase on future purchasers of high function HR and FM software. McAfee, 6/21/04, 2531:24-2532:14.

          7.3.4.34 The model does not necessarily predict a large price increase for all mergers of firms in such a market. The size of the predicted effect on price depends on the parameters and the identity of the merging firms. McAfee, 6/21/04, 2596:13-18.

          7.3.4.35 In this case, the parameters Dr. McAfee picked as consistent with the reality of today's market shares confirms that Oracle and PeopleSoft are often each others' best competition and are often the customer's top two choices. McAfee, 6/21/04, 2597:20-2598:4.

          7.3.4.36 This is the only model advanced by any economist as fitting the facts of the case. McAfee, 6/21/04, 2511:9-2512:12; Hausman, 6/29/04, 4066: 15-4070:23 at 4068:1-21.

          7.3.4.37 Dean Campbell suggested using bargaining theory which would obscure the competitive dynamic between Oracle and PeopleSoft in bidding to win business. Campbell, 6/22/04, 2740:22-2741:11.

          7.3.4.38 However, Dean Campbell did not suggest a particular bargaining theory. He merely noted that across all of the variations of such theories the prediction of a merger's effect on price is indeterminate. Campbell, 6/22/04, 2742:2-14.

          7.3.4.39 Although there was significant evidence in the case that three vendors (i.e. SAP, PeopleSoft, and Oracle) compete for procurements, Dr. McAfee's model demonstrated significant price effects due to a merger where two bidders are competing in the final round of the competition. McAfee, 6/21/04, 2563:1-5.

        7.3.5 Dr. McAfee Conducted Three Independent Analysis to Assess the Competitive Effects of an Oracle/peolplesoft Merger. McAfee, 6/21/04, 2532:15-2534:2.

          7.3.5.1 Dr. McAfee's first analysis, case studies, demonstrated that competition from PeopleSoft resulted in significantly greater discounts from Oracle, and hence, significantly lower prices in the order of 20 to 57% lower. McAfee, 6/21/04, 2532:15-22.

          7.3.5.2 The second type of analysis Dr. McAfee performed were regressions, that demonstrated, as a market-wide average, the presence of PeopleSoft as a competitor caused significantly greater discounts, and hence lower prices from Oracle. McAfee, 6/21/04, 2532:23-2533:3.

          7.3.5.3 The third analysis Dr. McAfee performed was the merger simulation that demonstrated market-wide, that an Oracle/PeopleSoft merger could cause price increases ranging from 5% to 30% for customers in these markets. P4024A, McAfee, 6/21/04, 2531:24-2532:14.

          7.3.5.4 All three of Dr. McAfee's analysis independently and collectively demonstrated that a significant unilateral effect would likely result following an Oracle/PeopleSoft merger. McAfee, 6/21/04, 2533:15-2534:2.

          7.3.5.5 Even when the bidders do not have good information about customers' preferences, elimination of one of the three choices will harm customers. McAfee, 2486:17-2490:5 at 2489:21-5; Elzinga 6/18/04: 2217:l-9;2245:4-2246:4.

          7.3.5.6 Dr. Elzinga explained how PeopleSoft constrains Oracle's prices today "when Oracle is asking for more discount help to try and cut a deal or gain a sale" and that the "push would be gone if the merger were consummated." Elzinga, 6/18/04, 2235:8-21 at 19-21.

          7.3.5.7 Dr. McAfee's analyses explain the impact on prices if the merger were consummated. McAfee, 6/18/04, 2532:15-2533:9.

          7.3.5.8 The case studies demonstrate the increase in prices (and decrease in actual amount paid) if PeopleSoft were not involved in the head-to-head bidding contests with Oracle. McAfee, 6/18/04, 2463:13-2464:3; 2484:21-2485:7.

          7.3.5.9 The regressions demonstrate the impact PeopleSoft's presence has on Oracle's discounts, impact that would not exist after Oracle and PeopleSoft merge. McAfee, 6/18/04, 2508:4-22.

          7.3.5.10 And the auction model simulates the predicted price increases, market-wide, following an Oracle/PeopleSoft merger. P4024A, McAfee, 6/18/04, 2531:24-2532:14.

      7.4 The Merger Raises Concern for Possible Additional Harmful Anticompetitive Effects.

        7.4.1 Mergers In Highly Concentrated Markets Can Result in Harmful Coordinated Effects.

          7.4.1.1 Dr. Elzinga concluded that Oracle, PeopleSoft and SAP are the three competitors in the market for high function HRM and FMS software. Elzinga, 6/18/04, 2158:22-2159:9. The merger of Oracle and Peoplesoft would create a duopoly in the market for high function HRM and FMS software.

          7.4.1.2 As Dr. Elzinga stated, "I don't see any pro-competitive benefits from turning this market into a duopoly, and eliminating the very beneficial competition that now exists between Oracle and PeopleSoft." Elzinga, 6/178/04, 2245:4-2246:4.

          7.4.1.3 The market would have very few sellers - two, the fewest possible without becoming a monopoly.

          7.4.1.4 The market has high entry barriers. See section [[VHI._]] infra.

          7.4.1.5 The products involved do not have close substitutes. See section 3 supra.

          7.4.1.6 The market has a large number of customers. Catz, 6/25/04, 3494:24-3495:9 (Oracle has 4500 customers); See Section [[VI.E.l.e,]] infra (systems integrators report hundreds of implementations).

          7.4.1.7 Oracle and SAP Have Long-standing Vertical (or Industry) Specializations; In a Two-Firm Market, Such Specializations Could Become Even More Pronounced, Especially if the Two Closer Competitors (Oracle and PeopleSoft) Are Merged.

            7.4.1.7.1 Oracle, SAP and PeopleSoft have all had different strengths historically and Oracle and SAP have taken the major market share in those industries in which their respective strengths were important to the industry. For example, Oracle had very rich financial functionality and was dominant in industries in which financial functionality was most important. SAP had very rich supply chain and manufacturing capability and had won a majority of the market share in industries in which those characteristics were important. Keating, 6/10/04, 898:11-899:3.

            7.4.1.7.2 SAP is stronger in manufacturing and consumer products. Knowles, 6/23/04, 2943:12-15.

            7.4.1.7.3 Keating testified that Oracle and PeopleSoft historically have sold their products in modules that could be separately licensed and implemented, while SAP, at least until more recently, was all-encompassing and not licensed in separate modules. Keating, 6/10/04, 901:6-20. The customer had to buy the entire SAP suite. Keating, 6/10/04, 946:18-20. The modular architecture gave Oracle and Peoplesoft advantages in industries, such as telecommunications, in which the customers had other key systems that they needed to interface with the Oracle and PeopleSoft products. SAP took the lead in industries in the all-in-one software solution best matched the customer's requirements. Keating, 6/10/04, 901:20 to 902:15.

            7.4.1.7.4 SAP sometimes cannot penetrate particular verticals because it does "not have exactly the right functionality to serve that particular industry." Knowles, 6/23/04, 2943:1-11.

            7.4.1.7.5 For many customers, Oracle and PeopleSoft are the two best options. See section [[III.B.l]], supra.

            7.4.1.7.6 PeopleSoft and Oracle are "very close competitors in some of the service- related industries like financial services and utilities and telecommunications." PeopleSoft has an advantage over Oracle in the staffing and consulting and manufacturing industries. Wilmington, 6/16/04, 1815:23 -1816:8.

            7.4.1.7.7 Dr. Elzinga found that Oracle's discount approval forms demonstrated that Oracle competed against Peoplesoft more than SAP. Elzinga, 6/18/04, 2176:9-2179:17, at 2177:3-11: P3175.

            7.4.1.7.8 Oracle describes PeopleSoft as its #1 competitor. P2093 at ORCL-EDOC- 00042674.

            7.4.1.7.9 Vendors generally know against whom they are competing. See [[VH.C3.e- g]], supra.

            7.4.1.7.10 The vendors generally learn who wins a procurement, even in the existing three-firm market. [[VH.C3.e-g]], supra.

            7.4.1.7.11 Customers believe that there will be less incentive for vigorous competition after a merger of Oracle and PeopleSoft. Customers, such as Verizon, believe that neither SAP nor Oracle will be as "accommodating" and willing to negotiate after the merger as they are now. Bradley, 6/9/04, 635:22-636:4.

            7.4.1.7.12 Ms. Bradley explained the reason that Verizon decided to license with SAP rather than waiting to see the outcome of Oracle's attempt to merge with PeopleSoft. "[We] have seen in many technology markets and many technology negotiations that when the number of competitors is just two, its just a less vibrant market." Bradley, 6/9/04, 612:1-18.

          7.4.1.8 SAP and Oracle Have a Shared Economic Interest in the Continuing Benefit to Both From a Joint Business Relationship: SAP's Value-Added Reselling of Oracle's Database.

            7.4.1.8.1 SAP is Oracle's number one reseller and partner for Oracle databases. P3037, Knowles Dep., 5/3/04, 373:6-374:5.

            7.4.1.8.2 As the largest reseller of Oracle databases, SAP generates approximately $100 million in reseller revenues to Oracle annually, which "dwarfs" the database royalties of any other company to Oracle. P3033, Henley Dep., 5/4/04, 85:11-20.

            7.4.1.8.3 Sixty-six percent of SAP installations are on the Oracle database. P0139 at ORLITF0057636.

            7.4.1.8.4 Oracle's ongoing business relationship with SAP includes a small Oracle team stationed in Germany working with SAP to make its applications operate better on the Oracle database. P3033, Henley Dep., 5/14/04, 81:12-82:20.

      7.5 The Predicted Harmful Effects Are Likely to Be Substantial. The Volume of Commerce in FMS and HRM Software Affected By the Merger of Oracle and PeopleSoft Would Be Large.

        7.5.1 the Combined Software License and Maintenance Price Will Be Escalated as a Result of the Merger. Elzinga, 6/18/04, 2400:11-2401:18. (Responding to Court's questions.)

        7.5.2 Substantial Annual Licensing Revenues Would be Affected By the Merger.

          7.5.2.1 For customers representing FMS license revenues of at least $500,000, total license revenues for U.S. sales in 2003 are about $44.7 million for SAP, about $35.7 million for PeopleSoft, and about $19.2 million for Oracle, for a total across these three vendors of about $99.6 million. P3177.

          7.5.2.2 For customers representing HRM license revenues of at least $500,000, total license revenues for U.S. sales in 2003 are about $65.1 million for PeopleSoft, about $22.9 million for Oracle, and about $36.7 million for SAP, for a total across these three vendors of about $124.7 million. P3178.

        7.5.3 In addition, Substantial Streams of Maintenance Revenue Would be Affected by the Merger.

          7.5.3.1 When a customer buys enterprise software, the customer gets more than just the license. The nature of the product is that it requires continual updating, as tax and regulatory rules change. Elzinga, 6/18/04, 2396:6-14. In addition to the one-time license fee, customers pay annual maintenance fees to receive a steady stream of innovation and improvement in the software product. Wohl, 6/25/04, 3379:20-3380:9; 3386:13-23.

          7.5.3.2 In competing for a sale, a vendor takes into account that with success will come not only a license fee but a stream of income through a maintenance contract. Elzinga, 6/18/04, 2400:21-2401:12.

          7.5.3.3 PeopleSoft's initial maintenance contracts can extend more than five years. P0031 at SAP-05769. Customers' maintenance relationships with enterprise software vendors are long term, lasting five to ten years or more. [REDACTED TEXT]

          7.5.3.4 Ongoing annual maintenance fees are typically pegged to about 20% to 22% of the one-time license fee. Elzinga, 6/18/04, 2396:1-5; P4965 at PS-TE2251795.

          7.5.3.5 Defendant's economic expert, Dr. Teece admits that his calculations do not include maintenance fees. Teece, 7/1/04, 4478:1-4479:5.

          7.5.3.6 The market power opportunity to raise license fees post-merger would be leveraged to maintenance fees as well. Elzinga, 6/18/04, 2397:1-4.

        7.5.4 Not Only Would Maintenance Streams Linked to New License Sales be Affected by the Merger, But Also the Ongoing Maintenance Revenues Flowing From the Vendors' Installed Bases of Existing Customers.

          7.5.4.1 One of Oracle's rationales for acquiring PeopleSoft is PeopleSoft's strong revenue stream from maintenance, upgrades and support. P2041 at ORCL-EDOC-00192190.

          7.5.4.2 Of PeopleSoft's total revenues in FY2002, 32% were from maintenance. P2041 at ORCL-EDOC-00192194. PeopleSoft's total revenues in 2002 were about $1.95 billion, 32% of which would be about $624 million. D7149 at 25.

          7.5.4.3 About 20% of Oracle's total license revenues for FY2002 were from applications. D7147 at 20. Oracle's total product support revenues for FY2002 were about $3.9 billion, 20% of which would be about $770 million. D7147 at 22.

          7.5.4.4 Once the term of the initial maintenance contract (or the period of the negotiated price cap) has passed, maintenance fees are subject to upward revision. P0031 at SAP-05769. The possible increase could run as high as 98%. P0031 at SAP-05770.

          7.5.4.5 In renegotiating a maintenance contract, a buyer's ability to negotiate a better price is, no less than in the case of a license purchase, a function of the alternatives open to the buyer. The proposed merger, if permitted to go through, would eliminate one of those alternatives. Elzinga, 6/18/04, 2236:9-17.

          7.5.4.6 The competition eliminated by the merger is a particularly attractive for enterprises that have a split applications footprint of PeopleSoft HRM and Oracle FMS software.

            7.5.4.6.1 Given a split HRM and FMS footprint, the alternative of consolidating onto a single existing vendor has advantages over switching one of the applications to a new software vendor. Consolidation offers better integration across the applications (Wohl, 6/25/04, 3412:1-17), and avoids the cost and risk of retraining personnel on a new software package [REDACTED TEXT].

            7.5.4.6.2 There are a lot of enterprises that have PeopleSoft HRM and Oracle FMS software installed. Wohl, 6/25/04, 3411:22-24. Based on market research and professional experience, Mr. Keating (BearingPoint) believes there are more than 1200 enterprises with such a split footprint. Keating, 6/10/04, 931:12-932:7.

        7.5.5 Substantial Future Sales Of High Function Software Are Projected By Competitors, Customers and Systems Integrators.

          7.5.5.1 Analysts project software license revenues to grow "upwards of 6% per year through 2007." D7041, at PS-TH2802745.

          7.5.5.2 Competitors are projecting significant opportunities for sales of high function software.

            7.5.5.2.1 At this time there is "no saturation" with respect to sales of HRM and FMS core functionality in the ERP market. Knowles 6/23/04, 2966:6-2966:9.

            7.5.5.2.2 A PeopleSoft study identified "over 1,500 named accounts" between 500 million and 5 billion in revenue in the US to target for commercial HCM sales. More than a "fourth of which are on legacy and home grown systems...and burning money on them..." D7041, at PS-TH2802748.

            7.5.5.2.3 PeopleSoft sees significant sales opportunity in selling to customers with a "homegrown...or custom system that an account has put in place that's generally costing too much money...while being inefficient...ineffective...and difficult to modify as the business needs change..." D7041, at PS-TH2802747.

          7.5.5.3 Customers anticipate the need to purchase additional HRM and FMS software in the future.

            7.5.5.3.1 Where Oracle and PeopleSoft both have software installed at a customer, there is opportunity for additional sales, and with both companies in the market, for competition. Oracle and PeopleSoft "have plenty of common customers," i.e., customer with both Oracle and PeopleSoft applications, such as Oracle FMS and PeopleSoft HRM. Such common customers sometimes ask Oracle for quotes to replace the PeopleSoft application and Oracle sales people pursue opportunities to displace Peoplesoft from those accounts. P3033, Henley Dep., 5/04/04, 146:12-147:09; 147:11-149:13.

            7.5.5.3.2 Competition may occur with existing clients facing a significant upgrade. When Nextel was confronted with a decision on whether to upgrade its Oracle FMS software to 11i, it brought in consultants to look at their options and to evaluate Oracle, PeopleSoft and SAP. Since the technical scores were all close, SAP was eliminated because PeopleSoft and Oracle were already incumbents (HRM and FMS, respectively). PeopleSoft believes it won the FMS business from Oracle on product grounds - better technology and an Enterprise Planning module - and by focusing on business practices. P2593 (PS-TE2131284-85). See Wesson 6/14/04 1162:08-25.

            7.5.5.3.3 When customers upgrade their ERP systems, they routinely expand their existing licensing to include additional geographic regions. [REDACTED TEXT]

            7.5.5.3.4 [REDACTED TEXT]

            7.5.5.3.5 After Novell completes an acquisition of a company, it attempts to "rationalize" on a single application in order not to have to bear the additional cost of consolidating information from two sets of core functions such as FMS. As a result, Novell opts to integrate its acquisitions on a single pillar solution, which would require it to purchase additional software. P3188 at p.3-4, [REDACTED TEXT].

          7.5.5.4 Systems Integrators Are Projecting Substantial Sales of HRM and FMS Software In The Near Term.

            7.5.5.4.1 . [REDACTED TEXT]

            7.5.5.4.2 [REDACTED TEXT]

            7.5.5.4.3 Bearing Point. During the past three years, Bearing Point's Global Solutions Enterprise group has done more than 2000 Tier One core ERP projects. Of those 900 were Oracle projects, 250 were PeopleSoft and 1,000 were SAP. Keating, 6/10/04, 866:22-867:4, P 200. This business has produced three-year revenues for Bearing Point in excess of $2.2 billion. P200 at 1.

            7.5.5.4.4 In 2003, Bearing Point booked 172 projects for PeopleSoft/JD Edwards, 163 projects for Oracle and 158 for SAP. Keating, 6/10/04., 882:18-883:6.

            7.5.5.4.5 The vast majority of the Bearing Point ERP implementations have been FMS and HRM systems. Keating, 6/10/04, 875:10-18.

            7.5.5.4.6 Mr. Keating testified that the majority of the growth in Bearing Point's ERP services has been with clients that have systems from multiple vendors and wish to reduce the number of systems that they operate. For example, a customer may have Oracle for FMS and PeopleSoft for HRM and will wish to reduce the number of software platforms and applications it is running. Keating, 6/10/04, 930:20-931:11.

            7.5.5.4.7 A company that has Oracle and PeopleSoft and is consolidating to reduce costs will retain one of them and migrate its licenses to a single vendor, rather than buying a new product and driving up the cost. Keating, 6/10/04, 1000:6-12.

            7.5.5.4.8 Bearing Point has identified more than 1200 companies that have "split footprint" Oracle FMS software and PeopleSoft HRM software installed. Keating, 6/10/04, 931:12-932:3.

            7.5.5.4.9 Mr. Keating testifed that he does not anticipate any decline in the next two to three years in Bearing Points business for HRM or FMS implementation services. Keating, 6/10/04, 932:18-23.

            7.5.5.4.10 IBM Global Services. IBM is seeing similar behavior by customers. Nancy Thomas of IBM testified that there are a variety of circumstances that would cause clients to use IBM's services for FMS selections: 1) clients might be seeking to replace inefficient legacy systems; 2) they may have been involved in a merger; 3) their current system may be undergoing a significant upgrade, which causes them to look at alternatives prior to committing to the cost of an upgrade; and 4) regulatory requirements such as Sarbannes-Oxley or Basil II may be causing them to look at additional FMS functionality. Thomas, 6/9/04, 468:17-469:10.

            7.5.5.4.11 IBM's head of software, Steven Mills, meets regularly with customers, mostly larger-size businesses. He finds that there is a lot of discussion about how an integrated ERP suite would improve overall efficiency for their business and reduce the number of elements that have to be integrated. P3193, Mills Dep., 5/27/04, 65:12-21. Customers are making a choice as to which ERP environment to standardize on, and they view it as an opportunity to simplify some aspects of their applications environment and realize more operational efficiency by getting to common structures in their business processes. P3193, Mills Dep., 5/27/04, 66:21-67:13.

            7.5.5.4.12 Many companies are still focused on basic FMS consolidation - getting to one general ledger. "In many cases by reasons of mergers and acquisition and legacy, they're running multiple pieces of software today and they're looking to get one common ledger for their company, and so that is a discussion I do have with businesses." P3193, Mills Dep., 5/27/04, 65:4-66:3.

            7.5.5.4.13 IBM sees companies buying multi-pillar suites all the time, where multi- pillar refers to basic financial and human resources functions. P3193, Mills Dep., 5/27/04, 66:4-8; 69:18-70:13.

            7.5.5.4.14 The costs of adopting a suite vary depending on how many different systems a company is currently operating. IBM refers to these different systems that a company has accumulated over time as "clutter." P3193, Mills Dep., 5/27/04, 68:6-22.

            7.5.5.4.15 Conversion costs are one factor that will influence whether a company goes to a suite approach, but the company will also evaluate the business benefits that they can get by having one common base-level set of infrastructure software that supports those common ERP activities. P3193, Mills Dep., 5/27/04, 68:23-69:8.

            7.5.5.4.16 IBM would not generalize that large organizations have already standardized their ERP systems across the organization. In fact, the opposite might be true in as much as small organizations are unlikely to be running five different bookkeeping systems. Whereas a large organization may have multiple ERP systems across different parts of the business. P3193, Mills Dep., 5/27/04, 73:24-74:19.

            7.5.5.4.17 IBM believes that companies continue to evaluate whether to consolidate to a single HRM or FMS system as they change the way they operate their businesses, or as they grow through mergers and acquisitions. IBM sees customers continuing to evaluate these integrated suites for the foreseeable future. P3193, Mills Dep., 5/27/04, 130:18-134:19.

            7.5.5.4.18 Deloitte and Touche. Deloitte and Touche has had substantial sales of ERP software in the recent past and projects continued demand for new sales.

            7.5.5.4.19 [REDACTED TEXT]

            7.5.5.4.20 [REDACTED TEXT]

            7.5.5.4.21 [REDACTED TEXT]

            7.5.5.4.22 As customers complete their implementations of FMS systems, there will remain opportunities for additional software sales. [REDACTED TEXT]

            7.5.5.4.23 [REDACTED TEXT]

            7.5.5.4.24 [REDACTED TEXT]

            7.5.5.4.25 [REDACTED TEXT]

            7.5.5.4.26 [REDACTED TEXT]

        7.5.6 The Likely Anticompetitive Effect Includes A Decrease In Service Quality As The Pressure To Maintain Satisfied Customers As References and To Prevent Switching Is Diminished.

          7.5.6.1 References from similarly-sized and complex customers are very important for customers considering purchasing high-function software. Bradley, 6/9/04, 595:4-597:12; 600:25-602:18; P0071 (Verizon formal reference form submitted to PSFT to be used as a reference for PSFT's bid for the state of Ohio.) As Ms. Bradley testified, "The current customers for a product need to be taken care of at a reasonable price, or they will poison the well for new customers...." Bradley, 6/9/04, 636:4-18.

          7.5.6.2 With PeopleSoft in the market today, Oracle cannot "dare put Verizon at a significant disadvantage with poor maintenance or exorbitantly high costs," because it runs the risk of losing Verizon as a reference customer. The software vendors' need to compete for "new customers coming along" means that existing customers receive maintenance "at a reasonable price." Without this effect of current competition, "then the old customers could likely be milked for a lot higher support costs." Bradley, 6/9/04, 595:4-596:7.

          7.5.6.3 However, the level of maintenance and support is not guaranteed. Oracle's maintenance contracts do not "guarantee any enhancement or bug fixes or upgrades or anything." Catz, 6/25/04, 3526:14-20. Ms Catz stated, "What we do is we provide, when and if available, all of those things to the customers, as long as they're current on their maintenance fees. Q: All right, but the contract itself doesn't, for example, guarantee any minimum amount of upgrades or enhancements. A: Absolutely not." Catz, 6/25/04, 3526:14-25.

          7.5.6.4 Mr. Wohl testified that he has considered retaining only 10% of the PeopleSoft development staff to add features to the existing PeopleSoft platform in order to follow through on Oracle's public commitment to continue enhancing the PeopleSoft products. Wohl Dep., 4/30/04, at 491:9-493:1.

          7.5.6.5 Ms. Catz stated that Oracle plans to keep some of PeopleSoft's R&D group, but "we have not done an analysis of whose coming to do a [new software] feature and who is doing updates . . " Catz, 6/25/04, 3551:18-3552:6.

          7.5.6.6 As Mr. Wolfe explained, North Dakota is obligated to pay $750,000 in maintenance fees "regardless of whether we get a new release" and that "[c]ompetition motivates the vendor to continue to improve their product every 12 to 18 months to maintain a competitive market place. So they leapfrog each other. ... So, through that direct head-to-head competition they leapfrog each other, coming out with new releases, with new functionality, with new improvements. Either market place demands or they feel important for their product to sell it; and if that motivation isn't there, if there is no competition and so there's no reason for them to give you a new release every 12 to 18 months, they know they are going to get that maintenance fee every year regardless, and so, if they slow down that update process from 18 months to 36 months, it saves them money with the State still paying that million dollars a year to afford those services." Wolfe, 6/16/04, 1594:2-1595:25. (Responding to questions by the Court.)

          7.5.6.7 Daimler Chrysler relies on PeopleSoft to provide timely upgrades and maintenance of its HRM software. Gorriz, 06/15/04, 1383:2-4. Upgrades typically include improved functionality which the software vendor has developed over time and which incorporates the "best practice processes from other companies." Gorriz, 06/15/04, 1383:9-14. A company needs upgrades to keep the system running. Gorriz, 06/15/04, 1384:16-25. If upgrades were not provided, Daimler Chrysler would need to develop improved functionality on its own, which would be more costly. Gorriz, 06/15/04, 1383:15-21.

          7.5.6.8 Although Oracle has stated they would maintain the PeopleSoft HRM software in its present release, customers like Daimler Chrysler's Mr. Gorriz are skeptical, "I doubt that they will further develop the software for a long time, and therefore, I think this would not be beneficial to us." Gorriz, 06/15/04, 1390:9-14.

          7.5.6.9 Mr. Patel at Ford expressed similar concerns. If the proposed merger occurs, and Oracle does not develop a next generation PeopleSoft software, Oracle would have no incentive to upgrade the existing PeopleSoft software that Ford is using. P3041, Patel Dep., 6/03/04, 120:5-8, 18.

          7.5.6.10 For Verizon, "[m]aintenance and support means the evolution of the product." Bradley, 6/9/04, 634:3-8. Ms. Bradley does not believe that Verizon will have sufficient leverage if Oracle acquires PeopleSoft to keep Oracle from moving Verizon off the PeopleSoft platform, which she testified "[would] not compensate us for the fact that it would cost tense (sic) of millions of dollars to accomplish that move." Bradley, 6/9/04, 633:10-20. According to Mrs. Bradley, [I] just want to say that we have options; but they are all unpalatable options. I mean, you can tell me that I have two loathsome options, and you are not going to make me much happier that I have two." Bradley, 6/9/04, 637:6-13.

          7.5.6.11 Ms. Catz' testimony confirms the fears expressed by Verizon, Ford and Daimler Chrysler. She testified that Oracles has no intention to upgrade PeopleSoft's "disparate architecture" or their PeopleTools layer. "The next move, the next big architectural upgrade for the PeopleSoft products would move to a single database, with a single data schema, built on Oracle with the Oracle data base and application server." Catz, 6/25/04, 3450:3-18; 3527:16-3528:4.

          7.5.6.12 While Oracle touts that its promise to offer PeopleSoft's current customers a free enterprise software licenses for equivalent products, PeopleSoft's current customers will be forced to re-invest in implementation costs post-merger if and when they switch to Oracle's applications. Block Dep., 1/5/04, 460:7-462:47.

        7.5.7 The Likely Anticompetitive Effect Includes Increased Future Maintenance Fees to Existing Customers As The Pressure to Maintain Satisfied Customers Is Diminished.

          7.5.7.1 Unsatisfactory maintenance services can have costly consequences for customers. Mr. Peters of Emerson Electric testified that he paid 20 percent of his license fee of $7 million, ($1.4 million per year) in annual maintenance fees. When his maintenance services (from one of the major software vendors) became unsatisfactory he negotiated a 75 percent reduction in his annual maintenance fee. Nonetheless, Emerson still paid 25% of the maintenance fee for unsatisfactory service (a cost of approximately $350,000). Peters, 6/14/04, 1281:13-1283:5 (Responding to questioning by the Court.)

          7.5.7.2 Verizon also expects that maintenance prices will rise and maintenance quality would decline, post-merger because Oracle is buying the PeopleSoft customers, and it uninterested in the PeopleSoft code base. Oracle will not be developing new releases on the PeopleSoft code base or new PeopleSoft customers. Bradley, 6/9/04, 595:4-595:23.

        7.5.8 Another Anticompetitive Effect Is The Loss of Innovation Competition Between Oracle and PeopleSoft to Provide New Features And Functions To Customers and Leapfrog One Another To Gain Sales.

          7.5.8.1 Innovation is important to customers because, "They want to run their businesses more efficiently next year than the previous year. So the better information systems we provide, the more efficiently they can run their business and ... the better decisions they can make." Ellison, 6/30/04, 4312:17-24.

          7.5.8.2 Innovation is "guaranteed by market dynamics but not guaranteed in the maintenance contract." Ellison, 6/30/04, 4312:25-4313:4.

          7.5.8.3 Although monopolists have a financial incentive to innovate, they are less likely to do so, despite the financial incentives. The merger of Oracle and PeopleSoft will lead to a reduced incentive to innovate. Having three independent firms competing is more likely to come up with a leapfrog technology than two firms, or one firm. Elzinga, 6/18/04, 2379:2-2380:10.

          7.5.8.4 Mr. Ellison stated, "I guess if there's no competition, innovation would be a wasted effort." Ellison, 6/30/04, 431325-4314:24.

          7.5.8.5 As described supra, there is substantial innovation today as a result of PeopleSoft-Oracle competition and a substantial likelihood of loss of innovation competition as a result of the merger.

          7.5.8.6 While it is difficult or impossible to quantify the value of lost future innovation, descriptions from market participants of the scope and significance of the current innovation, as well as of concerns about future innovation, suggests that the loss is would be substantial.

          7.5.8.7 PeopleSoft's R&D budget is roughly $420 million. By comparison, SAP's is approximately $1 billion. Bergquist, 6/8/04, 337:2-4; 337:19; 338:5-6. According to its 2003 10K filing, Oracle's total R&D expenditure for all product lines in 2003 was $1.18 billion. P2051. Oracle has roughly twice the number of software developers that PeopleSoft has for an equivalent product footprint. Berquist, 6/8/04, 337:2-10.

          7.5.8.8 Mr. Berquist of Peoplesoft testified that despite the difference in R&D spending between PeopleSoft and Oracle, it has not prevented PeopleSoft from competing and competing effectively. Bergquist, 6/8/04, 338:13-339:3

          7.5.8.9 Vendors try to keep up with one another in terms of identifying deficiencies in their products and closing gaps, as well as trying to be the first one to innovate new features to gain competitive advantage. P3171, Ellison Dep., 01/20/04, 110:19-112:7.

          7.5.8.10 PeopleSoft closely monitors what its competition - including Oracle - is doing in order to stay ahead in the functionality race. Where it sees a competitor offering something new, it tries to match or even leapfrog past the others. Bergquist, 6/8/04, 339:20-340:8.

          7.5.8.11 For example, both Oracle and PeopleSoft claim to have led the market in application software to web-based architectures. Compare the testimony of Mr. Ellison and Mr. Berquist.

            7.5.8.11.1 One example of vendors leapfrogging one another is when Oracle moved from client server to Internet architecture and later virtually everyone followed. P3171, Ellison Dep., 01/20/04, 114:21-116:7.

            7.5.8.11.2 Oracle and SAP also respond to what PeopleSoft is offering. Thus, PeopleSoft's release of AppConnect two years ago was followed within months by SAP's release of a comparable product that it calls Netweaver and Oracle's recent release of a similar product. Bergquist, 6/8/04, 341:9-12; 341:15-342:9.

          7.5.8.12 Systems integrators such as Accenture observe innovation competition between Oracle, PeopleSoft and SAP.

            7.5.8.12.1 [REDACTED TEXT]

            7.5.8.12.2 [REDACTED TEXT]

            7.5.8.12.3 [REDACTED TEXT]

            7.5.8.12.4 [REDACTED TEXT]

          7.5.8.13 There is a significant difference in the cost between upgrading from one version of a vendor's software to another and going to a different vendor's product. This cost difference reflects the fact that different vendors solve a business problem differently. Shifting from one vendor to another requires a change in business model. Bergquist, 6/8/04, 336:1-337:1.

          7.5.8.14 Choosing a vendor's product because of innovation competition stands in stark contrast to the necessity of switching products as a result of a merger, which is the result customers believe they will be faced with if Oracle and PeopleSoft merge. Ford expects significant anticompetitive consequences from the merger.

            7.5.8.14.1 Ford expects to have to switch to Oracle HRM software because it does not expect its existing PeopleSoft software to be updated with new features as it would be if there were competition. Ford was advised by the Oracle CFO that, after the merger, Oracle would not be developing a next generation PS software product.

            7.5.8.14.2 The impact of this on Ford is significant. Ford's current PeopleSoft software will be of little use after a period of time, causing harm to the company's businesses, efficiency and effectiveness. In addition, Ford would have to devote substantial costs, time and resources to move from PeopleSoft to an alternative software product. P3041, Patel Dep., 6/03/04, 117:19-21, 24-25, 118:10-14, 118:17-119:3.

            7.5.8.14.3 Oracle's refusal to develop a next generation PeopleSoft product will bring process inefficiency to Ford because its personnel will have to depend on a PeopleSoft software product which does not reflect the latest technical innovations or the appropriate maturity of usage of Ford's PeopleSoft product. This process inefficiency impacts Ford's business operations and means the company overall would suffer because information technology costs and people costs are so substantial. Anything that requires more work to be done for the same amount of output imposes inefficiency. Mr. Patel believes that Ford would suffer harm as a result. P3041, Patel Dep., 6/03/04, 120:20-122:1.

          7.5.8.15 Ford estimates that it would be required to spend $100 million to replace its PeopleSoft software if the merger occurs. P3041, Patel Dep., 6/03/04, 123:13-22.

          7.5.8.16 Daimler Chrysler predicts similar consequences if Oracle fails to maintain and upgrade the PeopleSoft HR product Daimler Chrysler currently uses, it may be forced to switch platforms, which could cost it as much as $50-100 million. Gorriz, 06/15/04, 1390:15-22; 1427:13-1428:1.

            7.5.8.16.1 Daimler Chrysler does not believe that its negotiating leverage will insulate if from such competitive effects.

            7.5.8.16.2 If as a result of the merger, innovation for PeopleSoft products was slowed down, Daimler Chrysler "couldn't defend" themselves because they are "locked in." It would be forced to do all the innovation and customization itself, which would cost $1-2 million a year, far more than the $500,000 that it is currently paying PeopleSoft for maintenance and free upgrades. Gorriz, 06/15/04, 1434:10-1436:1.

          7.5.8.17 Cox Communications is likewise concerned about a potential migration to Oracle software if this merger is consummated. Such a migration could cost $1-2 million - "a fair amount of pain to Cox for no real gain." Hatfield, 6/7/04, 142:14-143:7, 144:1-145:5.

            7.5.8.17.1 In addition to increased cost, there are other adverse consequences that would result from the merger. The loss of PeopleSoft personnel could affect Cox Communications negatively insofar as certain personnel have developed "tribal knowledge" of Cox, such as history of the company's unique implementation and integration, that cannot be passed on to successors. Hatfield, 6/7/04, 147:5-148:9.

          7.5.8.18 Verizon anticipates that it will be required to spend "tens of millions" to move from PeopleSoft to the Oracle platform should the merger be completed. Bradley, 6/9/04, 633:10-23.

    8. Anticompetitive Effects - Entry/Repositioning. Entry is not likely to alleviate the anticompetitive effects of the merger. Nor is repositioning.

      8.1 JD Edwards' (JDE) failed attempt to enter the high-function software markets shows that entry is difficult.

        8.1.1 In the early 1990s, JDE decided to reposition itself to be able to sell to the up- market because it saw substantial revenue opportunities there. Allen, 6/10/04, 770:22-771:10.

        8.1.2 To enter the up-market took JDE roughly six to seven years and approximately $1 billion. It hired hundreds of additional programmers and about 1000 sales and marketing personnel. JDE targeted the employees of SAP, Oracle and PeopleSoft because it believed their experience with up-market customers would help translate JDE's mid-market programs. Allen, 6/10/04, 774:5-776:5.

        8.1.3 As part of its repositioning strategy, JDE had to create a new product architecture that would allow its applications to run on multiple hardware platforms, with multiple databases and multiple operating systems. Allen, 6/10/04, 771:11-21.

        8.1.4 After the toolset was built to support this new product architecture, around 1994, the application programmers were able to start to produce enterprise applications. JDE first built and then released an FMS product for use only on the IBM AS 400 computer in late 1995. Because up-market customers preferred additional hardware platforms, JDE next developed an FMS product for the UNIX and NT platforms in late 1997. JDE then expanded its product offerings to include HR and SCM functionality by the end of 1998. Allen, 6/10/04, 772:19-774:4.

        8.1.5 JDE decided to abandon its attempt to enter the upmarket in 2001 after determining that despite nearly a decade of work and hundreds of millions in expenses, that it "didn't have the products, services, and ultimately reputation necessary to satisfy the requirements of upmarket customers." Allen, 6/10/04, 777:1-13.

        8.1.6 The decision to abandon the up-market was made after careful consideration and after enlisting McKinsey & Co., a consulting firm, to assess its products and performance. Its recommendation - and the recommendation of JDE's internal team - was to return to JDE's roots in the mid-market. Allen, 6/10/04, 777:14-778:5; P4875.

        8.1.7 JDE believes it failed in reaching the up-market, despite all the time and money spent on the effort, because its products lacked the deep, robust feature and function necessary to succeed; lacked the necessary scalability; lacked the reputation and reference accounts it needed; and lacked the broad industry expertise that the up market demanded. Additionally, systems integrators weren't willing to build a practice around JDE unless they believed JDE was going to be successful with larger customers. Allen, 6/10/04, 782:13-784:7; 784:8-786:3.

          8.1.7.1 An example of the missing functionality was the sophisticated software that multinational companies or multidi visional companies require to move data from different divisions and subsidiaries, and to develop a consolidated view of the entire company. Allen, 6/10/04, 784:8-17.

          8.1.7.2 Another example of missing functionality was a global payroll system. Allen, 6/10/04, 784:18-22.

          8.1.7.3 JDE also lacked sophisticated HR functionality, in part based on JDE's historical focus on six asset and manufacturing-intensive industries. Allen, 6/10/04, 784:8-785:7.

          8.1.7.4 JDE also had only a very, very basic treasury system, and relied on a third party to provide the bulk of that functionality. Allen, 6/10/04, 785:8-14.

          8.1.7.5 Similarly, JDE's reporting capabilities were inadequate for up-market customers and JDE relied on a third party to provide those functions. Allen, 6/10/04, 785:15-19.

        8.1.8 Before abandoning the effort, JDE calculated that to truly bring a competitive product to the up-market would have required an additional three to five years of work and a "boatload" of money. Allen, 6/10/04, 786:4-787:2 (emphasis added).

        8.1.9 JDE was unable to reposition itself into the up-market despite the fact that it was the fourth largest EAS provider with 5500-6000 customers, 5000 employees, 1500 software engineers/programmers, and nearly $1 billion in annual revenue. Allen, 6/10/04, 746:16-747:14.

        8.1.10 JDE did not abandon the effort to reach the up-market because it saw the up-market as saturated or that it would have better revenue opportunities in the mid-market. Mr. Allen testified that he used the word "saturated" to indicate that the "marketplace was saturated with competitors, those being PeopleSoft, Oracle and SAP in this case." Allen, 6/10/04, 814:18-815:19.

      8.2 Oracle's, PeopleSoft's and SAP's entry into and repositioning within the high function software markets has been time-consuming and expensive.

        8.2.1 Oracle's efforts to enter the HRM high-function market and to reposition based on competitive pressures have taken years

          8.2.1.1 Until recently, Oracle was "behind" PeopleSoft in the HRM software market. This was true even though Oracle already had a high-function FMS product. Oracle "played catchup" and it took Oracle "a number of years to actually get enough functionality to be competitive." Oracle now offers enough HRM functionality that it is "very competitive" with PeopleSoft. Jeff Henley, Oracle's CEO, is "convinced" that Oracle currently is "very competitive [with PeopleSoft] across the board." Oracle's software is "equal" to or better than PeopleSoft's offering in most areas of HRM. But it took Oracle "three or four years" to reach the point where it was "starting to be reasonably competitive" with PeopleSoft in HR functionality. P3033, Henley dep., 5/4/04, 223:3-228:3; 228:11-229:23.

          8.2.1.2 Even for established high-function vendors, when one vendor "leap frogs" another in terms of functionality, it takes the trailing vendor years to catch up. Specifically, when SAP leap-frogged Oracle by coming out with its R-3 suite product, it took Oracle a few years to catch up to SAP in terms of developing functionality to serve multi-national customers. P3171, Ellison dep., 1/20/04, 116:22-118:8.

        8.2.2 PeopleSoft's move into FMS was time-consuming and expensive. Even after it already had a high-function HRM product, it took PeopleSoft three years from beginning to work on an FMS product to its initial release. It took five to seven years to have a truly competitive product capable of serving the global requirements of high-function FMS software customers. Bergquist, 6/8/04, 264:4-12.

        8.2.3 Similarly SAP has taken many years to reposition itself in certain areas of the high function software market.

          8.2.3.1 SAP has been unable for some time to reposition to a #1 position in the United States. Elzinga, 6/18/04, 2369:9-21.

          8.2.3.2 SAP traditionally trailed Oracle and PeopleSoft in terms of enterprise application software revenues in the United States prior to Oracle's announcement of its intent to acquire PeopleSoft in June 2003. Knowles, 6/23/04, 2931:3-20.

          8.2.3.3 PeopleSoft estimated that it would take SAP roughly two years to reposition itself to compete effectively in the services sector. Bergquist, 6/8/04, 321:19-22

          8.2.3.4 In addition to the time it would take SAP to alter the product to make it more attractive to a different industry sector, there is also the time it takes to develop reference accounts to allow it to compete effectively - roughly one to three years. Bergquist, 6/8/04, 321:23-322:1; 322:10-19; 322:21-325:6.

          8.2.3.5 SAP testified that a vendor of high function software must account for industry differences which requires a huge capital investment to learn and then transfer that knowledge into automation or into writing code to solve the business process issues facing that industry. P3037, Knowles dep., 5/3/04, 68:16-70:15.

          8.2.3.6 SAP further testified that it can be a long and involved process to successfully enter a new industry vertical.

            8.2.3.6.1 As a specific example, while SAP has a successful adoption rate with the global banking community, it is considerably less successful with banks in the United States. Even though SAP's product has the functionality to support European and other area's banking needs, it does not support the needs of the US banks. In order to develop a product to meet the needs of these banks, SAP had to "make a partnership with Accenture, we had to put additional R&D research dollars to work. We've had to create an additional sales force totally dedicated to that industry in order to get focus and dedicated industry -- people that can speak the language of that industry." Knowles, 6/23/04, 2943:20-2944:20.

            8.2.3.6.2 To serve a new industry, a vendor must learn that industry's regulations and processes so one can build those into the product, and that can be complex and involved. A vendor must also build into its software the core functionality that varies across industries, as well as the industry-specific functionality that varies amongst customers in that industry. Knowles, 6/23/04, 2944:21-2945:22.

            8.2.3.6.3 SAP testified that it would take approximately one to three years for it to develop a complete product for financial services companies. Knowles, 6/23/04, 2945:23-2946:5.

          8.2.3.7 Bass testified that SAP began negotiating with Accenture to enter the banking and insurance industry prior to the announcement of the Oracle takeover bid; therefore, it does not reflect re-positioning in light of the merger. Bass, 6/16/04, 1699:10-16. The announcement of the merger had no impact on SAP's plans. Bass, 6/16/04, 1701:6-12.

      8.3 Microsoft has no plans to enter the high-function HRM or FMS markets, and, even were it to develop such plans, its entry could not be timely or sufficient to alleviate the anticompetitive effects of the merger.

        8.3.1 Microsoft's prior acquisitions involved mid-market products and have not put Microsoft in a position to compete for high function software sales.

          8.3.1.1 Microsoft has a company-wide taxonomy that it uses as a guide for sales, marketing and product development. This taxonomy ranges from Lower Small Business to Global, Strategic and Major (GSM) accounts, and has numerous characteristics for each segment. The characteristics include the number of PCs, number of employees, number of worldwide entities, and average IT spend. [REDACTED TEXT] 54); Burgum, 6/23/04, 2980:4-2985:10. Looking at number of employees alone, the midmarket is defined as businesses with 50-1,000 employees; the Corporate Account market consists of businesses with up to 5,000 employees; and the Global, Major and Strategic space ("GSM") accounts are businesses with above 5,000 employees. P3255, [REDACTED TEXT]

          8.3.1.2 Microsoft did not have any business applications before acquiring Great Plains in 2001. Burgum, 6/23/04, 2975:18-2976:1.

            8.3.1.2.1 At the time it was acquired by Microsoft, Great Plains had two products: Great Plains and Solomon. Burgum, 6/23/04, 2976:2-10.

            8.3.1.2.2 The Solomon product is sold to small and mid-market customers. Burgum, 6/23/04, 2997:25-2998:6.

            8.3.1.2.3 The Great Plains product is primarily mid-market, with some customers that fit the Corporate Account Space (CAS: 1000-5000 employees). Burgum, 6/23/04, 2999:11-19.

            8.3.1.2.4 Microsoft will continue to improve the Great Plains product, but the improvements will not expand the Great Plains product's ability to serve large enterprise customers because Microsoft is not targeting that segment. Burgum, 6/23/04, 3001:17-3002:1.

          8.3.1.3 In 2002, Microsoft acquired Navision, a mirror company to Great Plains in that it developed business applications for mid-market companies in different, mostly European, geographies. Burgum, 6/23/04, 2972:14-2973:9.

            8.3.1.3.1 At the time it was acquired by Microsoft, Navision had two products: Navision and Axapta. Burgum, 6/23/04, 2976:11-21.

            8.3.1.3.2 The Navision product is targeted at the lower and mid-market segments. Burgum, 6/23/04, 2998:7-21.

            8.3.1.3.3 The Axapta product is targeted at the upper mid-market, Corporate Account Space (1000-5000 employees), and divisions or independent subdivisions of large organizations. Burgum, 6/23/04, 3002:2-21.

          8.3.1.4 None of Microsoft's current products target what Microsoft calls the "Global, Strategic and Major" (GSM) accounts. Burgum, 6/23/04, 3023:17-21. Moreover, sales into the Corporate Account Space (CAS), representing those companies between 1,000- 5,000 employees and just below the GSM segment, are opportunistic and only supported by 18 sales representatives. Burgum, 6/23/04, 2990:4-22.

        8.3.2 Rather than move up-market, Microsoft has continued to refocus its efforts on the mid-market.

          8.3.2.1 Microsoft does not view the large enterprise market as a "natural extension of its current business model," and "[r]ather than pushing to move up into the enterprise space, [Microsoft is] looking to move out into a broader range of global markets and industries within the small and mid-sized business organizations and divisions of large organizations." P0839 at MS-OPSUB 000000004412; [REDACTED TEXT]

          8.3.2.2 Rather than expanding to reach the up-market, Microsoft recognizes that even at the mid-market, "our eyes are bigger than our stomach," and has been in a period of boundary-setting and reassessment as to how best to properly target and focus on the mid- market for the next few years. P0831R at MS-OPSUB 000000001416; [REDACTED TEXT]

            8.3.2.2.1 [REDACTED TEXT]

            8.3.2.2.2 [REDACTED TEXT]

            8.3.2.2.3 ; [REDACTED TEXT] Burgum, 6/23/04, 3018:12-3019:10 (Business Desk or Technical Implementation Desk to help "avoid overselling the product and avoid bad customer experiences").

          8.3.3 Microsoft does not view the high function software market as a good fit for its business model.

            8.3.3.1 [REDACTED TEXT]

            8.3.3.2 Microsoft does not have the implementation and consulting capability to serve a large, complex enterprise, and has no intention of building such capability. Burgum, 6/23/04, 2995:3-18.

            8.3.3.3 Microsoft's basic strategy is to obtain license revenue -- not service revenue, as is required in the high function software market. [REDACTED TEXT]

            8.3.3.4 Implementation is usually handled by partners and Microsoft does direct implementation only on a very limited basis. Microsoft has 6000 partners but only 100 individuals in the Microsoft consulting organization. Burgum, 6/23/04, 2992:19 - 2993:21.

            8.3.3.5 Microsoft has no plans to change from its practice of primarily selling its business applications through channel partners rather than a direct sales force. Microsoft's model, company-wide, is to sell indirectly - approximately [RED] of Microsoft's sales are through partners. [REDACTED TEXT]

            8.3.3.6 Microsoft does not wish to be the "prime" (e.g., the direct point of contact for the customer) on any sale to a customer. Mr. Burgum testified that if he "was going to joke," being the "prime" on a sale "would mean that we targeted the wrong customer." While Microsoft had a handful of deals where it was the "prime" in the past year, the goal for the coming year is zero. Burgum, 6/23/04, 2992:3-14.

              8.3.3.6.1 One example of Microsoft being the prime for a recent contract was Esselte.

                8.3.3.6.1.1 Esselte doesn't expect to complete its installation until 2008. D7159, Spund dep., 5/25/04, 62:12-19.

                8.3.3.6.1.2 Esselte describes the implementation as getting off to a "rough start" because Microsoft isn't "staffed up to handle very large world-class implementations in multiple countries. They're just not staffed for it and their - their consulting implementation partners are not either.... They just don't have the - the experience and it shows." D7159, Spund dep., 5/25/04, 167:7-168:4.

                8.3.3.6.1.3 Esselte believes Microsoft was not being an adequate prime contractor, and is currently in a billing dispute with Microsoft concerning bills for unauthorized work. D7159, Spund dep., 5/25/04, 191:19-192:3.

                8.3.3.6.1.4 Mr. Burgum testified that "the value may be negative, because I'm not sure anyone's going to make any money on it." Burgum, 6/23/04, 3100:2-21.

              8.3.3.6.2 Another example of Microsoft priming a contract is CDI.

                8.3.3.6.2.1 CDI wanted to contract directly with Microsoft and would not have purchased Great Plains through one of the Microsoft value-added reseller partners. CDI believed that because of its size and complexity, it did not want to contract with a small-sized partner. D8109, Holland dep., 5/24/04, 130:11-131:9.

                8.3.3.6.2.2 CDI believes Microsoft oversold its project management capability. D8109, Holland dep., 5/24/04, 145:22-146:5.

                8.3.3.6.2.3 Because of missed deadlines and added costs, CDI is a hesitant reference account for Microsoft. D8109, Holland dep., 5/24/04, 144:5-18.

            8.3.3.7 Mr. Burgum testified that being the prime is "a great way for us to lose money." Burgum, 6/23/04, 3019:11-24.

          8.3.4 Microsoft does not intend to enter the high function software market with the aid of "Big 5" systems integrators.

            8.3.4.1 Microsoft does not have a GSM product and does not plan on entering the GSM market alone or through alliances with systems integrators. [REDACTED TEXT]

            8.3.4.2 [REDACTED TEXT]

            8.3.4.3 Microsoft's recent agreement with BearingPoint, under Microsoft's "FastRamp" program, is not aimed at the upmarket, nor does it signify any change in Microsoft's target markets. Burgum, 6/23/04, 3052:24-3053:6; 3054:10-18; [REDACTED TEXT] .

              8.3.4.3.1 FastRamp was a program in the first six months of 2004 that focused on increasing Axapta sales and partner expertise in the United States. It had a small amount of funds, $3-5 million, spread across a number of partners. It does not focus on any other products or any markets outside Axapta's target range. Burgum, 6/23/04, 3054:20-3056:1.

              8.3.4.3.2 [REDACTED TEXT] , thus it will expire before Project Green is even released. [REDACTED TEXT].

          8.3.5 Public statements regarding Microsoft's intent to move up-market are inaccurate and usually refer to its efforts in the CAS space, not an intent to move into the GSM space. Microsoft continues to work to clarify any misconceptions about its plans to target the GSM market. P0835 at MS-OPSUB 479; P0836 at MSOPSUB 4560-61; [REDACTED TEXT]

            8.3.5.1 While widely reported, it is not true that Microsoft intends to enter the high- function software market. "I think people confuse the fact that Microsoft has a large balance sheet with ... the innate ability to buy your way into a space.... If there is a vector... specifically directed from Steve Ballmer, [it] is for us to move more downmarket into higher volume segments." Burgum, 6/23/04, 3026:11-3027:11.

          8.3.6 Microsoft does not intend to reposition its current products to build a product to enter the high function software market.

            8.3.6.1 Microsoft has never considered building a product or changing its product to enter the large enterprise space, due to the gap in functionality, gap in a direct sales force, and gap in consulting services. Burgum, 6/23/04, 3023:17-3024:2.

            8.3.6.2 Microsoft's Project Green will not target the high function software market.

              8.3.6.2.1 Project Green -- a Microsoft R&D product meant to create a single solution to replace Microsoft's mid-market products -- would not be ready for market until 2008, at the earliest. Green Version 1 will have less functionality than Microsoft's current mid-market products. Burgum, 6/23/04, 3056:9-3059:17.

              8.3.6.2.2 Green Version 2 will pursue functionality to try to catch up with Microsoft's current products and is not expected to be released until 2010 or 2011. Burgum, 6/23/04, 3061:7-3062:6.

              8.3.6.2.3 Microsoft's Project Green is aimed at rationalizing Microsoft's four products onto a new code base. That release is dependent upon the release of Longhorn, the next version of the Microsoft operating system. Iansiti, 6/17/04, 2058:25-2059:24.

              8.3.6.2.4 Project Green applications will not be able to serve the needs of complex enterprises that Oracle, PeopleSoft and SAP can serve today. Project Green will serve mid-tier customers, perhaps up to $1 billion in revenues. Iansiti, 6/17/04, 2061:11-24.

              8.3.6.2.5 Even Oracle's expert agrees that Microsoft's new business application product, currently under development with the code name "Project Green," has been underway for a number of years and is still years away from having an actual application to release to customers. Kutnick, 6/30/04, 4194:22-4195:16.

            8.3.6.3 Microsoft is not planning on "pouring a lot of money trying to enter into new markets." In fact, "one of the most widely reported and consistently incorrect view held by analysts and press elsewhere is that there's some, you know, unmovable desire for us to move up market, when, in fact, that never was a strategy of Navision, never a strategy of Great Plains, and just because we're owned by Microsoft doesn't suddenly anoint us with the capability to go in to do that." Burgum, 6/23/04, 3024:23-3025:25.

            8.3.6.4 [REDACTED TEXT]

            8.3.6.5 [REDACTED TEXT]

            8.3.6.6 [REDACTED TEXT]

            8.3.6.7 Mr. Burgum testified that to enter the GSM market would be a "formidable task that would take more money than I would be willing to recommend that Microsoft spend." Burgum, 6/23/04, 3024:3-22.

            8.3.6.8 Even if Microsoft changed its current position and decided to enter the high- function market, its entry would not be timely. Microsoft estimates that it will take them five or six years to replace some of their mid-market capability with the new Project Green. Anytime line for entering the high-function market would be longer than that. Burgum, 6/23/04, 3024:23-3025:8.

          8.3.7 Microsoft's potential purchase of SAP was the only way Microsoft would consider beginning to sell in the high-function software market.

            8.3.7.1 Microsoft began considering a merger with SAP in June of 2003, but discussions concluded with a decision not to go forward in the Spring of 2004. Burgum, 6/23/04, 3027:15-3028:10.

            8.3.7.2. [REDACTED TEXT] Siebel does not compete in HRM or FMS software. Ellison, 6/30/04, 4319:8-10.

            8.3.7.3. [REDACTED TEXT]

            8.3.7.4 [REDACTED TEXT]

            8.3.7.5 [REDACTED TEXT]

            8.3.7.6 In materials prepared for review by the Microsoft Board of Directors dated January 5, 2004, a key strategic rationale for the SAP merger is described as "development of innovative new products that break barriers between end users and backend enterprise data." P2640R at MS-OPSUB 000682: Burgum, 6/23/04, 3045:18-3047:9.

            8.3.7.7 Microsoft also flatly states in this Board presentation that achieving leadership in ERP applications for Fortune 500 companies is "an objective we will not pursue in the absence of this combination." [REDACTED TEXT]

            8.3.7.8 [REDACTED TEXT]

            [REDACTED TEXT]

          8.3.8 Other market participants agree that it will take Microsoft many years to enter the high function software market, if it enters at all.

            8.3.8.1 In an e-mail dated 7/2/03, Rick Bergquist, PeopleSoft's Chief Technology Officer, estimated that it would take until 2006 for Microsoft to re-write Navision and Great Plains into a single product with global capabilities, that it would be 2010 before it had a "solid mid-market product," and 2013-2015 before it could effectively compete for enterprise customers. P0807 (PS-C182036) at 182036.

            8.3.8.2 In 2003, Giga estimated that it would be another 3 to 4 years before Microsoft completes the re-write of Great Plains and Navision and can offer a new product. P4009 (PS-TE2360060-1).

            8.3.8.3 A December 2003 META Group analysis of Microsoft's MBS business and its business applications, titled "Microsoft Business Solutions, Strategic Next Step or Misstep?," concluded that "Microsoft's foray into business applications has been targeted specifically at small and medium businesses, SMBs, businesses with less than $1 billion in annual revenue." P3314. Kutnick, 6/30/04, 4197:10-23.

            8.3.8.4 META Group's assessment of Microsoft's MBS also was that the business was "in fact, a small and unprofitable division of Microsoft with a patchwork product portfolio of four overlapping ERP products...." P3314; Kutnick 6/30/04, 4197:24-4198:6.

            8.3.8.5 Mr. Kutnick's own assessment of Microsoft's applications business, made in a META Group report in May 2002, was that "We do not expect Microsoft to compete in the Global 2000 enterprise market (against SAP, PeopleSoft, et al.) but it will become a strong competitor against the major ERP vendors for mid-range sales, and in instances where Global 2000 organization requires a solution for a division or unit that will not be part of a global ERP deployment." P3315; Kutnick, 6/30/04, 4199:2-4200:9.

            8.3.8.6 Microsoft does not succeed in expanding to the highest reaches of, or in dominating, the business for every product it sells just because it is Microsoft, just because it has a lot of money or because it has a monopoly in desktop operating systems. Microsoft has offered database software in competition with Oracle and IBM for somewhere near 10 years. It took Microsoft between seven and nine years of offering its SQL database product to reach rough parity with Oracle and IBM in share of the database sales. Kutnick, 6/30/04, 4193:20-4194:12.

            8.3.8.7 Mr. Keating testified that Microsoft acquired a number of Tier Two products, including Axapta and Great Plains, and those products do not have the scalability, the functionality, or the capabilities of Oracle, SAP or PeopleSoft in either HR or financial management. Keating, 6/10/04, 888:24-889:6.

            8.3.8.8 Mr. Keating testified that customers view Microsoft's products as focused on the middle market or general business and that Microsoft is also focused on that area. Keating, 6/10/04, 889:7-12.

            8.3.8.9 [REDACTED TEXT]

        8.4 Lawson is not likely to enter the high function software market

          8.4.1 Lawson's sales and marketing efforts over the past several years have been in the vertical areas of health care, retail and public sector. Coughlan, 6/28/04, 3662:17-24.

          8.4.2 In its June 24, 2004 earnings call, Lawson stated that its vertical focus would not change, regardless of the outcome of the antitrust trial in San Francisco. Coughlan, 6/28/04, 3666:17-3667:1.

          8.4.3 Despite a four year effort, the financial services vertical has yet to generate at least five percent of Lawson's overall revenues. Coughlan, 6/28/04, 3668:17-20.

          8.4.4 Financial services does not get any separate research and development or marketing dollars because it is not a delineated vertical within Lawson. Coughlan, 6/28/04, 3677:18-3678:9.

          8.4.5 Lawson's ability to successfully expand beyond its delineated verticals is affected by its lack of industry knowledge. Coughlan, 6/28/04, 3673:10-20; [REDACTED TEXT]

          8.4.6 It would not be a good use of Lawson's resources to compete with Oracle, PeopleSoft and SAP in every vertical or industry, because Lawson is smaller in size and Oracle, PeopleSoft and SAP are embedded in a lot more industries. Coughlan, 6/28/04, 3679:9-15.

          8.4.7 Lawson does not believe it would be a good strategy to compete against Oracle, PeopleSoft and SAP, but rather, it should stay focused on its verticals. Coughlan, 6/28/04, 3679:16-22.

        8.5 Other firms are not likely to enter into the high-function software markets.

          8.5.1 [REDACTED TEXT]

          8.5.2 [REDACTED TEXT]

          8.5.3 According to Oracle's own expert, little or no venture capital investment is being directed at financial management or human resources management applications, with the exception of a few, very narrow peripheral applications. Kutnick, 6/30/04, 4192:18-4193:12.

          8.5.4 Mr. Kutnick further believes that venture capital funding for applications to compete with Oracle or PeopleSoft or SAP are very unusual and unlikely. Kutnick, 6/30/04, 4193:13-18.

          8.5.5 IBM does not intend to enter the high function software market, nor any portion of the applications business.

            8.5.5.1 Currently, IBM does not make any business applications software. P3193, Mills dep., 5/27/04, 24:9-22.

            8.5.5.2 IBM decided in 1998 to focus on middleware technologies and not the applications business, preferring to partner with applications companies rather than compete with applications companies. P3193, Mills dep., 5/27/04, 16:7-15.

            8.5.5.3 The benefits to IBM of partnering its large portfolio of hardware, services, and middleware technologies with applications software gives IBM an enormous financial advantage. P3193, Mills dep., 5/27/04, 16:16-25.

            8.5.5.4 In evaluating its potential response to the proposed Oracle/PeopleSoft merger, IBM reaffirmed its commitment to stay out of the applications business. Concerning making an acquisition itself, IBM "talked about not doing that. We talked about how our strategy was still sound, that not being in the applications business continued to make sense, that this acquisition [Oracle/PeopleSoft] was not going to change our strategy in the marketplace because it didn't in any fundamental way upset the economic value proposition that IBM had been benefitting from in choosing to partner with application vendors." P3193, Mills dep., 5/27/04, 112:19-113:8. No specific acquisitions were suggested by the staff. P3193, Mills dep., 5/27/04, 114:2-13. Moreover, Mills testified that instead of considering an applications vendor purchase, "all the conversations I've participated in and all the conversations I know of actually were to the contrary." P3193, Mills dep., 5/27/04, 117:23-118:7.

            8.5.5.5 Further, the 1998 declaration by IBM that it would not compete in the applications business is "something we always return to as a grounding principle, and there are some who at times will come forward and say, well, shall we rethink that and the consistent answer is no." P3193, Mills dep., 5/27/04, 113:16-25. Mills testified that "since 1998 we've been absolute in our position on this. Since I am the guy who leads the software business, I've refused to look at any and all application company evaluations of purchase." P3193, Mills dep., 5/27/04, 118:23-119:4.

        8.6 Further evolution of the "infrastructure layer" will not make new entry into the high-function applications markets timely, likely, or sufficient.

          8.6.1 Whatever further evolution of middleware or the "infrastructure layer" occurs, those middleware elements cannot and will not replace the application software that automates the business knowledge and the business processes in an application.

          8.6.2 Infrastructure products like SAP's NetWeaver and Peoplesoft's AppConnect are not substitutes for HRM or FMS applications. Instead, they are complements to applications which help to integrate applications but which have no capability of their own to process information or transactions. They do not replace the transaction processing engine of the application. Wilmington, 6/16/04, 1813:18-1814:20. See also P3193, Mills dep., 5/27/04, 17:21-18:13 ("Middleware doesn't provide the application function. Middleware is not about the ERP function. It's not about the accounts receivable, accounts payable. Middleware is merely about how you get things connected together.")

          8.6.3 The HRM and FMS applications reflect and automate the business process and functionality, for example, hiring an employee, terminating an employee, changing jobs, etc. The infrastructure layer "doesn't do the business logic. It provides services and capabilities that are made use of by that enterprise software." Bergquist, 6/8/04, 331:22-332:13.

          8.6.4 The bulk of what is being done today in high-function applications is the business logic that reflects an understanding of the processes within a business, the industry specialization, the global specialization, etc. Bergquist, 6/8/04, 333:8-15

          8.6.5 Customers "absolutely" will still require an application separate from any "application services" to provide the actual business functions and processes provided today by applications. Kutnick, 6/30/04, 4187:25-4188:8.

          8.6.6 For example, many of Bearing Point's clients use integration products such as NetWeaver to connect their core FMS and HR systems with other point solutions, but still must rely on applications from Oracle, PeopleSoft, and SAP for their core FMS and HR support. Keating, 6/10/04, 995:4-9.

          8.6.7 Even if infrastructure were to evolve to be much more capable in the future, application developers would still have to understand the details of the underlying business processes they were attempting to automate and then write actual software code that makes the application perform the relevant business process. Kutnick, 6/30/04, 4189:5-16.

          8.6.8 As Dr. Iansiti described the difference, "FMS and HRMS applications distinguish themselves because of the business processes that they enable, and that knowledge has nothing to do with some technology that sits underneath the stack. They are no substitute for each other. There's a richness of features and functionality that cannot be easily substituted by some new player coming along.... [Y]ou can make it incrementally easier to write an application. It's easier now to write an application than it was five years ago or ten years ago. What's not easy is understanding what the large organizations want and what's not easy is to take that understanding and translate into code. The technology itself makes it much more easy to do this, but it doesn't create a substitute for what weVe been talking about." Iansiti, 6/17/04, 2076:13-2077:11.

          8.6.9 Thus, all of the complexity that exists today in developing the portion of an application that automates the underlying business processes will not be replaced or eliminated by greater infrastructure or integration capability of middleware. Kutnick, 6/30/04, 4189:17-21.

          8.6.10 As a result, entry into enterprise applications will remain formidable no matter what developments occur in the infrastructure layer. Bergquist, 6/8/04, 332:24-333:15 (new entrants "would still face a formidable undertaking to produce the same functionality that is provided by PeopleSoft in these areas.").

          8.6.11 While Dr. Teece asserts that the infrastructure layer will make best of breed applications more successful, he has not done and is not aware of any quantitative analysis assessing the extent of any such impact on entry cost or whether that impact would be sufficient to allow additional firms to compete in high-function human resource and financial management software markets. Teece, 7/1/04, 4447:6-7.

          8.6.12 Oracle continues to believe that one of the value propositions of its eBusiness Suite is that it offers a wide range of integrated functionality. P3044, Wohl dep., 4/29/04, 226:22-24.

          8.6.13 Oracle's strategy of offering infrastructure technology that makes it easier to integrate applications from various vendors together does not undermine the value of Oracle's integrated business suite by making it easier for best of breed vendors to replace Oracle's applications. P3044, Wohl dep., 4/29/04, 226:25-227:3 ("Q [I]f you offer technology for integrating with other functionality, do you undercut the value of your integrated functionality? A Not in the least.").

          8.6.14 The Oracle also asserts that the evolution of the infrastructure layer enables the entry of so-called "composite applications." Teece, 7/1/04, 4452:9-15.

            8.6.14.1 Composite applications represent an evolution of the business functionality provided by the large enterprise applications vendors. They provide a variety of functionality that can work across traditional applications such as FMS and HR, automating many of the tasks that users formerly had to do manually. These composite applications are bringing all that content together, so people don't have to act as the glue to composite things together in a manual fashion. Bergquist, 6/8/04, 325:14-327:10; 328:23-329:7.

            8.6.14.2 The continued development of composite applications will not make it significantly easier for new entrants or smaller vendors to develop applications that have the same capability and functional as high-function applications offered by Oracle, PeopleSoft, and SAP for complex customers. Bergquist, 6/8/04, 329:8-13.

            8.6.14.3 Because composite applications will merely augment, rather than replace, HRM and FMS applications, any new HRM or FMS entrant still would have to match the formal capabilities of those applications in terms of functionality, global scalability, and their other dimensions. Bergquist, 6/8/04, 328:4-329:20.

            8.6.14.4 The idea of composite applications is not new; they have been evolving over the last three years. Bergquist, 6/8/04, 327:19-328:3.

            8.6.14.5 Composite applications remain difficult to develop. Teece, 7/1/04, 4448:1-21; P3339 (Gartner research report noting that, "[d]espite advances in technology, composite applications are difficult to develop.").

            8.6.14.6 Composite applications are meant to be complements to, not substitutes for, HRM and FMS applications software. P3342 at page 21; Bergquist, 6/8/04, 329:8-20. "Composite characteristics" need to be added into the business processes involved in HRM and FMS applications. Bergquist, 6/8/04, 327:11-18.

            8.6.14.7 Even where composite applications are developed, the core functionality performed by HRM or FMS applications still needs to be performed by those applications. Bergquist, 6/8/04, 328:15-20.

            8.6.14.8 Rather than replacing or eliminating the need for high-function HRM or FMS applications offered by Oracle and PeopleSoft, composite applications instead will augment the functionality of those traditional applications to make users more effective and efficient. Bergquist, 6/8/04, 329:8-20 ("This is an augmentation process, not a replacement process.").

            8.6.14.9 While certain PeopleSoft documents from early 2003 indicate initial concern over whether resources would be available to enable PeopleSoft to develop and sell composite applications, PeopleSoft in fact has developed and is shipping applications that have varying degrees of composite characteristics. Bergquist, 6/8/04, 324:20-325:14.

        8.7 Expert testimony supports the conclusion that entry or repositioning is unlikely to stem the anticompetitive effects of the proposed merger

          8.7.1 Dr. Elzinga concluded that the adverse competitive effects of this merger won't be prevented or undercut by the entry of new firms or the expansion of existing firms. Elzinga, 6/18/04, 2160:12-18.

          8.7.2 When he first started studying the proposed merger, Dr. Elzinga thought entry might come from mid-market vendors moving into high-function software. But based on his research, Dr. Elzinga determined that was not likely to happen. Instead, the high-function vendors, Oracle, PeopleSoft and SAP are able to strip-out functionality and move down into the mid-market, while the mid-market vendors are unsuccessful at moving up into high-function software. Elzinga, 6/18/04, 2239:23-2241:12.

          8.7.3 JD Edwards failed attempt to enter the up market confirms that enterprise customers are not likely to be protected by entry from the mid-market. Elzinga, 6/18/04, 2240:16-2241:12.

          8.7.4 Dr. Elzinga testified that he did not believe Microsoft was positioned to constrain the pricing power of a combined Oracle/PeopleSoft. He based this conclusion on Microsoft's testimony and documentary evidence in this case, as well as his own prior experience consulting for Microsoft. Dr. Elzinga believes Microsoft's business model is one of high volume, low price, and low customer service. What "they're really good at is writing code and selling a product that is high volume, thousands and thousands, millions of units. They do it at a low margin, and they don't offer a lot of hand-holding and customer service." Elzinga, 6/18/04, 2242:20-2243:23.

          8.7.5 Dr. Elzinga believes that Oracle, PeopleSoft and SAP take the opposite approach. "These companies have large, skilled, dedicated sales forces. They spend months with a prospective customer, a customer they might not even gain.... [T]hen once the deal is cut and the customer chooses one of these three vendors, there's a great deal of service that's spent with them... It's a very personal business, even though it's high-tech. It is definitely not high volume." Elzinga, 6/18/04, 2243:24-2244:15.

      9 Anticompetitive Effects - Efficiencies. Oracle cannot demonstrate that verifiable, merger- specific efficiencies will reverse the anticompetitive effects of the merger.

        9.1 Oracle's alleged "efficiencies" defense is not based on an actual efficiency, i.e., "something that expands output." Rather, "[i]f somebody cuts out a product line and reduces output, we don't think of that as a social efficiency." Elzinga, 6/18/04, 2161:1-2162:1.

        9.2 Oracle's efficiencies claims are not verifiable.

          9.2.1 Oracle's efficiencies claims are not verifiable because they lack any supporting facts, documents or analyses. Zmijewski, 7/1/04, 4519:19-4520:17.

          9.2.2 Oracle's efficiencies claim is based on a one-year old spreadsheet from July 2003 ("July 2003 model"). Catz, 6/25/04, 3467:19-3471:20; P4184R; P3004R; D5613R; Zmijewski, 7/1/04, 4519:19-4520:23, 4530:13-22, 4532:14-4533:7, 4533:8-25, 4538:6-21.

          9.2.3 The calculation estimates cost savings in three areas of operation, Sales and Marketing ("S&M"), Research and Development ("R&D"), and General and Administration ("G&A"). D7132; P3357 at slide 6; Zmijewski, 7/1/04, 4523:21-4524:6.

          9.2.4 Oracle calculated the alleged efficiencies by subtracting S&M, R&D, and G&A cost estimates for fiscal year 2005 ("FY05") operations from PeopleSoft's and J.D. Edwards's fiscal year 2003 ("FY03") costs for S&M, R&D, and G&A. D7132; P3357 at slide 6; P3004 at 34 and 41; Zmijewski, 7/1/04, 4523:21-4524:6.

          9.2.5 The calculation is done at the aggregate level; there is no breakdown of the alleged S&M, R&D, and G&A cost savings categories by product, service, division or operating group. D7132; P3004R at 34 and 41.

          9.2.6 The FY05 cost estimates for the three areas of operation are what "drive" the alleged cost savings claim. Zmijewski, 7/1/04, 4524:14-23.

          9.2.7 The entire calculation of the FY05 cost estimates is contained in only two columns of the July 2003 model. P3357 at slides 7 and 8; P3004R at 34 (elm. R) and 41 (elm. P); Zmijewski, 7/1/04, 4525:12-4526:17; 4527:11-4528:13.

          9.2.8 The numbers that determine the amount of the FY05 cost estimates ("inputs") are simply typed-in numbers - i.e., "hard-coded" inputs. P3004R at 34 (elm. R), and 41 (elm. P); P3357 at slides 7 and 8; Zmijewski, 7/1/04, 4526:18-4527:3, 4529:15-4530:12.

          9.2.9 There are no factual bases for the FY05 cost estimates.

            9.2.9.1 There are no factual bases, such as supporting documentation, data, analyses or citations, for any of the inputs in the July 2003 model that determine the amount of the FY05 cost estimates. Zmijewski, 7/1/04, 4519:19-4520:23, 4530:13-22, 4532:14-4533:7, 4533:8-25, 4538:6-21.

            9.2.9.2 There are no factual bases, such as supporting documentation, data, analyses or citations, for the inputs used to determine the S&M component of Oracle's FY05 cost estimates. Zmijewski, 7/1/04, 4533:8-25, 4538:6-21.

            9.2.9.3 There are no factual bases, such as supporting documentation, data, analyses or citations, for the inputs used to determine the R&D component of Oracle's FY05 cost estimates. Zmijewski, 7/1/04, 4530:13-22, 4538:6-21.

              9.2.9.3.1 In late April 2004, eight months after the July 2003 model was prepared, Ron Wohl, Oracle's head applications developer had a conversation with Larry Ellison, in which Mr. Ellison asked Wohl for his "speculation as to what the right number" would be for Oracle's R&D budget if Oracle acquired PeopleSoft. P3067, Wohl dep., 4/30/04, 484:1-20.

              9.2.9.3.2 Wohl testified that during the conversation, Wohl and Ellison discussed the level of R&D staffing needed to maintain the PeopleSoft products and Wohl made "just a kind of wild guesstimate." P3067, Wohl dep., 04/30/04, 485:21-486:7. Wohl testified that at this point he had not done any formal analysis and that "Larry and I were speculating together, he had just called me and asked me for an off-the-cuff guess ..." P3067, Wohl dep., 4/30/04, 487:1-3.

              9.2.9.3.3 Wohl testified that he and Mr. Ellison discussed the number of people needed to add new features to the current generation of PeopleSoft products and that he believed "that number by far is the most speculative number because we don't have a good way to gauge that without having any idea whatsoever what expectations PeopleSoft has set for their current customer base." P3067, Wohl dep. 4/30/04, 485:21-486:25. Wohl further testified "[s]o the enhancements that we would add. That one was the most speculative of numbers. And Larry asked me what I thought, and I told him literally I don't know because I know literally nothing about the expectations." P3067, Wohl dep., 4/30/04, 487:1-13.

              9.2.9.3.4 Wohl testified that as of the time of his late April 2004 conversation with Larry Ellison he was not aware of the planned reductions in R&D headcount. P3067, Wohl dep., 4/30/04, 488:3-17.

              9.2.9.3.5 Wohl testified that plans to retain only 50% of the PeopleSoft R&D staff were "pure speculation, the number would be at least 50 percent but could well be greater depending on what we found after we were able to get into detail about the nature of the expectations that PeopleSoft had set on behalf of their existing customers." P3067, Wohl dep. 4/30/04, 491:8-492:20; 487:14-18.

            9.2.9.4 There is no factual bases, such as supporting documentation, data, analyses or citations, for the inputs used to determine the G&A component of Oracle's FY05 cost estimates. Zmijewski, 7/1/04, 4532:14-4533:7, 4538:6-21.

          9.2.10 Ms. Catz testified that there are no supporting documents, cites, calculations or back-up materials for the inputs in the July 2003 spreadsheet.

            9.2.10.1 Ms. Catz provided the inputs for the July 2003 model, including the inputs used to calculate the FY05 cost estimates. Catz, 6/25/04, 3470:3-8.

            9.2.10.2 Ms. Catz testified that the July 2003 model does not explain how any of the inputs were achieved or developed. Catz, 6/25/04, 3558:9-18.

            9.2.10.3 Ms. Catz testified that the July 2003 model does not contain any citations for any facts or assumptions that she relied in determining the inputs. Catz, 6/25/04, 3560:10-13.

            9.2.10.4 Ms. Catz testified that there is no information in the July 2003 model that enables reconstructing of how the cost estimates were determined. Catz, 6/25/04, 3557:7-9.

            9.2.10.5 Ms. Catz testified that there is "no document that [she] know[s] of that explains how the inputs were developed. Catz, 6/25/04, 3558:16-21.

            9.2.10.6 Ms. Catz testified that the analysis for some of the inputs in the July 2003 model were determined by herself and Larry Ellison by sitting around and thinking about what sounded like a good number to plug into the model. Catz, 6/25/04, 3557:15-20.

            9.2.10.7 Ms. Catz testified that there is no back-up or underlying spreadsheets used to compute the inputs. Catz, 6/25/04, 3557:21-3558:8.

            9.2.10.8 Ms. Catz testified that, when she was working on the financial models, she did not consult with Ron Wohl, Oracle's head applications developer, concerning how he would allocate R&D resources if Oracle acquires PeopleSoft. P3348, Catz dep., 5/18/04, 147:13-20.

          9.2.11 Oracle's investment banker, CSFB, does not know the facts or assumptions relied on by Oracle to determine many of the inputs for the July 2003 model.

            9.2.11.1 CSFB's 30(b)(6) witness on the July 2003 model testified that CSFB received many of the inputs for the July 2003 model from "Oracle management." P3346, Scarborough dep., 6/3/04, 102:1-12, 104:16-19, 105:16-108:2, 111:7-112:15, 116:9-117:6, 130:14-131:8, 136:11-138:16, 150:11-151:19, 156:15-158:10, 193:21-195:13, 206:6-207:7, 208:7-211:6.

            9.2.11.2 CSFB's 30(b)(6) witness on the July 2003 model testified that he was not aware of the "facts" or "assumptions" relied on by Oracle management in determining many of the hard-coded inputs in the July 2003 model. P3346, Scarborough dep., 6/3/04, 91:10-92:13, 102:1-12, 104:16-19, 105:16-108:2, 111:7-112:15, 116:9-117:6, 136:11-138:16, 150:11-151:19, 154:20-155:8, 156:15-158:10, 193:21-195:13, 206:6-207:7, 208:7-211:6.

          9.2.12 Oracle could have provided the Court with supporting facts for its efficiencies claims.

            9.2.12.1 "Oracle could have presented a lot of information, different types of sources of information, and conducted analyses to provide factual support for those numbers" in the July 2003 model. Zmijewski, 7/1/04, 4581:19-4582:16.

            9.2.12.2 Because "the cost structures and the relation between various levels of activity and output and cost should be at least generally the same for J.D. Edwards and Oracle and PeopleSoft...a lot of that information could have been used to measure the assumptions in [the July 2003 model's] calculations. Zmijewski, 7/1/04, 4534:23-4535:24.

            9.2.12.3 Given that Oracle and PeopleSoft are similar organizations, there is information concerning its S&M operations that "would be very relevant" to analyzing Oracle's cost savings claims with respect to S&M. Zmijewski, 7/1/04, 4534:23-4535:24.

            9.2.12.4 Given that Oracle and PeopleSoft are similar organizations, "there is a lot of that information could have been used to measure the assumptions in the [R&D cost] calculations." Zmijewski, 7/1/04, 4534:23-4535:24. For example, "someone could have done, at least, is go back to the Oracle documents, use Oracle documents, by product line, within research and development, and see how much -- how many people they use for different tasks...so that [one] actually saw the cost structure and the organization structure within research and development....." Zmijewski, 7/1/04, 4530:23-4531:24.

          9.2.13 Oracle and CSFB believe that Oracle's cost estimates will become less accurate over time.

            9.2.13.1 The July 2003 model contains S&M, R&D and G&A costs estimates for FY05 through FY08. P3004R, at 34 and 41.

            9.2.13.2 Ms. Catz testified, using a demonstrative exhibit prepared for this trial, rather than a contemporaneous business document, about the purported net present value of the alleged efficiencies amount for a five-year period, from FY05 through FY9. Catz, 6/25/04, 3493:11-3494:4.

            9.2.13.3 CSFB's 30(b)(6) witness on the July 2003 model testified that he had "received requests from people to get rid of '07 and '08" estimates because they were "extraneous information," that "no one, you know, believed in" because it was "[l]ess reliable, too far out in the future to be really predictive." P3346, Scarborough dep., 6/3/04, 225:18-227:10.

            9.2.13.4 CSFB's 30(b)(6) efficiencies witness testified that it is "very, very difficult" to do accurate projections for fiscal year 2007 and further that his "take" on financial models, such as the July 2003 model, is that "the only thing [he] knows for sure...is it's not going to be accurate." P3346, Scarborough dep., 6/3/04, 228:16-230:8.

            9.2.13.5 Ms. Catz testified that the "further out you go, the less likely" that the estimates in the July 2003 model will be accurate and that "FY08 is quite a long time from now, a lot of things could change by then." P3348, Catz dep., 5/20/04, 413:14-415:16.

        9.3 Oracle used the wrong cost baseline in computing the alleged efficiencies.

          9.3.1 Oracle computed the FY03 cost baseline for its efficiencies calculation by using separate cost structures from PeopleSoft and J.D. Edwards that existed prior to the merger of the two companies. Zmijewski, 7/1/04, 4520:24-4521:14, 4538:22-4540:5.

          9.3.2 Because Oracle used a cost baseline for FY03 that existed prior to the merger of PeopleSoft and J.D. Edwards, Oracle failed to subtract from its efficiencies claim potential efficiencies from the PeopleSoft/J.D. Edwards transaction. Zmijewski, 7/1/04, 4520:24-4521:14, 4538:22-4540:5; Catz, 6/25/05, 3494:5-13.

          9.3.3 Oracle has not attempted to estimate the potential efficiencies that may be achieved from the PeopleSoft/J.D. Edwards merger. Catz, 6/25/05, 3494:5-13.

        9.4 Oracle has not shown that the alleged cost savings efficiencies are independent of the substantial reductions in output projected in the July 2003 model.

          9.4.1 "[S]omething's economically efficient if you can produce the same rate of output using fewer inputs. Or the kind of flip side of that is, something becomes more efficient if you can take the same number of inputs and you can somehow squeeze out of those inputs a larger output or -- if not larger in terms of quantity, larger in terms of value or quality." Elzinga, 6/18/04, 2161:1-14.

          9.4.2 "Efficiencies means to use your resources more -- more effectively than you did before. To do that, it can't be the result of just producing less. So, if you produce less your costs should change. They should go down. But that's a reduction that's the result of volume." Zmijewski, 7/1/04, 4540:22-4541:5.

          9.4.3 "[I]f a cost reduction is related to change in output...[a]ccountants call that a volume variance, they don't call it an efficiency...." Zmijewski, 7/1/04, 4521:24-4522:14.

          9.4.4 "If somebody cuts out a product line and reduces output, [economists] don't think of that as a social efficiency. It might be a cost saying (sic) for the company." Elzinga, 6/18/04, 2161:15-2162:1.

          9.4.5 The July 2003 model predicts substantial declines in output for the combined Oracle/PeopleSoft firm relative to the combined output of Oracle and PeopleSoft as stand alone companies. P3357 at slide 24, P3004R at 31 and 38, P0901 at 41 (Ins. 70-73. clm. N).

          9.4.6 Oracle has not provided any factual bases, including documents, data, analyses or citations, demonstrating that the alleged efficiencies are independent of the July 2003 model's projected substantial declines in output of the combined Oracle/PeopleSoft firm. Zmijewski, 7/1/04, 4521:15-23, 4554:7-19.

          9.4.7 The July 2003 model projects declines in new customers of PeopleSoft and J.D. Edwards software.

            9.4.7.1 The July 2003 model projects that from FY03 to FY05 the number of new customers of PeopleSoft and J.D. Edwards software declines from 560 to 0 (100%). P3357 at slides 16 and 24; P3004R at 31 (ln. 15) and 38 (ln. 15); P0901 at 41 (ln. 70 clm. N).

            9.4.7.2 The July 2003 model projects that from FY03 to FY05 the number of new customers of PeopleSoft software declines from 348 to 0 (100%). P3357 at slides 24 and 27; P3004R at 31 (ln. 15).

            9.4.7.3 The July 2003 model projects that from FY03 to FY05 the number of new customers of J.D. Edwards software declines from 212 to 0 (100%). P3357 at slides 24 and 34; P3004R at 38 (ln. 15); P0901 at 41 (ln. 70 clm. N).

          9.4.8 The July 2003 model projects declines in existing customers of PeopleSoft and J.D. Edwards software.

            9.4.8.1 The July 2003 model projects substantial declines in the number of existing customers that use PeopleSoft and J.D. Edwards software. From FY03 to FY05, the number of users of PeopleSoft and J.D. Edwards software declines from 9,687 to 8,063 (17%), and declines to 4,930 (49%) by FY08. P3357 at slide 17; P3004R at 31 (ln. 24) and 38 (ln. 24).

            9.4.8.2 The July 2003 model projects substantial declines in the number of existing customers that use PeopleSoft software. From FY03 to FY05, the number of users of PeopleSoft software declines from 4,997 to 4,103 (18%), and declines to 2,370 (53%) by FY08. P3357 at slides 24 and 28; P3004R at 31 (ln. 24).

            9.4.8.3 The July 2003 model projects substantial declines in the number of existing customers that use J.D. Edwards software. From FY03 to FY05, the number of users of J.D. Edwards software declines from 4,690 to 3,960 (16%), and declines to 2,560 (45%) by FY08. P3357 at slide 35; P3004R at 31 (ln. 24).

          9.4.9 The July 2003 model projects declines in the total size of the current PeopleSoft and J.D. Edwards customer base (including "Oracle migrations"), and thus of the total customer base of the merged firm.

            9.4.9.1 The July 2003 model projects that some of the current PeopleSoft and J.D. Edwards customers will "migrate" to Oracle software products ("Oracle migrating customers"). P3004R at 31 (ln. 42) and 38 (ln. 42).

            9.4.9.2 Including the Oracle migrating customers, the July 2003 model still projects substantial declines in the size of the PeopleSoft and J.D. Edwards customer base ("Total PS/JDE customers") from pre-merger levels. From FY03 to FY05, the number of Total PS/JDE customers declines from 9,687 to 8,664 (11%), and declines to 7,719 (20%) by FY08. P3357 at slides 17, 24 and 25; P3004R at 31 (ln. 21) and 38 (ln. 21); Zmijewski, 7/1/04: 4543:22-4546:4.

            9.4.9.3 The July 2003 model projects substantial declines in the size of the PeopleSoft customer base ("Total PS customers") from pre-merger levels, even when the Oracle migrating customers are included. From FY03 to FY05, the number of Total PS customers declines from 4,997 to 4, 364 (13%) and declines to 3,829 (23%) by FY08. P3357 at slides 24, 25 and 28; P3004R at 31 (ln. 21).

            9.4.9.4 The July 2003 model projects substantial declines in the size of the J.D. Edwards customer base ("Total JDE customers") from pre-merger levels, even when the Oracle migrating customers are included. From FY03 to FY05, the number of Total JDE customers declines from 4,690 to 4,300 (8%) and declines to 3,890 (17%) by FY08. P3357 at slides 24, 25 and 35; P3004R at 38 (ln. 21).

          9.4.10 The July 2003 model projects declines in total licenses purchased.

            9.4.10.1 The July 2003 model projects substantial declines in the number of licenses purchased by the PeopleSoft and J.D. Edwards customer base, even including the projected number of licences purchased by customers that "migrate" to Oracle products ("Total Licenses"). From FY03 to FY05, Total Licenses purchased by PeopleSoft and J.D. Edwards customers declines from 2,319 to 1,026 (56%), and declines to 908 (61%) by FY08. P3357 at slides 18, 24 and 25; P3004R at 31 (ln. 18) and 38 (ln. 18); P0901 at 41 (ln. 73, elm. N); Zmijewski, 7/1/04: 4547:1-22.

            9.4.10.2 From FY03 to FY05, Total Licenses purchased by PeopleSoft customers declines from 1, 253 to 411 (67%), and declines to 458 (63%) by FY08. P3357 at slides 24, 25 and 29; P3004R at 31(ln. 18).

            9.4.10.3 From FY03 to FY05, Total Licenses purchased by J.D. Edwards customers declines from 1,066 to 615 (42%), and declines to 450 (58%) by FY08. P3357 at slides 24, 25 and 36; P3004R at 38 (ln. 18); P0901 at 41 (ln. 73, clm. N).

          9.4.11 The July 2003 model projects substantial reductions in employee headcount, which according to CSFB, will come mostly from PeopleSoft.

            9.4.11.1 The July 2003 model projects substantial reductions in the total employee headcount of a combined Oracle/PeopleSoft company relative to the pre-merger headcount levels. From FY03 to FY05, headcount declines by 10,055 (76% of the total PS/JDE 13,267 pre-merger headcount). P3357 at slide 21; P3004R at 34 (ln. 208) and 41 (ln. 203); Zmijewski, 7/1/04, 4549:21-4550:23.

            9.4.11.2 The July 2003 model projects substantial reductions in the S&M headcount of a combined Oracle/PeopleSoft company relative to the pre-merger headcount levels. From FY03 to FY05, sales quota representative headcount declines by 635 (89% the total PS/JDE 710 pre-merger headcount). From FY03 to FY05, other sales and marketing personnel headcount declines by 1,874 (99% of the total PS/JDE 1,891 pre-merger headcount, respectively). P3004R at 34 (lns. 176, 180) and 41 (lns. 170, 174).

            9.4.11.3 The July 2003 model projects substantial reductions in the R&D headcount of a combined Oracle/PeopleSoft company relative to the pre-merger headcount levels. From FY03 to FY05, headcount declines by 2,077 (57% of the total PS/JDE 3,652 pre- merger headcount). P3004R at 34 (ln. 168) and p. 41 (ln. 162).

            9.4.11.4 The July 2003 model projects substantial reductions in the G&A headcounts of a combined Oracle/PeopleSoft company relative to the pre-merger headcount levels. From FY03 to FY05, headcount declines by 1,342 (87% of the total PS/JDE 1,537 pre- merger headcount). P3004R at 34 (ln. 193) and 41 (ln. 188).

            9.4.11.5 CSFB's 30(b)(6) witness on the July 2003 model testified that Oracle management told him that "most of the employees that would be impacted would be from the PeopleSoft side." P3346, Scarborough dep., 6/3/04, 79:2-21.

          9.4.12 Oracle's estimated declines in costs track the estimated declines in output.

            9.4.12.1 The July 2003 model projects declines in output (customers and licenses) that track the estimated declines in cost and headcount. P3357 at slides 23-25, 33, 40; P3004R at 31-34 and 38-41; P0901 at 41-42.

            9.4.12.2 Oracle intends to stop all active marketing and selling of PeopleSoft and J.D. Edwards' products. Catz, 6/25/04, 3541:18-25; 3544:13-15; 3546:4-3547:3.

            9.4.12.3 Oracle will not develop a next generation of PeopleSoft and J.D. Edwards' products. P4327 at 3; Wohl, 6/25/04, 3419:15-3420:13; Catz, 6/25/04, 3527:20-3528:9.

            9.4.12.4 There are no factual bases, including documents, data, analyses or citations, for Ms. Catz's testimony that the output estimates in the July 2003 model are worst case scenarios. Zmijewski, 7/1/04: 4537:8-4538:5.

            9.4.12.5 Oracle has not presented any factual bases, including any documents, data, analyses or citations, showing that the July 2003 model's estimated declines in costs are unrelated to the substantial projected declines in customers and license output. Zmijewski, 7/1/04, 4521:15-23, 4554:7-4555:1.

          9.4.13 S&M, R&D and G&A are not always fixed costs

            9.4.13.1 S&M costs are not always fixed. Coughlan, 6/28/04, 3687:8-15.

            9.4.13.2 R&D costs are not always fixed. Zmijewski, 7/1/04, 4552:4-4553:9.

            9.4.13.3 G&A costs are not always fixed because "as the company grows, G&A costs go up...there is a correlation between G&A costs and company costs." Zmijewski, 7/1/04, 4553:10-4554:6.

            9.4.13.4 Lawson's President testified that R&D, S&M, and G&A costs are related to changes in revenues. Coughlan, 6/28/04, 3685:11-16, 3687:8-20.

              9.4.13.4.1 Lawson's President testified that Lawson tries to keep "R&D anywhere from 15 to 18 percent of sales." Coughlan, 6/28/04, 3685:11-16.

              9.4.13.4.2 Lawson's President testified that "you're either adding or reducing a sales head count, based on the activity you see." Coughlan, 6/28/04, 3687:8-15.

              9.4.13.4.3 Lawson's President testified that "G&A usually runs 5 to 7 percent" of sales. Coughlan, 6/28/04, 3687:16-20.

          9.4.14 Oracle could have provided the Court with supporting facts and analyses for its claim that its S&M, R&D and G&A costs are fixed.

            9.4.14.1 Oracle could have used its own data to demonstrate that its alleged cost reductions in S&M, R&D and G&A are not related to the July 2003 model's projected substantial declines in output. Zmijewski, 7/1/04, 4554:20-4555:1.

            9.4.14.2 Oracle's data could have shown the "relations between different levels of activity and these different type of expenses [because t]hat's an empirical question you can answer using data and standard analyses." Zmijewski, 7/1/04, 4554:20-4555:1.

        9.5 No third party verified Oracle's efficiencies claims

          9.5.1 None of Oracle's four experts provided any input into or verified Oracle's alleged efficiencies claims. Catz, 6/25/04, 3564:19-3565:4; Zmijewski, 7/1/04: 4518:24-4519:9.

          9.5.2 Oracle has not requested that CSFB provide it with a "fairness opinion" with respect to Oracle's attempt to acquire PeopleSoft or to advise Oracle whether the tender offer price is fair or reasonable. P3347, Armstrong dep., 5/13/04, 11:5-18.

        9.6 Oracle's efficiencies claims do not specifically account for substantial merger integration costs or potential integration risks

          9.6.1 The July 2003 model estimates $1.293 billion of integration costs and $92 million in additional fees and expenses. P3004R at 16 (lns. 52 and 53).

          9.6.2 Oracle did not deduct these approximately $1.385 billion in integration and additional fees and expenses from the alleged efficiencies amount. D7132; P3357 at slide 6; P3004R at 34 and 41.

          9.6.3 The July 2003 model does not contain any explicit calculation or analysis of the potential integration risks involved in achieving the alleged efficiencies. P3004R. Oracle's investment banker, CSFB, believes integration risks exist in mergers, including the "multitude of reasons that meshing two companies together can be difficult." P3346, Scarborough dep., 6/3/04, 58:12-18, 58:18-62:2; 65:5-66:3; 70:7-15; P3004R.

          9.6.4 Oracle's analysis of efficiencies does not consider certain inefficient costs that Oracle's actions would impose on others, mainly customers.

            9.6.4.1 PeopleSoft customers would face an expensive transition to Oracle. Bradley, 6/9/04, 633:10-20; P3041, Patel dep., 6/3/04, 120:20-122:1, 123:13-22.

              9.6.4.1.1 Technical training, user training to learn the new application and reporting tools, rewriting all of the interfaces and reports would all make a transition to Oracle two three times more expensive than a PeopleSoft upgrade. P3045 at ORLITF0057307.

              9.6.4.1.2 The transition from PeopleSoft to Oracle is not as easy as Oracle has implied in its public statements, because there is a lot of money, human capital and emotional intelligence built into these systems. P3037, Knowles dep., 5/3/04, 156:21-158:17. "Since the PeopleSoft product will be killed, the strategy is to upgrade PeopleSoft customer to the next release of Oracle, a task that isn't nearly as easy or painless as Oracle is trying to make it sound." P4482 at SAP62528.

              9.6.4.1.3 As Debra Anderson, Novell's CIO, recognizes, a vendor's support personnel gain significant knowledge of a customer's unique configuration of the software. The knowledge maintenance personnel have about a customer's systems is "innate" and can not easily be transferred - it can't be written down. Thus, losing that person is much more significant than simply switching among plumbers. P8107, Anderson dep., 6/3/04, 44:8-20; 45:18-46:13.

            9.6.4.2 Such transition costs are not "exogenous," but rather are tied directly to the merger, because an owner of PeopleSoft that was not PeopleSoft's key competitor would not, applying "common sense," discontinue the product line and diminish its quality by decreasing the upgrades available. Johnsen, 6/16/04, 1750:21-1751:14, 1756:24-1757:15.

        9.7 Oracle's innovation claims are not merger specific.

          9.7.1 Oracle could build on its own all of the alleged modules it claims it will develop if it acquires PeopleSoft. Catz, 6/25/05, 3536:19-22.

          9.7.2 Ms. Catz testified that Oracle's applications products already have massive overlap with PeopleSoft's products today. Catz, 6/25/05, 3537:6-7.

          9.7.3 Oracle could acquire "scale," to the extent it needs it, by acquisitions other than PeopleSoft.

            9.7.3.1 Oracle can achieve the scale it wants without acquiring PeopleSoft, e.g., through other acquisitions. Ellison, 6/30/04, 4294:19-4295:24.

            9.7.3.2 There are many acquisition candidates available to achieve the scale Oracle thinks it needs, and Oracle is currently considering three or four. Ellison, 6/30/04, 4296:11-24.

            9.7.3.3 Before making the offer for PeopleSoft, Oracle looked at acquiring other companies such as Cerner, Lawson, and J.D. Edwards. P3171, Ellison dep., 1/20/04, 199:10-200:10.

            9.7.3.4 If this transaction is not permitted, Oracle would still be able to do other transactions to acquire the scale needed to compete with Microsoft, IBM and others. P3174, Ellison dep., 5/23/04, 145:5-146:12.

            9.7.3.5 If the PeopleSoft acquisition does not go through, Oracle will look again at acquiring Cerner or Lawson. P3171, Ellison dep., ½2O/O4, 200:11-18.

            9.7.3.6 If the PeopleSoft acquisition does not go through, Oracle can increase its scale by buying companies in specific verticals. P3171, Ellison dep., ½2O/O4, 202:11-203:1.

            9.7.3.7 Oracle has been in touch with [REDACTED TEXT]

            9.7.3.8 Oracle has been in touch with [REDACTED TEXT]

          9.7.4 Oracle could acquire "scale" by internal growth through competing for customers.

            9.7.4.1 Oracle can also achieve scale by continually adding new customers. Ellison, 6/30/04, 4297:22-4298:7.

            9.7.4.2 As Oracle gains more customers every year they come closer to achieving the scale they need without acquiring another company. P3174, Ellison dep., 5/23/04, 150:15-151:4.

        9.8 Oracle's various efficiencies claims related to innovation improvements are speculative.

          9.8.1 Ms. Catz testified that she did not know when Oracle would provide the alleged "superset" of products. Catz, 6/25/04, 3533:8-16.

          9.8.2 There are no contemporaneous business documents that discuss Oracle's alleged plans for a superset product. Catz, 6/25/04, 3534:20-3535:7.

          9.8.3 Oracle has not prepared an individual cost budget for any of the alleged modules or features for its purported "superset" product. P3348, Catz dep., 5/18/04, 108:9-14; P3349, Catz dep., 05/20/04, 420:15-424:10.

          9.8.4 Ms. Catz testified that she was not aware of whether Oracle had ever determined the amount of resources needed to build a pension module on its own. P3348, Catz dep., 5/18/04, 125:2-15.

          9.8.5 Ms. Catz testified that she was not aware of the number of employees Oracle would need to develop a pension module if Oracle acquires PeopleSoft. P3348, Catz dep., 05/18/04, 127:1-5.

          9.8.6 Ms. Catz testified that she was not aware of how much it would cost for Oracle to develop a mobile sales solution that is comparable to that of PeopleSoft's. P3348, Catz dep, 5/18/04, 271:3-275:17.

          9.8.7 Ms. Catz testified that Oracle had not prepared any projection of the number of employees, the costs or the time involved in developing an advisor module if Oracle acquires PeopleSoft. P3348, Catz dep., 5/18/04, 275:18-276:9.

          9.8.8 Dr. Hausman considers a technological breakthrough in the industry unlikely. Although he sees the advent of the stack being potentially important, he doesn't see a whole new approach coming out. Hausman, 6/29/04, 4022:5-12.

      10. Oracle's "Stack" or "Infrastructure Layer" Arguments Do Not Alter the Antitrust Analysis of the Proposed Acquisition or Mitigate Its Anticompetitive Impact in the Relevant Markets

        10.1 Background

          10.1.1 The term "stack" is used within the IT industry to mean many different, vertically organized products or technologies - system stacks, applications stacks, telecommunications stacks, etc. P3193, Mills dep., 5/27/04, 20:25-21:15.

            10.1.1.1 Oracle's arguments focus on the system stack, which is, generically, the hardware platform, the operating system, other infrastructure components, and applications on top. P3193, Mills dep., 5/27/04, 21:16-22.

            10.1.1.2 "Application Services" are infrastructure components that essentially perform services for an application - tasks such as security services, managing communication services, exchanging information with other applications --and thereby make it easier for the application to execute its functions. Kutnick, 6/30/04, 4138:18-4139:11; 4187:10-17; Bergquist, 6/8/04, 330:14-331:4.

            10.1.1.3 Applications services are also sometimes referred to as the "infrastructure layer." Kutnick, 6/30/04, 4139:12-16.

            10.1.1.4 "Web services" are a means of allowing different applications or programs to interconnect together through the means of standard protocols. P3173, Ellison dep., 5/23/04, 74:10-76:13 ("Now that there is this global standard for interconnecting machines, we can build a program that uses these standards and these are called Web Services that adopts that standard way of interconnecting.").

            10.1.1.5 Web services can be used to assist in greater integration between different software systems inside one company, as well as to more effectively automate business transactions between companies. Web services provide a way for applications to more easily exchange information, but they do not eliminate the difficulty of applications communicating or interacting; it is still necessary for the applications to understand what to do with the web services. Wohl, 6/25/04, 3428:15-3429:23.

          10.1.2 Oracle, through its two "industry" experts, Dale Kutnick and Dr. David Teece, asserts that the ongoing evolution of applications services or the "infrastructure layer" (or, as Dr. Teece calls it, the "integration layer") represent nothing less than a "tectonic shift," a "new world," and the most significant change in the information technology industry in the last 15 years. Kutnick, 6/30/04, 4215:24-4216:9. Oracle argues that the emergence of applications services will fundamentally change the "bases of competition". Teece, 7/1/04, 4396:23-4398:1 and 4361:4-15.

          10.1.3 Oracle's dire predictions are unfounded and wrong. Far from constituting an imminent, sudden "paradigm shift," the "infrastructure layer" or application services have been in existence for a number of years, have steadily and gradually evolved during that time, and will continue to evolve.

            10.1.3.1 Application servers, one of the basic integration products in the infrastructure layer, have been available in the market for seven or eight years. Kutnick, 6/30/04, 4200:22-24. Oracle has offered an application server for some time and has improved it through several different versions. Wohl, 6/25/04, 3414:2-18.

            10.1.3.2 Larry Ellison testified that web services as a significant means of doing application integration have been around for four years. P3173, Ellison dep., 5/23/04, 69:11-20.

            10.1.3.3 Various industry standards that facilitate interoperation and integration through middleware or an "integration layer" also are not new, but have been in the market five years or more.

              10.1.3.3.1 XML, one of the first web services standards which, in essence, standardizes the definition of a document and allows for easier exchange of information between applications, has existed for over five years. Kutnick, 6/30/04, 4201:10-18; Bergquist, 6/8/04, 328:4-14; Iansiti, 6/17/04, 2076:1-12.

              10.1.3.3.2 XML derivative standards such as UDDI or SOAP (Simple Object Access Protocol), which further define some of the objects within particular documents, have been developing and are continuing to develop as industry standards. Kutnick, 6/30/04, 4202:2-4203:2 ("Q And by that, would it be fair to say that those are - are -- have been developing and are continuing to develop as industry standards? A Absolutely.").

            10.1.3.4 From IBM's perspective, the IBM Software Group has been dominated by middleware technologies and helping customers do horizontal integration for 20 years. Horizontal integration is an IBM competency, and businesses have faced the challenge of horizontal integration for many years. P3193, Mills dep., 5/27/04, 30:16-31:5.

          10.1.4 Oracle's claims at trial about the significance of developments in the "infrastructure layer" and the consequences for competition in high-function applications are unfounded and contradicted by Oracle's contemporaneous business statements and conduct outside of this litigation.

            10.1.4.1 If Oracle's experts' speculation was correct about this "tectonic shift" and its likely drastic consequences for Oracle and its enterprise applications, Oracle would have been acutely aware of the developments and their dangers.

            10.1.4.2 Mr. Kutnick acknowledged that, if his opinion about a coming "inflection point" were right, such a change "absolutely" would pose a serious threat to Oracle's business and Oracle "absolutely" would have recognized that threat. Kutnick, 6/30/04, 4216:10-18.

            10.1.4.3 Yet, no Oracle documents suggest that it perceived any such "inflection point'' in the infrastructure space, or that any motivation for its attempt to acquire PeopleSoft was to better compete with IBM, BEA, Microsoft, and SAP in the "stack" or infrastructure.

            10.1.4.4 Instead, Oracle's business documents regarding the rationale for and expected benefit from the acquisition reveal that Oracle perceived no new kind of competition in "infrastructure" and no risks to its business from such new competition. Oracle's documents also contain no indication that Oracle believed, other than for litigation purposes, that it needed to acquire PeopleSoft in order somehow to help it better compete in the infrastructure layer with vendors such as IBM and BEA.

              10.1.4.4.1 For example, Oracle's June 5, 2003 presentation to its Board of Directors regarding the proposal to launch a takeover attempt of PeopleSoft lists various rationale for the acquisition. There is no mention anywhere in the document of developments in the infrastructure layer; of infrastructure or stack competition with IBM, Microsoft, BEA, or others; or of any need to acquire PeopleSoft in order to better compete in infrastructure. P3340.

                10.1.4.4.1.1 Instead, the stated rationale include strengthening Oracle's ability to compete against SAP and, in the long-run, Microsoft; improving Oracle's competitive position versus point product vendors; and providing complementary application expertise. P3340 at pp. 4; attached "Talking Points" p. 1.

              10.1.4.4.2 Similarly, a transcript of a conference call between Oracle executives and various financial analysts on June 6, 2003, submitted to the SEC as part of a Form 8-K filing, reveals no Oracle concern over infrastructure competition, no recognition of any "inflection point" or shift in the nature of competition for applications, and no thought that acquiring PeopleSoft was necessary to make Oracle a more viable infrasturcture competitor. Rather, Oracle's stated concern was limited to becoming a stronger applications competitor against SAP. P2039; P2039 at pp. 16, 18, 19, 24-25. A presentation to shareholders by Oracle's then-CFO Jeff Henley and then-Executive Vice President Charles Phillips explained the proposed acquisition and the rationale for it but never mentioned infrastructure competition, inflection points, or the need acquire PeopleSoft in order to make Oracle a more viable infrastructure competitor. P2041.

                10.1.4.4.2.1 Instead, the presentation states that the applications market is ripe for consolidation, P2041 at ORCL-EDOC-00192182, and that the acquisition would "significantly strengthen Oracle's ability to compete in enterprise applications against SAP, Microsoft and others." P2041 at ORCL-EDOC- 00192184 & 00192215 (emphasis added).

              10.1.4.4.3 As late as March 4, 2004, in its Answer to the Complaint in this case, Oracle showed no awareness and made no mention of the "infrastructure layer" or "stack" issues as looming inflection points, as factors affecting the competitive dynamics of high-function application markets, or as part of the rationale for Oracle's proposed takeover of PeopleSoft.

                10.1.4.4.3.1 Instead, in its answer to Paragraph 1 of the Complaint, Oracle stated that the acquisition "will permit Oracle to better compete with the clear market leader, SAP AG, and the numerous other existing competitors of the relevant software products, and will position Oracle to compete effectively with Microsoft Corp., which is aggressively expanding its position in enterprise applications software.'" There is no mention of infrastructure products or the need to compete with IBM, BEA, and other infrastructure vendors, (emphasis added)

                10.1.4.4.3.2 Similarly, in response to Paragraph 40 of the Complaint, which alleges various harmful effects of the acquisition, Oracle answered that the "acquisition will increase competition in the relevant markets by better positioning Oracle to compete against SAP, the clear market leader, and in the future against Microsoft...." There is no mention of competition in infrastructure or of competition against IBM or BEA.

              10.1.4.5 The fallacy of Oracle's infrastructure arguments is also demonstrated by the absence of statements by Oracle's executives and even its other experts recognizing or reflecting awareness of or concern about the impact of infrastructure developments.

                10.1.4.5.1 Oracle's executives did not mention the infrastructure layer or assert that it was a factor affecting competition in enterprise applications in the normal, pre- litigation course of their business.

                10.1.4.5.2 Dale Kutnick's first and only discussion with an Oracle business person came in late February or early March, 2003 when he spoke briefly by telephone with Charles Phillips, Oracle's co-president, about the possibility of being retained as an expert witness by Oracle. Kutnick, 6/30/04, 4185:22-4186:11.

                10.1.4.5.3 In that conversation, Mr. Phillips described various reasons why Oracle believed it needed to acquire PeopleSoft, but made no mention of the "infrastructure layer," "infrastructure competition" any coming "paradigm shift," or any need to be better positioned to compete with IBM, BEA, Microsoft, and SAP in infrastructure. Kutnick, 6/30/04, 4218:5-16.

                10.1.4.5.4 Rather, Mr. Phillips described to Mr. Kutnick Oracle's understanding of the applications market and Oracle's rationale for acquiring PeopleSoft. None of Mr. Phillips' explanation had anything to do with the infrastructure layer; instead, it was: Gaining additional applications and maintenance revenue; a need for the industry to be consolidated; and gaining greater "heft" to compete with SAP in applications. Kutnick, 6/30/04, 4217:5-4218:4.

                  10.1.4.5.4.1 Indeed, it was only Mr. Kutnick, and not Mr. Phillips or anyone from Oracle, who first made any mention of this purported "infrastructure competition." Kutnick, 6/30/04, 4218:5-4219:13.

                  10.1.4.5.4.2 After Mr. Kutnick described his opinions about the coming "tectonic shift" in infrastructure and his personal view that "Oracle's position in what [he] believed was the new world was very challenged," Mr. Phillips of Oracle laughed and then said something to the effect of, "That's all very interesting. Would you like to be an expert witness?" Kutnick, 6/30/04, 4219:11-4220:9 (emphasis added).

                10.1.4.5.5 Dr. Hausman, another Oracle expert witness, did not accept Oracle's attempt to portray developments in infrastructure as some sort of fundamental paradigm shift; he testified that, "I see the advent of the stack being important, but I don't see a whole new approach to this coming out." Hausman, 6/29/04, 4022:5-12.

              10.1.4.6 Plaintiffs' technical/industry expert, Dr. Iansiti, firmly rebutted the notion that the increasing capability of the infrastructure layer is a "paradigm shift" and testified that, instead integration technology has been evolving for many years. Iansiti, 6/17/04, 2077:12-2078:3.

            10.1.5 In fact, Oracle's acquisition of PeopleSoft will not result in Oracle obtaining any significant infrastructure technology that it needs to compete better in the infrastructure layer.

              10.1.5.1 Oracle's expert Dale Kutnick, who has held his general opinions about changes in the infrastructure layer for a number of years, did not perceive any connection between Oracle's proposed acquisition and the purported need for Oracle to compete better in infrastructure before he was contacted to serve as an expert for Oracle in this matter.

                10.1.5.1.1 Mr. Kutnick testified that it was his opinion that the acquisition of PeopleSoft by Oracle would help Oracle compete better in the new "application services" competition he believed was emerging. Kutnick, 6/30/04, 4211:19-4212:2.

                10.1.5.1.2 However, as late as March 2004, at the time of Mr. Kutnick's telephone call with Charles Phillips, Mr. Kutnick was still "curious" about why Oracle would want to acquire PeopleSoft, and he "basically questioned" Mr. Phillips about why Oracle was attempting the acquisition. Kutnick, 6/30/04, 4212:3-4213:13.

                10.1.5.1.3 Mr. Kutnick did not understand how the acquisition would improve Oracle's infrastructure. Kutnick, 6/30/04, 4213:25-4214:23 ("Q. In fact, sir, in your deposition, didn't you tell me that you didn't understand how the acquisition would improve Oracle's infrastructure? A. Yes, that's correct.")..

              10.1.5.2 Acquiring PeopleSoft would not bring Oracle any significant additions to its infrastructure technology. Larry Ellison agreed that Oracle was "not buying PeopleSoft in order to acquire their technology." Ellison, 6/30/04, 4315:19-22; P3174, Ellison dep., 5/23/04, 147:8-13 (acquiring PeopleSoft will not give Oracle any technology to help compete with SAP's NetWeaver).

              10.1.5.3 Instead, Oracle's only claim regarding a possible connection between its acquisition of PeopleSoft and an enhanced ability to compete in the "infrastructure layer" is that, by acquiring PeopleSoft, it will acquire (rather than having to compete for) more applications customers. The result, Oracle claims, is that it will have more revenue to spend on R&D and a larger base of customers to use its infrastructure products. Kutnick, 6/30/04, 4215:16-23; Teece, 7/1/04, 4406:16-4407:15; Ellison, 6/30/04, 4294:8-14.

                10.1.5.3.1 But neither Dr. Teece or Mr. Kutnick attempted to explain why Oracle, the second-largest software company in the world, needs to gain additional "scale" to be able effectively compete in R&D and innovation with firms such as BEA whose existing scale is already substantially smaller than Oracle's. Dr. Teece acknowledged that he had done no quantitative assessment of the "critical mass factor that will make the Oracle integration layer a more viable competitor to IBM, Microsoft, BEA and SAP." Teece, 7/1/04, 4406:16-4407:15; 4476:4-8.

                10.1.5.3.2 In any event, objective evidence of actual R&D expenditures refutes Dr. Teece's claim that the merger is needed to enable Oracle to achieve some critical mass to be a more viable competitor BEA, SAP, and others. SAP's revenues of slightly over $7 billion, Knowles, 6/23/04, 2806:5-6, and BEA's revenues of about $1 billion, P0035 at 10, are substantially lower than Oracle's revenues of just over $10 billion. Catz, 6/25/04, 3435:2-17. Yet SAP and BEA have been successful in developing integration layer products that Professor Teece now views as superior to Oracle's.

              10.1.5.4 Moreover, there is "nothing special" about the acquisition of PeopleSoft in particular in terms of Oracle gaining the additional revenues it claims to need, other than PeopleSoft's relative size of loyal customers. Kutnick, 6/30/04, 4220:10-4222:9.

                10.1.5.4.1 One way for Oracle to acquire the larger base of customers it wants is to buy PeopleSoft; the other way is for it to buy a number of smaller applications companies. Larry Ellison testified that "[i]f we can't buy - there's two ways to attack this, to get to scale, you can buy one of the big cross-industry players or you can buy a number of the specialists. They are - if you will, there's some equivalency there." P3171, Ellison dep., 1/20/04, 201:22-203:1.

                10.1.5.4.2 Acquiring other applications vendors such as Siebel, Lawson, or others could provide Oracle with a larger customer base, if on a relatively smaller scale. Ellison, 6/30/04, 4294:15-4296:4. Dale Kutnick viewed Siebel as hypothetically "an interesting acquisition to acquire customers." Kutnick, 6/30/04, 4220:10-4222:9. Oracle is currently considering three or four other acquisitions. Ellison, 6/30/04, 4296:11-24. If Oracle is not permitted to acquire PeopleSoft, it might again consider acquiring Lawson, which it views as "an eager seller." Ellison, 6/30/04, 4295:5-12.

            10.1.6 Several fundamental premises underlying the opinions of Mr. Kutnick and Dr. Teece about developments in the "infrastructure layer" and the impact of those developments on competition for enterprise applications are wrong.

              10.1.6.1 The fundamental premise underlying Dr. Teece's opinions about the impact of the "infrastructure layer" on the relevant markets in this case and the nature of competition among enterprise applications vendors is his assumption that there is and increasingly will be tight "co-specialization" among different layers of the stack, leading to a series of closed or "verticalized" stacks. Dr. Teece assumes that the infrastructure layer is highly "co-specialized with applications" and that there is a strong trend towards close coupling and co-specialization in the industry. Teece, 7/1/04, 4378:19-4379:10.

              10.1.6.2 Dr. Teece is unambiguous that, if his assumption about tight coupling or high co-specialization is wrong, then so-called "complementarities" from the stack and infrastructure layer don't matter in this case for antitrust purposes. Teece, 7/1/04, 4376:15-4377:4 ("I would agree if you're -- if you don't have co- -- if you don't have tight coupling and you don't have co-specialization, complementaries (sic) don't matter for antitrust purposes here."); Teece, 7/1/04, 4377:8-20 ("interoperability may indicate that there's some loose coupling.").

              10.1.6.3 Similarly, Dr. Teece acknowledges that the emergence of industry standards for infrastructure layer technology would weaken his notion of stack competition and "the stack can be undone." Teece, 7/1/04, 4445:13-4446:1.

                10.1.6.3.1 Yet Dr. Teece had not studied and did not specifically know the extent to which various vendors" infrastructure layers and applications products are or are not, in fact, highly co-specialized. Teece, 7/1/04, 4429:10-4431:1. Indeed, he did not even know with which infrastructure layer products PeopleSoft's applications are compatible. Teece, 7/1/04, 4429:21-4430:16.

                10.1.6.3.2 In fact, the evidence at trial demonstrated that Dr. Teece's premise was simply wrong and, therefore, his conclusions baseless. The assumption of tight "co- specialization" among the database, applications services, and applications in each, separate competing stack is plainly contradicted by evidence that various stack layers currently are and increasingly will be open and interoperable with one another.

                  10.1.6.3.2.1 Currently, only Oracle chooses to design its enterprise applications such that they run using only Oracle's application services and only on Oracle databases. Kutnick, 6/30/04, 4203:3-13; Bergquist, 6/8/04, 333:19-25 ("Oracle has tended to do a verticalization of the stack."); Iansiti, 6/17/04, 2071:5-19.

                  10.1.6.3.2.2 Oracle's "closed stack" strategy is, in the words of Mr. Kutnick, "shortsighted." Oracle's applications could reach a broader potential market by running on other platforms as well. Moreover, according to Mr. Kutnick, in general products like Oracle applications need to be able to stand on their own and be subjected to market forces to help make them the best in the business. Kutnick, 6/30/04, 4204:19-4205:18.

                  10.1.6.3.2.3 In contrast to Oracle, PeopleSoft's applications currently run on multiple databases and use multiple vendors' middleware or applications services. Kutnick, 6/30/04, 4203:14-20. Bergquist, 6/8/04, 333:16-18; 334:1-18.

                  10.1.6.3.2.4 Today, SAP's applications run on multiple databases. Kutnick, 6/30/04, 4201:21-4204:1.

                  10.1.6.3.2.5 Going forward, if PeopleSoft remained independent it likely would continue to make its applications work with multiple stacks; that is, its applications would continue to run on various vendor's databases and use various vendors' applications services. Bergquist, 6/8/04, 334:15-18. ("Q. And is this strategic choice to have your application run on multiple middleware layers something that PeopleSoft plans to continue? A. Absolutely.").

                  10.1.6.3.2.6 Similarly, going forward, SAP's FMS, HR, and other business applications are likely to "enable a much better interoperability with other people's application services." In fact, SAP and Microsoft already have engaged in discussions regarding making SAP's applications interoperable with Microsoft's .NET platform over time. Kutnick, 6/30/04, 4204:3-18.

                    10.1.6.3.2.6.1 SAP's "NetWeaver" is an integration and applications platform that supports open standards - including Java J2EE and Microsoft's .NET - and can be used to interconnect or integrate existing systems or other software vendors' systems. SAP's goal for several years has been to make its applications more open and accommodating to other vendors' software that a user might want to use. Knowles, 6/23/04, 2885:15-2886:6

                    10.1.6.3.2.6.2 SAP likely will move in the direction of its having its enterprise applications utilize the services not only of its "NetWeaver" infrastructure product but also of other vendors' infrastructure products. Kutnick, 6/30/04, 4204: 3-4 ("Shai Agassi [SAP executive board member] basically says that -- that we [SAP] are going to support not only other people's applications but enable a much better interoperability with other people's application services.... [O]ver time there would have to be some level of more interoperability between everybody's applications."); Teece, 7/1/04, 4440:8-4441:6; Bass, 6/16/04, 1637:7-22 ("the NetWeaver stack today provides much more openness to enable more flexibility with working with the IBM stack or BEA or Microsoft dot-Net.").

                    10.1.6.3.2.6.3 An SAP document titled "SAP NetWeaver Platform Interoperability with IBM WebSphere and Microsoft.NET," indicates that "[a] key ingredient of SAP NetWeaver is complete interoperability with both Microsoft.NET and IBM WebSphere. That means IT organizations don't have to decide between the two technologies. Using SAP NetWeaver, they can weave their environments into a single, smooth fabric and do so at a lower cost of ownership." P3337 at p. 6; Teece, 7/1/04, 4435:25-4436:25.

                  10.1.6.3.2.7 In the future, even Oracle likely would modify its applications so that they could use the services of various vendors' application services. Kutnick, 6/30/04, 4211:8-18.

                    10.1.6.3.2.7.1 As Oracle makes its applications more and more Java enabled, those applications will increasingly interoperate with IBM's WebSphere, BEA's WebLogic, and Microsoft's .NET, for example. P3173, Ellison dep., 5/23/04, 38:7-20; 57:16-22; 67:18-68:1; 72:1-9. Oracle has announced an agreement with Microsoft supporting coexistence with Microsoft's .NET technology platform. P3173, Ellison dep., 5/23/04, 35:9-22.

                  10.1.6.3.2.8 Going forward, applications vendors are likely to make their applications able to work with various vendors' application services. This "open stack" strategy will be the best thing for customers and for the applications vendors as well. Having cross-stack applications means that those applications have to stand on their own in the market, ensuring that, in Mr. Kutnick's words, "it's the best. It sells the best. You know it offers the best functions, et cetera, et cetera." Open stacks also would be good because they would give users more choice in what application services they use with their application. Kutnick, 6/30/04, 4209:18-4210:16 ("Q Going forward, a decision by SAP to make its application's product work on various vendors' infrastructure services platforms ... would be a good thing because that would give users more choice in what application services they do use with their application? A That's correct.").

                  10.1.6.3.2.9 PeopleSoft believes its strategy of not providing its own infrastructure layer but instead working with multiple stacks is an advantage. It permits customers to choose whatever stack they want "to either mix and match them or to have competition across multiple vendors." Bergquist, 6/8/04, 334:19-335:1.

                10.1.6.3.3 Dr. Teece's key assumption of tight "co-specialization" among stack layers and the existence of closed, competing stacks was refuted not only by evidence of what vendors are doing in the market, but also by Oracle's other "infrastructure" expert, Mr. Kutnick, on whom Dr. Teece relied. Mr. Kutnick testified that applications vendors in the future will write their applications to work with several vendors' stacks, and that Oracle's applications "certainly could be modified to utilize the services of other vendors' application services." Kutnick, 6/30/04, 4211:8-18.

                  10.1.6.3.3.1 Dr. Teece himself did not do any detailed analysis of the costs to applications vendors of writing their applications to work with different stacks or infrastructure layers. Teece, 7/1/04, 4429:3-14.

              10.1.6.4 One of the fundamental premises underlying Dr. Kutnick's opinion about "application services" and the purported "need" for Oracle to acquire PeopleSoft is that Oracle's own infrastructure products are inferior and, without the acquisition, Oracle will not be an effective infrastructure competitor to IBM, Microsoft, BEA, and SAP. This assumption is wrong. Not only did Mr. Kutnick cite no evidence to support the assumption but, in fact, the evidence at trial established that Oracle's application server and its other middleware products are of high quality and capability and are competitive.

                10.1.6.4.1.1 According to Ron Wohl, Oracle's head of applications development, Oracle has steadily improved its application server. Currently, it is a very high-quality application server product and very competitive with application servers from IBM, Microsoft, BEA, and others. Wohl, 6/25/04, 3414: 2-24.

                10.1.6.4.1.2 Mr. Ellison testified that Oracle's application server is today a quality, competitive product that compares favorably with IBM's Websphere, NetWeaver and other significant infrastructure products in the market. Ellison, 6/30/04, 4324:8-16 ("From BEA, from Microsoft, absolutely").

                10.1.6.4.1.3 Similarly, Oracle's April 14, 2003 "Board Presentation on M&A Opportunities" describes Oracle's database, application server, application integration, and development tools as "competitive." P0035 at p. 3.

            10.1.6.5 Mr. Kutnick's predictions of changes in the infrastructure layer constituting a "tectonic shift," a "new world," and the most significant change in the information technology industry in the last 15 years, Kutnick, 6/30/04, 4215: 24-4216: 9, are speculative and unreliable as a basis to ignore the otherwise clear anticompetitive impact of the proposed acquisition.

              10.1.6.5.1 Mr. Kutnick readily acknowledged the speculative nature of his opinions about the "infrastructure layer" when he explained why he saw no need to review Oracle business documents or speak to any Oracle business people before reaching his conclusions. When asked about his belief that he didn't need anyone else's opinion to help shape his own, Mr. Kutnick admitted, "That doesn't mean it was right, but it was my opinion." Kutnick, 6/30/04, 4185: 8-21 (emphasis added).

              10.1.6.5.2 Mr. Kutnick also acknowledged that his "opinions" frequently have been contradicted by the beliefs and business judgments of Oracle itself - on occasions "too many to count." Kutnick, 6/30/04, 4234: 9-21 ("Q. Do you find that you've disagreed in the past with Oracle on a number of matters? A. Too many to count, that's correct.").

              10.1.6.5.3 Mr. Kutnick, while referencing his experience working for industry analyst firms such as META Group, Inc., did not testify on behalf of or representing the META Group. Instead, his testimony at trial consisted of only his own, personal opinions. Kutnick, 6/30/04, 4186:12-18.

          10.1.7 Microsoft's reaction to Oracle's proposed acquisition of PeopleSoft does not support the notion of a coming paradigm shift or some significant changes in the "stack level competition" as described by Dr. Teece.

            10.1.7.1 In early June, 2003, Microsoft began exploring the possibility of acquiring SAP. Burgum, 06/23/04, 3027:12-3029:4; 3069:22-3071:24; D5324R at MS-OPCID 8865.

            10.1.7.2 Also in June, 2003, after Oracle's bid for PeopleSoft was announced, Microsoft executives internally expressed some initial concerns that what Microsoft saw as the "duopolistic long-term future" envisioned by Oracle - that is, a two-player market in enterprise applications -- might negatively impact Microsoft's database business if Oracle steered PeopleSoft applications customers to the Oracle database. Burgum, 06/23/04, 3030:2-3032:9; D5822R at MS-OPCID 19180; Burgum, 06/23/04, 3069:22-3072:22; D7173R, Burgum dep., 05/13/04, 189:25-192:18.

            10.1.7.3 One Microsoft executive's early evaluation of the proposed Oracle/PeopleSoft transaction, expressly based on "high-level" assumptions and without benefit of input from senior executives or personnel in the relevant business groups, [REDACTED TEXT] - suggested that if the acquisition resulted in a tighter IBM/SAP relationship, that could hurt Microsoft's database and infrastructure business. D5822R at MS-OPCID 19181; Burgum, 06/23/04, 3035:4-3036:11; [REDACTED TEXT].

            10.1.7.4 At the same time, Microsoft also recognized that a possible combination of Microsoft and SAP could yield a key strategic benefit of permitting "back-end" enterprise applications to better interact with and deliver data to Microsoft's desktop Office productivity suite. Burgum, 06/23/04, 3032:10-3033:22; D5822R at MS-OPCID 19176 ("Key position to realize vision of moving enterprise applications from specialists' to everyone's desks").

            10.1.7.5 Between June and December, 2003, Microsoft continued to contemplate a possible combination with SAP; at a meeting of representatives of the two companies sometime in the Fall Microsoft first raised the possibility of a combination. Burgum, 06/23/04, 3036:12-25.

            10.1.7.6 By December, 2003, Microsoft's understanding of the enterprise applications market had developed and it realized that the economics of that market would require that, if it purchased SAP, SAP's applications would have to continue to be cross platform - that is, continue to run on various databases. Burgum, 06/23/04, 3041:23-3044:6 ("financial and economic necessity require that post-acquisition that Microsoft maintain SAP's heterogenous environment for its - its customers. Meaning, again, that... to have the SAP business continue to grow and prosper that you would have to provide cross- platform support for all of those Sagittarius [SAP] applications.")(discussing P841R at MS- OPCID 19500).

            10.1.7.7 Similarly, by December, 2003, Microsoft realized that, even if it did not acquire SAP and SAP instead forged a closer relationship with IBM, SAP would have an economic need to continue to have its applications be cross platform. Burgum, 06/23/04, 3044:7-3045:6. This economic need reduced any risk to Microsoft's business from closer relations between IBM and SAP.

            10.1.7.8 Thus, Microsoft's initial June, 2003, thinking about any potential database or infrastructure impact of the proposed Oracle/PeopleSoft acquisition was of significantly less concern by December. In fact, materials prepared by Microsoft in preparation for a December 5, 2003, meeting with SAP did not even mention the possibility of any impact on Microsoft of IBM and what SAP might do in connection with IBM if there was not a Microsoft/SAP combination. Burgum, 06/23/04, 3044:7-3045:6; P841R.

            10.1.7.9 Instead, what had emerged by December as the "leading idea" for why a possible Microsoft/SAP combination made sense, the "highest potential opportunity for innovation" with SAP, was the benefit recognized back in June to better link "backend data" from enterprise applications to Microsoft Office. Burgum, 06/23/04, 3038:17- 3040:23; P841R at MS-OPCID 19500 (First item listed under "The Future: How a Combination Would Benefit Customers" is "Front end users better able to communicate with and use backend data - Great Office integration for customers").

            10.1.7.10 In fact, Microsoft's assessment of the market situation had evolved such that the ultimate success or failure of Oracle's bid for PeopleSoft do not have an impact on Microsoft's consideration of what it should do with regard to the enterprise applications market. .[REDACTED TEXT]

            10.1.7.11 Similarly, in SAP-related materials prepared for review by the Microsoft Board of Directors dated January 5, 2004, a board presentation in which Microsoft CEO Steve Ballmer was instrumental, [REDACTED TEXT], a key strategic rationale is described as "development of innovative new products that break barriers between end users and backend enterprise data." P2640R at MS-OPSUB 000681; Burgum, 06/23/04, 3045:18-3047:9.

            10.1.7.12 Microsoft also flatly states at this point that achieving leadership in ERP applications for Fortune 500 companies is "an objective we will not pursue in the absence of this combination." P2640R at MS-OPSUB 000682. Cindy Bates, who in late 2003 was managing director of Microsoft's corporate development and strategy group, testified that, other than through a possible acquisition of SAP, getting into the enterprise applications market was not even "an objective" for Microsoft - "We don't have plans to." P3251, [REDACTED TEXT].

            10.1.7.13 Microsoft's January 5, 2004 presentation for its Board and Ms. Bates testimony refute any notion that lingering infrastructure concerns would compel Microsoft in the future to obtain a presence in the enterprise application space. P2640R at MS- OPSUB 000681.

          10.1.8 IBM's reaction to Oracle's proposed acquisition of PeopleSoft does not indicate that IBM recognized or feared the projected "tectonic shift" described by Oracle's experts. Nor did IBM see significant changes in the "stack level competition" as described by Dr. Teece.

            10.1.8.1 Overall, IBM reviewed the possible implications of the proposed PeopleSoft acquisition and, despite a projected loss of revenue, determined that the loss was manageable and that IBM should take no action. While IBM's staff proposed various recommendations for consideration, IBM management acted on none of those recommendations.

            10.1.8.2 IBM is opposed to the proposed acquisition and expects to lose some middleware business if the acquisition goes through. P3193, Mills dep., 5/27/04, 93:6-94:2. However, after carefully considering the issue, IBM's overall conclusion was that the acquisition would not have a substantially negative effect on IBM's business. P3193, Mills dep., 5/27/04, 119:5-10. IBM's response to the proposed acquisition has been to take no action at all. P3193, Mills dep., 5/27/04, 100:11-101:3.

            10.1.8.3 IBM staff organizations and strategy and planning groups assessed the impact of the proposed acquisition on IBM and made at least three presentations by the staff to the executive team concerning its potential impact. The executive team, not the staff, is responsible for determining IBM's response. P3193, Mills dep., 5/27/04, 96:9-97:5; 97:19-98:7; D5242R; D7143, Mills dep. (corrected), 102:6-103:2; D5240R; D7143, Mills dep. (corrected), 111:24-113:10; D5241R.

            10.1.8.4 If a "tectonic shift" or paradigm shift were in fact occurring, IBM, which has viewed integration and infrastructure technology as a core competency for 20 years, would be aware of it. Yet, none of the staff presentations make any reference to or recognize any such dramatic developments in the integration layer. D5240R; D5241R; D5242R.

            10.1.8.5 Instead, the staff presentations focus on the straight-forward possibility that some hardware, operating systems or middleware business would shift away from IBM technologies as a result of the proposed acquisition.

              10.1.8.5.1 One view of the IBM staff was that the acquisition would result in some migration away from IBM platforms, defined broadly to include hardware and system platforms. P3193, Mills dep., 5/27/04, 98:23-99:20; D5242R, at IBM500656.

              10.1.8.5.2 Another view of the IBM staff was that some customers might move off IBM middleware platforms. P3193, Mills dep., 5/27/04, 99:21-100:10; D5242R, at IBM500658.

            10.1.8.6 A variety of actions were suggested by the staff as possible responses but all were rejected by the executive team.

              10.1.8.6.1 The staff suggested that IBM "Strengthen IBM/SAP relationship. Push for embedded IBM middleware." Steve Mills reaction was that this suggestion was "a dumb idea and it wasn't going to happen." P3193, Mills dep., 5/27/04, 101:4-10.

              10.1.8.6.2 The staff also suggested that SAP would be more motivated to work more closely with IBM because SAP would perceive Oracle to become an even bigger competitor if the merger went through. P3193, Mills dep., 5/27/04, 102:16-103:23. But the executive team believed that IBM already spends a lot of money marketing and selling its products, and "there was no reason to act in any unique way relative to the PeopleSoft/Oracle acquisition, whether it was going to happen or not going to happen." P3193, Mills dep., 5/27/04, 101:24-102:7.

              10.1.8.6.3 The staff suggested that IBM "Identify emerging ISVs in the next tier under integrated accounts to offer additional IBM support to engender loyalty." But IBM did not act on this recommendation, because it is "already doing these things" and already has many programs to reach out to ISVs. Many of these ISVs "perceive Oracle as a competitor or unreliable partner" and this provides IBM an opportunity to work closely with them. But that "condition has existed in the marketplace for years. There is nothing new here." P3193, Mills dep., 5/27/04, 103:24-104:21.

              10.1.8.6.4 At these executive strategy meetings, IBM also discussed working more closely with others in the industry, and the potential impact on the broad range of IBM hardware, software and services. But instead of dire consequences, IBM concluded "that much of the business that we do today would continue post acquisition." P3193, Mills dep., 5/27/04, 112:7-16.

              10.1.8.6.5 The staff characterized the merger situation by stating that "We are exposed to potential consolidation of the ISV space, especially to acquisitions by hostile players." D7143, Mills dep. (corrected), 105:15-17; D5240R at p.6 (bates number illegible). The staff "intent was to say that the consolidation of independent software vendor application space represented a potential problem for the IBM company...". D7143, Mills dep. (corrected), 105:22-106:5.

              10.1.8.6.6 But the reaction by the IBM executive team to this "hostile player" slide was much the same as it was to all the other suggestions. The executive team discussed IBM's relationship with Oracle. "In fact, it was one of the reasons why we concluded that although the staff had, you know, spent time pulling together charts, it was thank you very much, we don't need to talk about this anymore." D7143, Mills dep. (corrected), 106:11-19. (emphasis added)

              10.1.8.6.7 The reasons the IBM executive team did not act on the staff suggestions "are quite simple. We - we have a lot of business that we do around Oracle applications today. Oracle applications run on IBM hardware. Oracle applications are things that are installed by IBM services people. ... And we have such a broad and diverse software portfolio that there are elements of our software portfolio that are sold into accounts that use Oracle applications." D7143, Mills dep. (corrected), 106:19-107:3.

              10.1.8.6.8 In short, "the prospects of Oracle acquiring PeopleSoft, though we didn't see it necessarily as being a favorable outcome in terms of parts of our middleware business, we did not conclude that it had a massive effect on the IBM company, given the diversity of relationships we have in the marketplace." D7143, Mills dep. (corrected), 107:4-13. (emphasis added)

              10.1.8.6.9 Concerning generally the presentation of D5242R "Oracle Acquisition of PS & JDE," and the specific option concerning "Lobbying efforts against ISV concentration with customers, influencers, and regulators," Mills testified that

              10.1.8.6.10 We went through it relatively quickly as a presentation. There was a lot of material here, a lot of which we didn't pause on during the meting and everything came down to a general discussion of the implications of this on our business and does this mean that we should change anything strategically for IBM. The conclusion that was reached in the meeting was no. P3193, Mills dep., 5/27/04, 114:14-115:14. (emphasis added).

              10.1.8.6.11 IBM's final conclusion was that the proposed Oracle acquisition of PeopleSoft would not have a substantially negative effect on IBM's business in total. P3193, Mills dep., 5/27/04, 119:5-10. See also P3193, Mills dep., 5/27/04, 100:2-101:3 (not pursue recommendations); D7143, Mills dep. (corrected), 104:20-25 ("We reviewed the material, talked about what was happening in the marketplace, and decided to do nothing.")

            10.1.8.7 As part of its analysis, IBM also reaffirmed its commitment to stay out of the applications business.

              10.1.8.7.1 IBM makes hardware, operating systems, and middleware, but does not make business applications software. P3193, Mills dep., 5/27/04, 24:9-22.

              10.1.8.7.2 Concerning making an acquisition itself, IBM "talked about not doing that. We talked about how our strategy was still sound, that not being in the applications business continued to make sense, that this acquisition [Oracle/PeopleSoft] was not going to change our strategy in the marketplace because it didn't in any fundamental way upset the economic value proposition that IBM had been benefitting from in choosing to partner with application vendors." P3193, Mills dep., 5/27/04, 112:19- 113:8.

              10.1.8.7.3 No specific acquisitions were suggested by the staff. P3193, Mills dep., 5/27/04, 114:2-13.

              10.1.8.7.4 Moreover, Mills testified that instead of considering an applications vendor purchase, "all the conversations I've participated in and all the conversations I know of actually were to the contrary." P3193, Mills dep., 5/27/04, 117:23-118:7.

              10.1.8.7.5 Further, the 1998 declaration by IBM that it would not compete in the applications business is "something we always return to as a grounding principle, and there are some who at times will come forward and say, well, shall we rethink that and the consistent answer is no." P3193, Mills dep., 5/27/04, 113:16-25.

              10.1.8.7.6 Mills testified that "since 1998 we've been absolute in our position on this. Since I am the guy who leads the software business, I've refused to look at any and all application company evaluations of purchase." P3193, Mills dep., 5/27/04, 118:23-119:4.

            10.1.8.8 At these executive strategy meetings, IBM did discuss the marketplace response to the proposed acquisition. "Our perspective of that was that customers would have a far more limited choice where they're trying to get to a consolidated ERP environment, that they're going to be choosing between two vendors instead of three, choosing between SAP and Oracle instead of Oracle, PeopleSoft, and SAP, that SAP would, in fact, likely be a beneficiary in some instances where customers were not comfortable doing business with Oracle." P3193, Mills dep., 5/27/04, 110:25-111:19.

          10.1.9 At the end of the day, there is nothing about the ongoing evolution of application services or the infrastructure layer that is likely to result in Oracle, PeopleSoft and SAP no longer being the only companies that provide high-function HRM and FMS applications. Iansiti, 6/17/04, 2080:4-11.

            10.1.9.1 Whatever further evolution of middleware or the "infrastructure layer" occurs, those middleware elements cannot and will not replace the application software that automates the business knowledge and the business processes in an application.

            10.1.9.2 The HRM and FMS applications reflect and automate the business process and functionality, for example, hiring an employee, terminating an employee, changing jobs, etc. The infrastructure layer "doesn't do the business logic. It provides services and capabilities that are made use of by that enterprise software." Bergquist 6/8/04, 331:22-332:13.

            10.1.9.3 The bulk of what is done today by high-function applications is the business logic that reflects an understanding of the processes within a business, the industry specialization, the global specialization, etc. These are things that will not become part of the infrastructure layer. Bergquist 6/8/04, 333:8-15.

            10.1.9.4 Infrastructure products like SAPs NetWeaver and Peoplesoft's AppConnect are not substitutes for HRM or FMS applications. Instead, they are complements to applications which help to integrate applications but which have no capability of their own to process information or transactions. They do not replace, for example, the "transaction processing engine" of the application. Wilmington, 6/16/04, 1813:18-1814:20.

            10.1.9.5 Middleware doesn't provide the application function. Middleware is not about the ERP function. It's not about the accounts receivable, accounts payable. Middleware is merely about how you get things connected together. P3193, Mills dep., 5/27/04, 17:21-18:14.

            10.1.9.6 Customers "absolutely" will still require an application separate from any "application services" to provide the actual business functions and processes provided today by applications. Kutnick, 6/30/04, 4187:25-4188:8.

            10.1.9.7 For example, many of Bearing Point's clients use integration products such as NetWeaver to connect their core FMS and HR systems with other point solutions, but still must rely on applications from Oracle, PeopleSoft, and SAP for their core FMS and HR support. Keating, 6/10/04, 995:4-9.

            10.1.9.8 As Dr. Iansiti described the difference, "FMS and HRMS applications distinguish themselves because of the business processes that they enable, and that knowledge has nothing to do with some technology that sits underneath the stack. They are no substitute for each other. There's a richness of features and functionality that cannot be easily substituted by some new player coming along.... [Y]ou can make it incrementally easier to write an application. It's easier now to write an application than it was five years ago or ten years ago. What's not easy is understanding what the large organizations want and what's not easy is to take that understanding and translate into code. The technology itself makes it much more easy to do this, but it doesn't create a substitute for what we've been talking about." Iansiti, 6/17/04, 2076:17-2077:11.

            10.1.9.9 Even if infrastructure were to evolve to be much more capable in the future, application developers would still have to understand the details of the underlying business processes they were attempting to automate and then write actual software code that makes the application perform the relevant business process. Kutnick, 6/30/04, 4189:5-16.

            10.1.9.10 Thus, all of the complexity that exists today in developing the portion of an application that automates the underlying business processes will not be replaced or eliminated by greater infrastructure or integration capability of middleware. Kutnick, 6/30/04, 4189:17-21.

            10.1.9.11 As a result, entry into enterprise applications will remain formidable no matter what developments occur in the infrastructure layer. Bergquist 6/8/04, 332:24-333:15 (new entrants "would still face a formidable undertaking to produce the same functionality that is provided by PeopleSoft in these areas."). See Section 8.6, supra.

            10.1.9.12 While Dr. Teece asserts that the infrastructure layer will make best of breed applications more successful, he has not done and is not aware of any quantitative analysis assessing the extent of any such impact on entry cost or whether that impact would be sufficient to allow additional firms to compete in high-function human resource and financial management software markets. Teece, 7/2/04, 447:6-17.

            10.1.9.13 Oracle's strategy of offering infrastructure technology that makes it easier to integrate applications from various vendors together does not undermine the value of Oracle's integrated business suite by making it easier for best of breed vendors to replace Oracle's applications. P3044, Wohl dep. 4/29/04, 226:25-227:3 ("Q [I]f you offer technology for integrating with other functionality, do you undercut the value of your integrated functionality? A Not in the least.").

          10.1.10 Oracle also asserts that the evolution of the infrastructure layer enables the creation of so-called "composite applications." Teece, 7/1/04, 4452:9-15.

            10.1.10.1 Composite applications represent an evolution of the business functionality provided by the large enterprise applications vendors. They provide a variety of functionality that can work across traditional applications such as FMS and HR, automating many of the tasks that users formerly had to do manually. These composite applications are bringing all that content together, so people don't have to act as the glue to composite things together in a manual fashion. Bergquist, 6/8/04, 325:15-327-10; 328:23-329:7.

            10.1.10.2 The idea of composite applications is not new; they have been evolving over the last three years. Bergquist, 6/8/04, 327:19-328:3. This ongoing "emergence" will continue over the period 2005 through 2015. D7021 at p. 14.

            10.1.10.3 Composite applications are meant to be complements to, not substitutes for, HRM and FMS applications software. P3342, at page 21; Bergquist, 6/8/04, 329:8-20. "Composite characteristics" need to be added into the business processis involved in HRM and FMS applications. Bergquist, 6/8/04, 327:11-18.

            10.1.10.4 Even where composite applications are developed, the core functionality performed by HRM or FMS applications still needs to be performed by those applications. Bergquist, 6/8/04, 328:15-20.

            10.1.10.5 Rather than replacing or eliminating the need for high-function HRM or FMS applications offered by Oracle and PeopleSoft, composite applications instead will augment the functionality of those traditional applications to make users more effective and efficient. Bergquist, 6/8/04, 328:15-329:7 ("This is an augmentation process, not a replacement process.").

            10.1.10.6 Composite applications remain difficult to develop. Teece, 7/1/04, 4448, 1-21; P3339 (Gartner research report noting that, "[d]espite advances in technology, composite applications are difficult to develop.").

            10.1.10.7 While certain PeopleSoft documents from ealry 1993 indicate initial concern over whether resources would be available to enable PeopleSoft to develop and sell composite applications, PeopleSoft in fact has developed and is shipping applications that have varying degrees of composite characteristics. Bergquist, 6/8/04, 328:19-25.

            10.1.10.8 The continued development of composite applications will not make it significantly easier for new entrants or smaller vendors to develop applications that have the same capability and functional as high- function applications offered by Oracle, PeopleSoft, and SAP for complex customers. Bergquist, 6/8/04, 328:15-329:17.

      11 Many of the Opinions Offered by Oracle's Expert Witnesses Are Unfounded, Contradicted by Other Oracle Witnesses or the Trial Record, or Otherwise Not Credible

        11.1 Dr. David Teece

          11.1.1 A fundamental premise underlying Dr. Teece's stack and "integration layer" opinions is wrong. His analysis depends on the assumption that there is and increasingly will be tight "co-specialization" among different layers of the stack, leading to a series of closed or "verticalized" stacks. Dr. Teece testified that the infrastructure layer is highly "'co-specialized with applications" and that "there is a strong trend towards close coupling and co-specialization" in the industry. Teece, 7/1/04, 4378:19-4379:10.

            11.1.1.1 Dr. Teece admits that, if his assumption about tight coupling or high co-specialization is wrong, then so-called "complementarities" from the stack and infrastructure layer don't matter in this case for antitrust purposes. Teece, 7/1/04, 4376:15-4377:4 ("I would agree if you're -- if you don't have co- --if you don't have tight coupling and you don't have co- specialization, complementaries (sic) don't matter for antitrust purposes here."); Teece, 7/1/04, 4377:8-20 ("interoperability may indicate that there's some loose coupling."). He also acknowledges that the emergence of industry standards for infrastructure layer technology would weaken his notion of stack competition and "the stack can be undone." Teece, 7/1/04, 4445:13-4446:1.

            11.1.1.2 Dr. Teece had not studied and did not know the extent to which various vendors' infrastructure layers and applications products are or are not, in fact, highly co- specialized. Teece, 7/1/04, 4429:10-4431:1.

            11.1.1.3 In fact, Dr. Teece's assumption of increasing close coupling and co- specialization among the layers of separate, competing stacks was contradicted by the evidence. Just the opposite is true; various stack layers currently are and increasingly will be open and interoperable with one another. See Paragraphs 10.1.6.3.2, et seq., supra.

            11.1.1.4 In particular, the testimony of Mr. Kutnick, upon whom Dr. Teece frequently relied for support, is inconsistent with and substantially undercuts Dr. Teece's premises about the existence of and a trend toward more and more close coupling and co- specialization. See, e.g., Kutnick, 6/30/04, 4201: 14-20 (applications vendors in the future will write their applications to work with several vendors' stacks, and Oracle's applications could be modified to utilize the services of other vendors' application services); Kutnick, 6/30/04, 4211:8-18 (PeopleSoft's applications currently run on multiple databases and use multiple vendors' middleware or applications services); Kutnick, 6/30/04, 4201:21-25 (SAP's applications today run on multiple databases); Kutnick, 6/30/04, 4201:25-4202:18 (SAP's applications are moving toward "a much better interoperability with other people's application services" and SAP and Microsoft are discussing interoperability with .NET); Kutnick, 6/30/04,4208:3-10 (SAP interoperability); Kutnick, 6/30/04, 4211:8-18 (even Oracle is likely modify its applications in the future to use other vendors' application services).

          11.1.2 While Dr. Teece faulted the Plaintiffs for not considering the impact of stack or integration layer in assessing competition, his own assessments of that impact were often vague and non-specific.

            11.1.2.1 Dr. Teece did not attempt to identify or articulate any specific definition of the relevant markets based on his notions of stack competition. Instead, he testified generically that, "But I'm saying at a minimum in terms of relevant market, I have not been asked to determine what the outer scope of the market is, but I know for sure that it includes a whole bunch of products that the government has left out and that also brings in a new bunch of competitors." Teece, 7/1/04, 4371:24-4372:12 (emphasis added).

            11.1.2.2 Dr. Teece's view of the impact on customers of stack was similarly vague: "Because this typical sale is -- as you recognize, your Honor, is not just of a piece of software. There's a whole bunch of other stuff that gets sold along with it, and it's the fact that when this is priced, one takes into account other products that are being sold and takes into account the future." Teece, 7/1/04, 4402:9-4403:11.

              11.1.2.2.1 In spite of this generic and repeated assertion that stack-level competition somehow disciplines pricing of high-function application software, Teece, 7/1/04, 4373:11-23, Dr. Teece did not attempt any comprehensive or econometric analysis of pricing or discounts for high-function HRM and FMS applications. Teece, 7/1/04, 4424:11-4425:13 ("No, 1 did not look at the particular dynamics of -- of pricing in -- in applications....").

            11.1.2.3 In response to a question about the durability of Oracle's, PeopleSoft's, and SAP's dominance in ERP applications, Dr. Teece again resorted to vague, generic predictions: "I think the integration layer is going to change a lot of that, because it's going to enable composite, it's going to enable a whole bunch of new developments and bring into play Microsoft and IBM and several others." Teece, 7/1/04, 4458:25-4460:20 (emphasis added); P3338.

          11.1.3 Dr. Teece disagreed with Plaintiffs that price discrimination occurs in the relevant markets. Teece, 7/1/04, 4494:22-4495:9. He testified in response to the Court's questions that, "first of all, it's intellectual property, Your Honor. So, typically when we think about price discrimination, in all the textbooks the word price discrimination [sic] in terms of physical products." Teece, 7/1/04, 4496:17-4497:4. But in fact, "[p]rice discrimination is prevalent in the software industry ...." William J. Baumol & Daniel G. Swanson, The New Economy and Ubiquitous Competitive Price Discrimination: Identifying Defensible Criteria of Market Power, 70 Antitrust L.J. 661, 680 (2003); X Areeda, Antitrust Law, ¶ 1769d2A (Supp. 2d ed. 2004) ("[a]nother common instance of price discrimination occurs in the licensing of patents, copyrights, and trademarks.").

          11.1.4 Dr. Teece was uninformed about a number of basic facts underlying his opinions or contained in materials on which he relied.

            11.1.4.1 The key underpinning of Dr. Teece's testimony is his assumption that there is and increasingly will be tight "co-specialization" among different layers of the stack, yet he had not studied and did not specifically know the extent to which various vendors' infrastructure layers and applications products are or are not, in fact, highly co- specialized. Teece, 7/1/04, 4429:10-4431:1. Indeed, he did not even know with which infrastructure layer products PeopleSoft's applications are compatible. Teece, 7/1/04, 4429:21-4430:16.

            11.1.4.2 Although he asserted that applications vendors would find it too costly to write their applications to work with different stacks or infrastructure layers, Dr. Teece did not do any detailed analysis of those costs or their likely impact. Teece, 7/1/04, 4429-3:14.

            11.1.4.3 In response to questions by the Court, Dr. Teece was unable to define or identify a number of terms in documents on which he relied. Teece, 7/1/04, 4388:11- 4389:13 (meaning of "MW" in D5240R); Teece, 7/1/04, 4389:20-4390:20 (same exhibit); Teece, 7/1/04, 4492:1-4494:5 (meaning of term in P3338).

            11.1.4.4 Dr. Teece also had difficulty explaining several concepts he introduced in his testimony. Teece, 7/1/04, 4404:17-4405:2 ("installed base effects"); Teece, 7/1/04, 4414:18-4415:15 (correct denominator for a percentage he calculated relating to Dr. Elzinga's relevant product market).

            11.1.4.5 Dr. Teece in his testimony tried to minimize the conflict between Mr. Kutnick's testimony about increasing interoperability across stacks and his own claims about increasing co-specialization.

            11.1.4.5.1 Dr. Teece first tried to draw an analogy involving telephones, and explained that "a landline phone system is interoperable with cell phone system, but you can't plug your cell phone into the land line. So, they are interoperable, but the cell phone system is still co-specialized as between the -- the handsets and the equipment." Teece, 7/1/04, 4483:24-4484:9.

            11.1.4.5.2 He was then asked by the Court to define portability as he was using it, and replied, "Portability -- we use the term, for instance, with respect to the cell phones, number portability is where the number can work perfectly no matter whose system you're on. And so, portability is, if you will, a higher level of integration. THE COURT: Higher level of interoperability? THE WITNESS: Well, it requires a higher level of integration, I suppose, or at least some common standards and protocol." Teece, 7/1/04, 4484:24-4485:24.

      11.2 Dale Kutnick

        11.2.1 Mr. Kutnick's view of the industry is contradicted by Oracle's business documents, activities, and statements in the marketplace.

          11.2.1.1 Mr. Kutnick's opinions predict industry upheaval as a result of developments in applications services or the infrastructure layer - what he variously describes as a "tectonic shift," a "new world," and "the most significant change" in the information technology industry in the last 15 years. Kutnick, 6/30/04, 4215:24-4216:9.

          11.2.1.2 Yet in formulating his opinions Mr. Kutnick deliberately avoided examining any Oracle business documents or the views of any Oracle executives to determine whether there was evidence that, in the real world, Oracle perceived any such threat. Kutnick, 6/30/04, 4184:1-16; 4185:8-4186:21 (Mr. Kutnick said he wanted his expert report to be his opinion and, while he didn't want to be cynical, he did not think he would you get any information that would modify that opinion by talking to Oracle executives).

          11.2.1.3 Mr. Kutnick chose not to examine any actual business documents or other evidence from Oracle even though he recognized that, if his opinion about a coming "inflection point" was right, such a change "absolutely" would pose a serious threat to Oracle's business and Oracle "absolutely" would have recognized that threat. Kutnick, 6/30/04,4216:10-18.

          11.2.1.4 The one view of the industry Mr. Kutnick did receive from an Oracle executive, Mr. Phillips, was inconsistent with Mr. Kutnick's theories. Kutnick, 6/30/04, 4217:5-4218:16; see generally Paragraphs 10.1.4.5, et seq., supra.

          11.2.1.5 The evidence in the record establishes that Oracle, in the course of its pre- litigation business activities, did not recognize or fear the paradigm shift Mr. Kutnick now hypothesizes, nor did it see its proposed acquisition of PeopleSoft as necessary to better compete with IBM, BEA, Microsoft, and SAP in the "stack" or infrastructure.

        11.2.2 While Mr. Kutnick now opines that the acquisition of PeopleSoft by Oracle would help Oracle compete better in the new "application services" competition he predicts, Kutnick, 6/30/04, 4211:19-4212:2, as late as March, 2004, he did not understand how the acquisition would improve Oracle's infrastructure and questioned why Oracle was pursuing the acquisition. Kutnick, 6/30/04, 4213:2-4214:13.

          11.2.2.1 In his February or March 2004, telephone conversation with Charles Phillips, Mr. Kutnick was still "curious" about why Oracle would want to acquire PeopleSoft, and he "basically questioned" Mr. Phillips about why Oracle was attempting the acquisition. Kutnick, 6/30/04, 4212:3-4213:13.

          11.2.2.2 Mr. Kutnick did not understand how the acquisition would improve Oracle's infrastructure. Kutnick, 6/30/04, 4213:25-4214:23 (agreeing that, at the time of his deposition in May, "to this day, you didn't believe that PeopleSoft brings any significant new infrastructure capability to Oracle.").

          11.2.2.3 In fact, acquiring PeopleSoft would not bring Oracle any significant additions to its infrastructure technology. Ellison, 6/30/04, 4315:19-22 (agreeing that Oracle was "not buying PeopleSoft in order to acquire their technology."); P3174, Ellison dep., 5/23/04, 147:8-13 (acquiring PeopleSoft will not give Oracle any technology to help compete with SAP's NetWeaver).

        11.2.3 Mr. Kutnick's opinion about Oracle's "need" to acquire PeopleSoft in order have more customers and revenue to be an effective infrastructure competitor to IBM, Microsoft, BEA, and SAP is premised on the notion that Oracle's own infrastructure products are inferior. Kutnick, 6/30/04, 4175:9-4176:7.

          11.2.3.1 But Mr. Kutnick cites no evidence to support his assumption and, in fact, the uncontroverted evidence is that Oracle's application server and its other middleware products are of high quality and capability and are competitive. Wohl, 6/25/04, 3414: 2-24 (Oracle's application server has been steadily improved and currently is very high-quality and very competitive with application servers from IBM, Microsoft, BEA, and others); Ellison, 6/30/04, 4324:8-16 (Oracle's application server is a quality, competitive product that compares favorably with IBM's Websphere, NetWeaver and others); P0035 at 3 (April 14, 2003 Oracle Board presentation lists Oracle's database, application server, application integration, and development tools as "competitive.).

        11.2.4 Mr. Kutnick also acknowledged that one of his key conclusions, that application "suites are increasingly an endangered species" and "[b]est-of-breed is going to become more viable," is inconsistent with Oracle's business statements and conduct. Kutnick, 6/30/04, 4234:1-18.

          11.2.4.1 Mr. Kutnick conceded that he frequently has disagreed with Oracle in the past, on occasions "too many to count." Kutnick, 6/30/04, 4234:19-4235:2.

      11.3 Dr. Jerry Hausman.

        11.3.1 Dr. Hausman's testimony at trial revealed that he reached erroneous conclusions about market definition based on an incorrect application of economic principles. First, he failed to apply the hypothetical monopolist test.

          11.3.1.1 Dr. Hausman claimed not to have thought in detail about whether packaged software vendors could satisfy the hypothetical monopolist test with respect to financial or HR software. Hausman, 6/29/04, 3977: 18-3980:1.

          11.3.1.2 Dr. Hausman testified that he had not concluded whether Oracle, PeopleSoft, and SAP could satisfy the hypothetical monopolist test.

            11.3.1.2.1 Dr. Hausman testified that he had not tried to determine whether Oracle, PeopleSoft, and SAP could jointly raise price five to ten percent. Hausman, 6/29/04, 3967: 6-15.

            11.3.1.2.1 On redirect, Dr. Hausman testified that he had interpreted the above questioning to relate to whether Oracle, PeopleSoft, and SAP together or along with other firms satisfied the hypothetical monopolist test. Dr. Hausman testified that he thought the issue not to have relevance to this case. Hausman, 6/29/04, 4050:3-4051:12

            11.3.1.3 Dr. Hausman's testimony at trial about not having applied the hypothetical monopolist test and whether Oracle, SAP, and PeopleSoft satisfied the hypothetical monopolist test varied from his deposition testimony.

            11.3.1.3.1 Packaged software is a very broad term, which includes FMS and HRM software. Hausman, 6/29/04, 3974:19-3976:3

            11.3.1.3.2 At his deposition, Dr. Hausman testified that all packaged software jointly would not satisfy the hypothetical monopolist test. Hausman, 6/29/04, 3978:6-3980:9

            11.3.1.3.3 In addition, at his deposition, Dr. Hausman testified that PeopleSoft, Oracle, and SAP could not raise price jointly. Hausman, 6/29/04, 3967:18-3970:3.

          11.3.1.4 Dr. Hausman's inconsistent testimony at his deposition that packaged software would not satisfy the hypothetical monopolist test rested, in part, on misinterpretation of his own regressions.

          11.3.1.5 The first stated reason for Dr. Hausman's earlier opinion was that his empirical work showed that adding one or two vendors made no significant difference to the discount. Hausman, 6/29/04, 3985:5-9

          11.3.1.6 However, his regressions in this case do show an economic effect from adding one or two vendors generally. Hausman, 6/29/04, 3995:2-3998:8 (when Peoplesoft goes from facing two competitors to a single competitor, the discount falls eight percentage points and the customer pays 16% more); Hausman, 6/29/04, 4000:8-4002:21 (when Oracle goes from facing no competitors to facing a single competitor the discount increases 4.5 percentage points).

          11.3.1.7 Dr. Hausman's regressions in this case show an economic effect from adding SAP or Peoplesoft specifically. Hausman, 6/29/04, 4004:8-4005:9 ("Q Okay, so this regression would not support your claim that the empirical work shows that adding one or two competitors doesn't significantly increase the discount, correct? A. I would agree with that to the extent that it's either PeopleSoft or SAP, but if you adjust a regular competitor, as you note, it's, you know, essentially zero, but yes, for these, as I testified yesterday, either SAP or PeopleSoft does have an effect on Oracle's pricing."). Hausman, 6/29/04,4006:13-23.

          11.3.1.8 Dr. Hausman admitted that some customers get a better discount when they have PeopleSoft rather than SAP as a competitor to Oracle. Yet, he concluded that the merger is not anti-competitive despite his not knowing how many customers the merger will harm. Hausman, 6/29/04, at 4018:2-16.

        11.3.2 In order to avoid a contradictory opinion regarding the similarity between how enterprise application software markets and auction or other bidding markets function from the expert retained by Oracle during Plaintiffs' investigation, Dr. Hausman testified that "Dr. Warren Boulton in [his] view is an extremely poor economist." Hausman, 6/29/04, 3965:5- 3966:5, at 3966:3. Dr. Warren-Boulton is a former head of economics at the Antitrust Division (Hausman, 6/29/04, 3966:6-8) and testified against Dr. Hausman, as an expert economist for the Federal Trade Commission (the prevailing party), in FTC. v. Staples, 970 F.Supp. 1066 (D.D.C. 1997).

        11.3.3 Dr. Hausman contradicted the testimony of Dr. Teece and Mr. Kutnick portraying developments in the infrastructure layer as a fundamental "paradigm shift" or "tectonic shift." Instead, Dr. Hausman testified that, "I see the advent of the stack being important, but I don't see a whole new approach to this coming out." Hausman, 6/29/04, 4022:5-12.

      11.4 Dr. Thomas Campbell

        11.4.1 Dr. Campbell's conclusions about market definition were based on erroneous assumptions.

          11.4.1.1 Dr. Campbell was asked to assume that the Plaintiffs had alleged "single- purchaser" markets. Campbell, 6/22/04, 2690:1-24. Most of his conclusions were based on that assumption. Campbell, 6/22/04, 2701:5-2702:19.

          11.4.1.2 In fact, Plaintiffs did not allege or seek to prove "single-purchaser" markets, but rather alleged and offered proof about traditional product markets based on product characteristics. First Amended Complaint ITU 23-24; Elzinga, 6/18/04, 2229:1-2231:10 (questioning by the Court).

        11.4.2 Dr. Campbell's conclusions about the competitive significance of legacy systems also were based on erroneous assumptions.

          11.4.2.1 Dr. Campbell assumed that high-function HRM and FMS software is durable, i.e., that "[i]t goes on forever. There's nothing to wear out." Campbell, 6/22/04, 2694:15-24. This assumption was a foundation for his four conclusions. Campbell, 6/22/04, 2701:5-2702:19.

          11.4.2.2 In fact, customers explained that software, to be usable for the purposes for which it is purchased, must be maintained and upgraded. Wolfe (North Dakota), 6/16/04, 1572:14-24 ("The assumption is, you're going to get new releases of software every 12 to 18 months, and that's like a refresh of that product."); P3041, Patel (Ford) dep., 6/03/04, 118:17-119:3 ("After a period of time, we will find that our PeopleSoft product will be of little use within the company.... Our efficiency, and our effectiveness as a company would suffer. And after a period of time, we would have to seek an alternative product."); Johnsen (PepsiAmericas), 6/16/04, 1737:1-7 ("It doesn't answer the investment that would be made to enhance the functionality of the software."); Bradley (Verizon), 6/9/04, 607:16-608:8 ("You can't let these things simply quiesce and run your company the way they run it today. That's not an option."); Cichanowicz (Nextel), 6/14/04, 1081:1-16; Wesson (AIMCO), 6/14/04, 1142:10-22; 1155:13-16.

          11.4.2.3 Dr. Campbell acknowledged that many customers consider upgrades important and consider software durable only because they anticipate upgrades from the vendor over the life of the software. Moreover, Dr. Campbell did no study of the cost of upgrading. Campbell, 6/22/04, 2749:25-2752:2.

          11.4.2.4 Dr. Campbell acknowledged that software "is not durable, yes or no" but rather "more or less durable depending upon how simple it is to make the changes that are important." For example, changes in the tax laws or employment laws must be incorporated into the software. In some instances, the customer will be very dependent on the vendor. Campbell, 6/22/49, 2752:18-2754:10 (questioning by the Court).

          11.4.2.5 Dr. Campbell claimed that a customer could stay with its legacy system by adding best of breed options, but acknowledged that this would be an option only if it is cost effective. Dr. Campbell did not study this question and it depends on every individual circumstance. Dr. Campbell did not examine the testimony of Mr. Ellison on this point. Campbell, 6/22/04, 2739:5-19. Mr. Ellison testified that best of breed solutions can be unbelievably expensive for companies. P3172, Ellison dep., ½0/04, 135:21-136:17

        11.4.3 Dr. Campbell's assumptions about the likely competitive effects of the merger led to incredible conclusions.

          11.4.3.1 Dr. Campbell assumed that there was a well-defined three-firm market for high-function HR and FMS software in the United States. Campbell, 6/22/04, 2689:13-24.

          11.4.3.2 Dr. Campbell concluded that a lawful, open cartel in this market would have indeterminate effects on prices for consumers. Campbell, 6/22/04, 2737:24-2738:17.

          11.4.3.3 Dr. Campbell did not have a view on whether a merger of all three firms in this market would have harmful effects on prices for consumers. Campbell, 6/22/04, 2738:18-2739:4.


APPENDIX A

GOVERNMENT WITNESSES

1. Richard Allen (J.D. Edwards)

Mr. Allen was Executive Vice-President and Chief Financial Officer of JD Edwards at the time of its acquisition by PeopleSoft and had been the firm's senior most financial officer for the entire eighteen years he was employed there. He was also a member of the firm's Executive Committee and its Board of Directors, and in those roles was involved in setting corporate policy. At the time of its acquisition, JD Edwards was the fourth largest enterprise application software vendor with 5500-6000 customers, 5000 employees, including between 1000 and 1500 programmers and software engineers, and nearly $1 billion in annual revenue. Allen, 6/10/04, 746:18-748:17.

2. Lynn Anderson (Cap Gemini)

Ms. Anderson is a Vice President at Cap Gemini and the leader of the PeopleSoft service area. Ms. Anderson is responsible for Cap Gemini's PeopleSoft team in the U.S. and The Group's overall PeopleSoft relationship globally. Anderson has worked at Cap Gemini and its predecessor companies since 1993. She is a certified Human Resource Professional, certified PeopleSoft HRMS consultant, and has been working with the PeopleSoft package since 1991. P3200, Anderson dep., 5/13/04, 6:20-24; 8:17-9:17; P4188, at 1.

Cap Gemini, formerly called Cap Gemini Ernst and Young U.S. LLC until April 16, 2004, is the U.S. subsidiary of the French based firm Cap Gemini SA, also know as "The Group." The Group employs about 50,000 people worldwide and had 2002 revenues of about 7 billion Euros. The Group conducts business, including its global PeopleSoft business, through approximately 130 subsidiaries in about 30 countries. P3200, Anderson dep., 5/13/04, 6:25-8:13; 16:2-21.

3. Anthony Armstrong (Credit Suisse First Boston)

Mr. Armstrong has been a director at Credit Suisse First Boston ("CSFB") and its predecessor, Donaldson, Lefken and Jenerette for two years. P3347, Armstrong dep., 5/13/04, 7:14-8:16.

CSFB was retained in 2003 to advise Oracle in its acquisition of PeopleSoft. P3347, Armstrong dep., 5/13/04, 8:21-9:8.

4. Orlando Ayala (Microsoft)

Mr. Ayala is a Senior Vice President at Microsoft, responsible for the Small and Medium Business Solutions Group. He has held this position at Microsoft for one year. Mr. Ayala joined Microsoft in 1991 and was tasked with building the Latin America Division. In 1997, he began to run the Americas Division and within two years he was promoted to Head of Worldwide Sales for Microsoft. P3255, Ayala dep., 5/18/04, 7:8-8:4.

5. Christy Bass (Accenture)

Ms. Bass is the Global Managing Partner of Global Business Solutions, i.e. all the ERP products and services Accenture takes to market. Ms. Bass has worked for Accenture (or its predecessor, Andersen Consulting) for 23 years, and has held her current position since 2001. Bass, 6/16/04, 1605:22-1606:2; 1607:18-1608:12.

Accenture is a technology, business consulting, and outsourcing firm; according to Ms. Bass, Accenture is the largest system integrator in the world. Accenture employs 90,000 people in 48 countries. Accenture had revenue of $11.4 billion in 2003. Accenture has approximately 4500 clients, 1200 to 1500 of which are members of the Global 2000. Bass, 6/16/04, 1610:4-25; 1611:13-1612:2.

6. Cindy Bates (Microsoft)

Ms. Bates is General Manager, U.S. Small Business at Microsoft. She is responsible for the profit and loss associated with the U.S. Small Business group, which serves small companies of 15 employees or less. Ms. Bates reports to Richard Emerson, the Senior Vice President of Corporate Development. She joined Microsoft in 2000 as Director of the Corporate Development and Strategy Group. P3251, Bates dep., 5/7/04, 6:17-21; 12:21-13:12; 11:8-11.

7. Richard Bergquist (PeopleSoft)

Mr. Bergquist is Chief Technology Officer, Senior Vice President, and Fellow at PeopleSoft, Inc. Bergquist was the sixth employee hired by PeopleSoft and has been with the company since 1988. In addition to being responsible for ensuring that the firm's technology and applications are able to compete effectively in the market, he works extensively with customers and prospects about the direction of the company's products. Prior to joining PeopleSoft he worked for eleven years as a developer and project manager at AMS. Bergquist, 6/8/04, 255:13-257:3.

8. Keith Block (Oracle)

Keith Block is currently Oracle's Executive Vice President North America responsible for operating the North American licensed sales and consulting organization. He works in Waltham, MA. Mr. Block joined Oracle Corp. in 1986 and has been in his current position for approximately one and one-half years, after working his way up through various sales management positions, including running industry-specific sales operations. Mr. Block has a Bachelor's Degree in Information Systems and a Master's Degree in Management and Public Policy, both obtained from Carnegie-Mellon University. P3009, Block dep., 12/16/03, 5:7-9; 10:18-12:1; 14:6-16; 14:19-22.

9. Zena Brand (ADP)

Zena Brand is ADP's Division Vice President, Global Product Marketing for Employer Services International. Her responsibilities include defining ADP's product strategy for multinational clients. P3042, Brand dep., 5/18/04, 9:3-6, 9:24-10:6.

10. Laurette Bradley (Verizon)

Ms. Bradley has been Senior Vice President for Information Technology for two-and-one-half years. She is responsible for Verizon's finance, payroll, and human resource systems, Verizon's information technology off-shoring program, and Verizon's overall IT business planning. Bradley, 6/9/04, 577:10-578:5.

Verizon is a Fortune 20 telecommunications company providing local and wireless service to about one third of the United States. Verizon's annual revenues are $66 billion and it operates or has investments in 30 different countries. Bradley, 6/9/04, 577:12-15; 580:18-581:4.

Verizon uses PeopleSoft and SAP HRM software and PeopleSoft and SAP FMS software. Bradley, 6/9/04,583:16-23; 590:2-10.

11. Robert Bullock (CH2MHill)

Mr. Bullock is Chief Information Officer and Senior Vice President for CH2MHill. Bullock, 6/7/04, 193:3-5.

CH2MHill is a civil engineering and environmental engineering firm that had $2 billion in revenues (prior to acquiring another firm in December) in 2003. Bullock, 6/7/04, 193:8-18.

CH2MHill currently uses Oracle FMS software and is implementing Oracle HRM software. Bullock, 6/7/04, 196:25-197:6; 199:21-200:6.

12. Douglas Burgum (Microsoft)

Mr. Burgum is a Senior Vice President at Microsoft, responsible for the Microsoft Business Solutions business group. He reports directly to Steve Ballmer, CEO of Microsoft. Prior to joining Microsoft, Mr. Burgum was the CEO of Great Plains, where he held the title of CEO until Microsoft purchased Great Plains in April 2001. Mr. Burgum is responsible for the overall performance of Microsoft Business Solutions, including profit and loss, strategy, and product development. Burgum, 6/23/04, 2971:12-25; 2974:6-22.

13. Paul Ciandrini (Oracle)

Mr. Ciandrini was Senior Vice President of North American Application Sales for Oracle, and was an officer of Oracle at the time of his deposition. As Senior Vice President for Application Sales, Mr. Ciandrini reported to Keith Block, and managed application sales for Oracle in North America, with general vice presidents responsible for East and West geographic sales territories reporting to him. Mr. Ciandrini has since left Oracle. P3179, Ciandrini dep., 15:13-21;16:12-15.

14. Richard Cichanowicz (Nextel)

Mr. Cichanowicz is Vice President of Business System Integration at Nextel and has been at Nextel just over six years. Prior to his work at Nextel, Mr. Cichanowicz was employed at Arthur Andersen, MCI, and Global Satellite. Mr. Cichanowicz's responsibility at Nextel is to oversee "Next Step," an overhaul of Nextel's business processes. Cichanowicz, 6/14/04, 1052:9-12; 1053:9-13; 1054:16-1055:15.

Nextel is a Fortune 200 company that provides wireless services to primarily large business customers. Nextel had $8 billion in revenues for 2003 and has over 13 million subscribers. Cichanowicz, 6/14/04, 1052:23-1053:3.

Nextel is currently using PeopleSoft for HR, FMS, and supply chain management. Cichanowicz, 6/14/04, 1072:11-19; 1073:21-25.

15. Sebastian DeSimone (Metro-North)

Mr. DeSimone has been Chief Information Officer for Metro-North for six years. He is responsible for every electronic component and application owned by Metro-North. P3061, DeSimone dep., 5/19/04, 7:25-8:18.

Metro-North is a subsidiary of the New York Metropolitan Transit Authority. Metro-North has about 5,600 employees and annual revenues of about $1 billion. P3061, DeSimone dep., 5/19/04, 15:12-18; 18:23-19:5; 19:10-12.

Metro-North recently procured PeopleSoft HRM software. P3061, DeSimone dep., 5/19/04, 223:18-224:18.

16. David Dortenzo (Deloitte Consulting LLP)

Mr. Dortenzo is a principal at Deloitte Consulting LLP ("Deloitte"). Mr. Dortenzo manages the U.S. operations for Deloitte's Oracle consulting practice. He is responsible for all the Oracle consulting operations, partner deployment, and staff growth. From June 2001 until June 2002, Dortenzo was responsible for both the PeopleSoft Americas practice and the Oracle practice for Deloitte. P3183, Dortenzo dep., 5/5/04, 8:8-18; 9:7-12; 23:10-22.

Deloitte is a part Deloitte & Touche LLP USA, which in turn is a member of the Swiss Verein (a partnership holding company) Deloitte Touche Tohmatsu. P3183, Dortenzo dep., 5/5/04, 10:13-11:17 [Under Seal].

17. Richard Scott Elliott (Kerr-McGee)

Mr. Elliott is Human Resources Portfolio manager in the company's Information Management and Technology division. Mr. Elliott manages a group of eight information technology professionals that support human resources at Kerr-McGee, as well as the company's legal and medical groups. Mr. Elliott's responsibilities include advising his Kerr-McGee customers on software procurement issues. He served as Technical Lead in 2001 for Kerr-McGee's vendor selection process for its HRM software. P3062, Elliott dep., 5/20/04, 5:21-24, 8:15-9:10, 10:6-14, 55:22-25.

Kerr-McGee is a worldwide company involved in two major lines of business - oil and gas exploration and production, and chemical manufacturing. It has operations across the United States and in Brazil, Benin, Morocco, Australia, China, the Netherlands, Germany and the United Kingdom. P3062, Elliott dep., 5/20/04, 14:9-14, 14:20-15:3.

Kerr-McGee uses Oracle FMS software and PeopleSoft HRM software. P3062, Elliott dep., 5/20/04, 42:23-43:12; 46:17-47:4.

18. Kenneth Elzinga (University of Virginia)

Dr. Elzinga, Ph.D., is currently the Robert C. Taylor Professor of Economics at the University of Virginia. He has taught economics at the University of Virginia since 1967, and served as the Assistant Dean of the College of Arts and Sciences from 1971 to 1973. In 1992, Dr. Elzinga received the Thomas Jefferson Award, the highest honor bestowed by the University of Virginia. Elzinga 6/18/04, 2142:11-18; P4014A.

Dr. Elzinga has also held academic appointments at the University of Chicago, Cambridge University, Trinity University, Michigan State University, and the University of Bonn. P4014A.

Dr. Elzinga has published more than 80 scholarly articles and books, primarily in the fields of antitrust economics and industrial organization. He currently serves on the editorial board of the Antitrust Bulletin, and has served as President of the Southern Economic Association and the Industrial Organization Society. Elzinga 6/18/04, 2142:19-20; 2143:2-3; 2144:1-4; P4014A.

From 1970 to 1971, Dr. Elzinga was Special Economic Advisor to the Assistant Attorney General, Antitrust Division, U.S. Department of Justice. Dr. Elzinga has been and is currently a consultant on antitrust issues to the U.S. Federal Trade Commission. He is currently working with Judge Kaplan on the Sotheby's/Christie's auction house antitrust litigation. He has also served as an Administrative Law Judge for the Nuclear Regulatory Commission in antitrust matters. Elzinga 6/18/04, 2142:21-2143:1; 2143:19-23; P4014A.

Dr. Elzinga has studied the software industry. He served as testifying expert before Judge Kollar-Kotelly in U.S. v. Microsoft, in support of the remedy agreed to by the U.S. Department of Justice and Microsoft, when that remedy was challenged by several non-settling states. He is the author (with David E. Mills) of "PC Software," in the Antitrust Bulletin, and (with David S. Evans and Albert L. Nichols) of "U.S. v. Microsoft: Remedy or Malady?" in the George Mason Law Review. Elzinga 6/18/04, 2144:15-2145:8; P4014A.

19. Mary Elizabeth Glover (Greyhound Lines) Ms. Glover is Vice President, Information Technology for Greyhound Lines ("Greyhound"). She has been with Greyhound since 1992, working her way up from Programmer Analyst III to Vice President. As Vice President her responsibilites include strategic and operational effectiveness of the organization and getting the products that meet Greyhounds technological needs at the best cost. Glover, 6/15/04, 1462:25-1464:3.

Greyhound is a bus transportation company serving 3,700 destinations in the United States and Canada. Greyhound employs 16,000 people and generates average revenue of $1.2 billion. Glover, 6/15/04, 1460:1-24.

Greyhound uses Oracle financial management software and HRl human resource management software. Glover, 6/15/04, 1464:8-13; 1465:5-8.

20. Michael Gorriz (DaimlerChrysler)

Dr. Gorriz has been Vice President, IT Business Systems for DaimlerChrysler since 2000. He is responsible for procuring, installing, and maintaining the software for the functional units of DaimlerChrysler. Gorriz, 6/15/04, 1368:18-1369:7.

DaimlerChrysler manufactures cars and trucks in roughly 100 manufacturing facilities around the world. DaimlerChrysler employs 365,000 people, 100,000 of those in the United States. Gorriz, 6/15/04, 1368:2-13.

DaimlerChrysler uses predominately SAP financial management software and predominately PeopleSoft human resource management software. Gorriz, 6/15/04, 1369:25-1370:4; 1371:13-18.

21. Scott Allen Hatfield (Cox Communications)

Mr. Hatfield is Senior Vice President and Chief Information Officer for Cox Communications, and has held that position since 1996. Mr. Hatfield's responsibilities include all of the information systems used by Cox Communications, as well as the design, implementation, and operation of Cox Communications' business applications. Hatfield, 6/7/04, 86:23-87:17.

Cox Communications is the third largest cable operator in the United States, with over 6.5 million video subscribers in about 30 states. Cox Communications has 21,000 employees. Hatfield, 6/7/04, 89:7-90:8.

Cox Communications uses PeopleSoft human resource management software and J.D. Edwards financial management software. Hatfield, 6/7/04, 94:9-14; 100:8-10.

22. Jeff Henley (Oracle)

Mr. Henley is Chief Financial Officer and Chairman of the Board of Directors for Oracle.

23. Marco Iansiti (Harvard Business School)

Dr. Iansiti is the David Sarnoff Professor of Business Administration at Harvard Business School. He has performed research in the technology sector, has taught courses in management operations, innovation, information technology and product development, and has created a joint PhD program in information technology management. He has taught computer science courses to undergraduate and graduate students at Harvard. Iansiti, 6/17/04, 2018:5-9; 2019:6-24; 2020:11-17.

Dr. Iansiti's first book, "Technology Integration," is about the match between technology and its uses. Dr. Iansiti's most recent book deals with the effect that changes in technology have on established vendors. Iansiti, 6/17/04, 2020:3-10; 2078:23-2079:1.

He has consulted for a variety of companies, including NetDynamics, which was one of the first application server companies. He has been a board member on a variety of software start-up companies, including Model N, (where he was a co-founder) which designed enterprise level applications for pricing and contract management. Iansiti, 6/17/04, 2020:19-2021:13.

24. Kenneth Johnsen (Pepsi Americas)

Mr. Johnsen has been Vice President and Chief Information Officer for Pepsi Americas since 2001. Mr. Johnsen was Chief of Technology at Pepsi Americas from 1997 to 2001, and Director of Technology at Budget Rent-a-Car for seven years prior to joining Pepsi Americas. Mr. Johnsen leads the information technology ("IT") area for Pepsi Americas, overseeing over 200 IT employees and an IT budget of about $50 million. He is ultimately responsible for Pepsi Americas' software purchases. Johnsen, 6/16/04, 1725:9-22; 1726:2-1727:1.

Pepsi Americas is a Fortune 500 company with annual revenue of $3.2 billion and 15,000 employees. Pepsi Americas manufacures, sells, and distributes Pepsi products. Pepsi Americas operates in 18 states in addition to four countries in Central Europe and the Carribean region. Johnsen, 6/16/04, 1723:24-1724:13.

Pepsi Americas currently uses Peoplesoft human resources management software and financial management software in the United States and SAP human resources management and financial management software in Europe. Johnsen, 6/16/04, 1727:10-1728:18.

25. Ann Marie Janke (Target)

Ms. Janke is Director, Target Technology Services. She is responsible for project MAX, which involves the installation of a new HRM system for Target. P3020, Janke dep., 5/13/04, 10:8-13 [Under Seal].

Target recently purchased PeopleSoft HRM software. P3020, Janke dep., 5/13/04, 107:19-108:12 [Under Seal].

26. Perry Keating (BearinePoint)

Mr. Keating is a Senior Vice President at BearingPoint. Keating, 6/10/04, 857:7-12. Prior to joining BearingPoint, Mr. Keating worked for both PeopleSoft and Oracle. Mr. Keating is in charge of BearingPoint's Global Enterprise Solution Group, which delivers BearingPoint's World Class Finance and Workforce Oracle, SAP and PeopleSoft solutions. He directly supervises over 800 people in BearingPoint's Enterprise Solution Group Keating, 6/10/04, 861:15-21; 862:8-17; 861:22-862:7; 862:21-25; 863:1-6.

BearingPoint is one of the largest consulting companies in the world and the former consulting arm of the audit, tax, and consulting firm KPMG. BearingPoint's annual revenues are in excess of $3 billion and it has 15,000 employees worldwide. It does business in the United States, Europe, Asia-Pac, and Latin America. It has commercial clients as well as public service and federal and ther government agency clients. In providing Tier One core ERP services, BearingPoint helps its clients select software, implement software, upgrade from older software versions to newer versions and consolidate to a smaller number of ERP systems. Keating, 6/10/04, 857:13-25; 866:5-13; P0200, at 1.

27. Phillip Lynn Maxwell (Neiman Marcus Group)

Mr. Maxwell has been Senior Vice President and Chief Information Officer at Neiman Marcus Group ("Neiman Marcus") for four years. Prior to that, he was Chief Information Officer for Nissan North America for two years. As CIO for Neiman Marcus, Mr. Maxwell is in charge of all the firm's technology activities. Maxwell, 6/9/04, 650:8-651:7.

Neiman Marcus Group is a specialty retailer that does business under several brand names, including Neiman Marcus and Bergdorf Goodman, at several locations throughout the United States. Neiman Marcus group has 15,000 employees and over $3 billion in annual sales. Maxwell, 6/9/04, 651:23-652:13.

Neiman Marcus has implemented Oracle FMS and licensed Oracle HR software. Maxwell, 6/9/04, 652:14-17; 674:7-11.

28. R. Preston McAfee (California Institute of Technology)

Dr. McAfee, Ph.D., currently holds the J. Stanley Johnson Chair at the California Institute of Technology. Previously, Dr. McAfee held the Murray S. Johnson Chair at the University of Texas 16 at Austin. He has taught Industrial Organization at UT Austin and MIT. McAfee, 6/21/2004, 2443:10-2444:7.

Dr. McAfee has authored over 70 scholarly articles in a variety of topics, including antitrust economics. He is also an Associate Editor for the Journal of Economic Theory and has been co- editor for the past nine years for the American Economic Review. McAfee, 6/21/04, 2445:6-23. 2223Dr. McAfee has co-authored one book "Incentives in Government Contracting" (with John McMillan), and more recently, authored his own book, "Competitive Solutions: A Strategist's Toolkit." McAfee, 6/21/04, 2445:24-2446:9.

Professor McAfee has testified in several antitrust proceedings and has consulted for businesses on a range of economic issues. McAfee, 6/21/04, 2446:10-2447:12.

29. Steven Mills (IBM)

Mr. Mills is the senior vice president and group executive for IBM software. He has been with IBM for over 30 years, and has been head of the software group since 1995, when it was first formed. He is responsible for worldwide development, marketing and sale of software products, and for shaping IBM's overall software strategy. He reports to the IBM CEO, Sam Palmisano.P3193, Mills dep., 5/27/04, 6:8-7:24.

IBM's software group has approximately $12 billion in revenue and 40,000 employees, approximately 10,000 of whom are in sales. P3193, Mills dep., 5/27/04, 7:25- 8:14.

30. Donna Morea (AMS)

Ms. Morea is President of CGI/AMS. D7166, Morea dep., 5/7/04, 8:20-25.

AMS performs systems integration and consulting for government agencies. D7166, Morea dep., 5/7/04, 14:8-17.

31. Bipin Pushotam Patel (Ford)

Mr. Patel has worked for Ford for 23 years. He currently is responsible for Sarbanes-Oxley compliance for Information Technology within Ford, and is the Director for Application Development Services-India. Mr. Patel recently served as the company's Director of Management Systems for a four year period ending in December 2003, and was responsible for implementing Ford's global HRM and FMS software solutions. In that role, Mr. Patel reported directly to Ford's CIO, was responsible for supervising approximately 70 people, and had other personnel in other countries who matrix reported to him. Because of his experience at Ford and the importance of HRM and FMS software to the company, Mr. Patel stays current on relevant issues and what Ford is doing with the software in the company. P3041, Patel dep., 6/3/04, 5:21-6:1; 8:10-11; 26:2-8; 26:24-27:8; 28:20-29:10, 45:1-9.

Ford is a global manufacturer of cars and trucks, employing 350,000 persons and operating in 200 or so countries. Ford owns several large businesses, including a credit operation (Ford Credit) and a substantial car rental company (Hertz), as well as numerous automobile divisions responsible for specific brands (e.g., Ford Oval, Volvo, Land Rover, and Jaguar) within specific countries. P3041, Patel dep., 6/3/04, 8:12-20, 9:1-7.

As of December 2003, Ford had its PeopleSoft HRM software implemented for salaried personnel in the United States, Brazil, and Argentina, and for salaried and hourly personnel in Germany, the United Kingdom, and Sweden. The software is also running in about two thirds of Ford's Volvo operation in Belgium and in smaller areas in Asia Pacific. Ford is currently using PeopleSoft FMS software for general ledger in approximately 27 countries, encompassing more than 200 entities. Ford's objective is to implement PeopleSoft FM software for general ledger ("GL") and accounts receivable on a worldwide basis, together with Oracle's FMS software for payables. P3041, Patel dep., 6/3/04, 53:24-54:9; 28:23-29:10, 123:23-124:4.

32. Charles Phillips (Oracle)

Mr. Phillips is President of Oracle.

33. William Pollie (Microsoft)

Mr. Pollie is General Manager within Microsoft working in the Enterprise Business Sector. He reports to the General Manager in charge of enterprise sales operations, Phil Sorgen. Prior to joining Microsoft, Mr Pollie was the Vice President of Field Operations for SAP, he was at SAP for approximately 10 years. Pollie, 5/26/04, 20:10-13; 36:15-19; 13:18-14:4.

34. Frank Prestipino (Oracle)

Mr. Prestipino is Vice President for Product Marketing for Oracle, his position encompasses new growth opportunities, including mid-market opportunities. P3070, Prestipino dep., 5/18/04, 95:25-7:6.

35. Christopher Scarborough (Credit Suisse First Boston)

Mr. Scarborough has been a Vice President in the M&A group at Credit Suisse First Boston ("CSFB") for 18 months. He began at CSFB as an analyst in 1998, before being promoted to associate in 2000. P3346, Scarborough dep., 6/3/04, 9:14-10:11.

In early Spring 2003, Mr. Scarbourough was assigned to provide services to Oracle to pitch certain acquisition ideas. P3346, Scarborough dep., 6/3/04, 10:15-11:18.

36. Gary Spiekerman (Michigan)

Mr. Spiekerman has been manager of system support and operations for the Department of Information Technology, State of Michigan for five years. He is responsible for system management and application support for HRM as well as some peripheral systems. P3356, Spiekerman dep., 5/28/04, 5:10-6:4.

Michigan currently uses Lawson HRM software; however, the state has encountered performance issues with the Lawson product. P3356, Spiekerman dep., 5/28/04, 8:22-9:16.

37. Nancy Thomas (IBM)

Nancy Thomas is the Global and Americas Financial Management Solutions Leader at IBM, leading IBM's software selection implementation work for FMS software. Prior to the October 2002 acquisition of Price Waterhouse Coopers (PWC), Ms. Thomas was a PWC partner with FMS selection and implementation responsibilities. She has worked for clients in the distribution sector, consumer products, financial services, banking, insurance, investment management, and the industrial sector. Thomas, 6/9/04, 463:1-3; 462:20-25; 466:5-467:8.

The Financial Management Solutions, or FMS, group of IBM is part of Business Consulting Services, or BCS. BCS provides consulting and systems integration services. IBM's FMS group primarily focuses on SAP, Oracle and PeopleSoft applications to support financial management activities. Thomas, 6/9/04, 463:4-25.

38. Philip Wilmington (PeopleSoft)

Mr. Wilmington is Peoplesoft's Executive Vice President for the Americas, with responsibility for the US, Canada, and Latin America . He is directly responsible for Peoplesoft's product sales - including HRM software and FMS software - and pricing, as well as for the maintenance business. Sales organizations responsible for all regions of the US and the organizations responsible for sales to federal, state and local governments and to educational institutions report directly to him. Between his experience prior to joining PeopleSoft and his time at PeopleSoft, Mr. Wilmington has 20 years of experience in the application software business. Wilmington, 6/16/04, 1760:1-4; 1761:4-1763:21; 1760:5-9.

39. Randall Scott Wesson (AIMCO)

Mr. Wesson is Senior Vice President and Chief Information Officer of AIMCO. He has held that position since July of 1997. He is in charge of the budgeting and planning of technology, the assets, project management, as well as the purchase of hardware and software for the company. Previously, Mr. Wesson was Vice President of Information Services at Lincoln Properties. Before that, he was Director of Network Services at Lincoln Properties. Previously, he worked at ACP in the area of outsourcing. Wesson, 6/14/04, 1128:10-1129:3; 1148:11-23.

AIMCO is an apartment management company headquartered in Denver, Colorado. AIMCO is the nation's largest owner and operator of apartment buildings. AIMCO had approximately 2,000 apartment complexes, located in 47 states, Puerto Rico, and the District of Columbia. Approximately one million residents live in AIMCO apartment complexes. AIMCO has approximately 6,500 employees, and had 2003 revenue of approximately $1.5 billion. AIMCO has 25 regional centers, with properties that report up through them. In addition to being a real estate investment trust, AIMCO has a number of limited partnerships. Wesson, 6/14/04, 1129:4-25.

AIMCO currently runs PeopleSoft financial management system software and Lawson human resources management software. Wesson, 6/14/04, 1126:23-1128:9.

40. Curtis Wolfe (North Dakota)

Mr. Wolfe has been Chief Information Officer for the State of North Dakota since 1999. Mr. Wolfe is responsible for overseeing the state's technology initiatives as well as being the Director of the State Information Technology Department, the central IT authority for North Dakota. Mr. Wolfe has about 40 years of IT experience. Wolfe, 6/16/04, 1532:18- 1533:18.

North Dakota employs roughly 10,000 people in 58 agencies and departments. North Dakota has a biennial budget of $5 billion. Wolfe, 6/16/04, 1533:19-1534:4.

North Dakota uses PeopleSoft HRM and FMS software. Wolfe, 6/16/04, 1561:12-14.

41. Mark E. Zmijewski (University of Chicago, Graduate School of Business)

Dr. Zmijewski is Leon Carroll Marshall Professor of Accounting at the University of Chicago School of Business. He is also a Deputy Dean. P3319.

Dr. Zmijewski has taught Financial Accounting, Managerial Accounting, Corporate Finance, Security Market and Financial Statement Analysis and Cases in Financial Strategy. P3319.

Dr. Zmijewski is a Founding Partner and Principal of Chicago Partners LLC. P3319.

Dr. Zmijewski has testified and consulted in a number of antitrust cases and consulted for businesses on a range of valuation, synergy and efficiency assessment and other financial issues. Zmijewski, 07/1/04, 4512:15-4513:18; P3319.

2. ORACLE WITNESSES

Oracle has about 4500 applications customers. Catz, 6/25/04, 3494:24-3495:9. Oracle presented only two unpaid customer witness to testify live at trial from the perspective of a customer: Charles Peters of Emerson Electric and Brian Mearns of Bank of America.

1. Charles Peters (Emerson Electric)

Emerson enjoys a close long-term relationship with Oracle, including Mr. Peters personally consulting with Safra Catz regarding a CIO vacancy for the company, serving as a reference customer for Oracle on its website, and testifying in Europe and domestically in favor of the proposed merger with Peoplesoft. Peters, 6/14/04, 1268:10-14; 1242:15-21; 1247:17-24; 1275:8- 1276:1; P3022.

2. Robert Cushman (Lutheran Worker Benefit Plans)

Lutheran Worker Benefit Plans ("LWBP") is far smaller and simpler than the market 16 participants at issue in this litigation. LWBP is a department of the larger Lutheran-Church Missouri Synod - a tax-exempt religious institution. D8105, Cushman dep., 5/18/04, 82:2-14.

LWBP maintains a single office within the Church and currently employs approximately 90 full-time employees, including six full-time IT employees, including Mr. Cushman, an Oracle witness. D8105, Cushman dep., 5/18/04, 82:15-25; 84:1-20.

The entire annual IT budget of LWBP is approximately $2 million. D8105, Cushman dep., 5/18/04, 85:11-15.

3. Richard Knowles (SAP)

SAP can reasonably expect to benefit from a merger of Oracle and PeopleSoft, because it would become the only option to Oracle. Bradley, 6/9/04, 611:25- 612:13.

Oracle has an ongoing business relationship with SAP. P3033, Henley dep. 5/13/04, 81:12-82:20.


Dated: July 8, 2004Respectfully submitted,



  _______________/s/________________
Claude F. Scott, Esq.
Pam Cole, Esq. (CA Bar No. 208286)
Joseph M. Miller, Esq.
Owen M. Kendler, Esq.
U.S. DEPARTMENT OF JUSTICE
ANTITRUST DIVISION
450 Golden Gate Avenue, Rm. 10-0101
San Francisco, CA 94102-3478
(415) 436-6660
(415) 436-6683 (Fax)

Counsel for Plaintiff
United States of America



_______________/s/________________
Mark Tobey, Esq.
Assistant Attorney General
Office of the Attorney General
P.O. Box 12548
Austin, Texas 78711-2548
(512) 463-2185
(512) 320-0975 (Fax)

Mark J. Bennett, Esq.
Attorney General
State of Hawaii
425 Queen Street
Honolulu, Hawaii 96813
(808) 586-1600
(808) 586-1239 (Fax)

Timothy E. Moran, Esq.
Assistant Attorney General
Consumer Protection and Antitrust Division
One Ashburton Place
Boston, MA 02108
(617) 727-2200, ext. 2516
(617) 727-5765 (Fax)

Kristen M. Olsen, Esq.
Assistant Attorney General
Office of the Attorney General of Minnesota
445 Minnesota Street, Suite 1200
St. Paul, Minnesota 55101-2130
(651) 296-2921
(651) 282-5437 (Fax)

Jay L. Himes, Esq.
Chief, Antitrust Bureau
Office of the Attorney General of New York
120 Broadway, 26th Floor
New York, NY 10271
(212) 416-8282
(212) 416-6015 (Fax)

Todd A. Sattler, Esq.
Assistant Attorney General
Consumer Protection and Antitrust Division
600 E. Boulevard Ave., Dept. 125
Bismark, ND 58505-0040
(701) 328-2811
(701) 328-3535 (Fax)

Steven M. Rutstein, Esq.
Assistant Attorney General
55 Elm Street
Hartford, CT 06106
(860)808-5169
(860) 808-5033 (Fax)

Paul F. Novak, Esq.
Assistant Attorney General In Charge
Special Litigation Division
Michigan Department of Attorney General
P.O. Box 30212
Lansing, MI 48909
(517) 335-4809
(517) 373-9860 (Fax)

Mitchell L. Gentile, Esq.
Assistant Attorney General
Antitrust Section
Office of the Attorney General
150 E. Gay St., 20th Floor
Columbus, OH 43215
(614) 466-4328
(614) 995-0266 (Fax)

Ellen S. Cooper, Esq.
Assistant Attorney General
Chief, Antitrust Division
State of Maryland
200 St. Paul Place, 19th Floor
Baltimore, MD 21202
(410) 576-6470
(410) 576-7830 (Fax)

Counsel for Plaintiff States


CERTIFICATE OF SERVICE

I hereby certify that on July 8, 2004, 1 caused a true and correct copy of the foregoing
PLAINTIFFS' PROPOSED FINDINGS OF FACT AND CONCLUSIONS OF LAW, FILED UNDER SEAL, to be served on the individuals listed below.


       __________________________________
Joli M. Wilson

Via E-Mail:

Daniel Wall
Latham & Watkins
505 Montgomery Street, Suite 1900
San Francisco, CA 94111-2562

Mark Tobey
Office of the Attorney General
300 West 15th Street
9th Floor
Austin, Texas 78701

Jay L. Himes, Esq.
Chief, Antitrust Bureau
Office of the Attorney General of New York
120 Broadway, 26th Floor
New York, NY 10271


CERTIFICATE OF SERVICE

I hereby certify that on July 8, 2004, 1 caused a true and correct copy of the foregoing
PLAINTIFFS' PROPOSED FINDINGS OF FACT AND CONCLUSIONS OF LAW, PUBLIC VERSON, to be served on the individuals listed below.


       __________________________________
Joli M. Wilson

Via E-Mail:

Daniel Wall
Latham & Watkins
505 Montgomery Street, Suite 1900
San Francisco, CA 94111-2562

Mark Tobey
Office of the Attorney General
300 West 15th Street
9th Floor
Austin, Texas 78701

Jay L. Himes, Esq.
Chief, Antitrust Bureau
Office of the Attorney General of New York
120 Broadway, 26th Floor
New York, NY 10271