BMC Software, Inc. v. Int'l Bus. Machs. Corp.

Full title: BMC Software, Inc., Plaintiff, v. International Business Machines…

Court: United States District Court, Southern District of Texas

Date published: May 30, 2022

Civil Action H-17-2254 (S.D. Tex. May. 30, 2022)

Opinion

Civil Action H-17-2254

05-30-2022

BMC Software, Inc., Plaintiff, v. International Business Machines Corporation, Defendant.

Honorable Gray H. Miller Senior United States District Judge

FINDINGS OF FACT AND CONCLUSIONS OF LAW

Honorable Gray H. Miller Senior United States District Judge

I. Procedural History of the Case

BMC filed this case against IBM on July 21, 2017. Dkt. 1. In its initial and first amended complaints, BMC applied for a preliminary injunction to prevent AT&T and IBM from completing Project Swallowtail, an AT&T-initiated project involving the migration of BMC software to IBM software that BMC claimed caused ongoing damages and infringed on its trade secrets.¹ Id. at 19- 32; see also Dkts. 37, 38, 92, 170. United States Magistrate Judge Nancy K. Johnson held a preliminary injunction hearing from November 28 through December 1, 2017. Dkt. 219 at 1. Ultimately, Judge Johnson concluded that BMC failed to establish a substantial threat of irreparable injury if Project Swallowtail were allowed to proceed. Id. at 26-27. To the extent that BMC argued that Project Swallowtail caused ongoing damages, the court found that the evidence was “too speculative to form the factual basis for the issuance of an injunction.” Id. at 27. Turning to BMC's trade secrets claim, the court concluded that BMC's software patches did not contain protected source code, even as they were subject to confidentiality restrictions found in the 2008 Master Licensing Agreement (the “MLA”), one of the contracts governing the parties' relationship. See Id. at 29-30. This court adopted Judge Johnson's factual findings in full and her conclusion that BMC failed to show a substantial threat of injury justifying the extraordinary equitable relief it sought. Dkt. 270 at 9-11.

1 AT&T was one of BMC and IBM's mutual customers. Though this case concerns an AT&T-initiated project, AT&T is not a party to this case.

BMC also sought a jury trial through its initial and first amended complaints. See Dkts. 1 at 33, 37 at 55. IBM moved to strike BMC's jury demand on the theory that the MLA's waiver of jury trial provision applied to BMC's claims under the 2015 Outsourcing Attachment (the “2015 OA” or the “OA”), another contract governing the parties' relationship. See Dkt. 213. Looking to the contracts' plain meaning, Judge Johnson agreed that the MLA and the 2015 OA formed an integrated whole. See Dkt. 256 at 14-21. On January 25, 2019, this court adopted Judge Johnson's conclusion in full because the “straightforward language of the 2015 OA indicates the parties' clear intention for the 2015 OA and the MLA to form an integrated contract.” Dkt. 287 at 6. BMC filed a Second Amended Complaint (the “SAC”) on February 22, 2019, wherein it raised twelve causes of action, including: breach of sections 5.4, 5.1, and 1.1 of the 2015 OA; anticipatory breach of contract; breach of section 8 of the MLA; fraudulent inducement; breach of the duty of good faith and fair dealing; tortious interference; common law misappropriation of trade secrets; misappropriation of trade secrets under Texas Uniform Trade Secrets Act (“TUTSA”) and Federal Defend Trade Secrets Act (“DTSA”); and unfair competition. See Dkt. 295 at 23-48. In its Answer, IBM raised ten affirmative defenses including, in relevant part, that section 5.4 of the 2015 OA is unenforceable under New York law and that the equitable doctrine of unclean hands barred BMC's claims because BMC failed to perform under the 2015 OA. See Dkt. 299 at 22-27.

In July 2020, BMC moved for partial summary judgment on IBM's breach of sections 1.1 and 5.4 of the 2015 OA (Dkt. 381), construction on damages limitations provisions in the MLA (Dkt. 385), certain IBM affirmative defenses (Dkt. 387), and IBM's counterclaim for breach of the 2015 OA's most-favored customer provision (dkt. 391). With respect to its motion for partial summary judgment on the MLA's damage limitations, BMC argued that MLA section 10 authorized the recovery of licensing fees that are “payable” to BMC. Dkt. 385 at 16-18. In its motion on IBM's affirmative defenses, BMC did not move for summary judgment on IBM's defense that section 5.4 is an unenforceable restrictive covenant under New York law. See generally Dkt. 387.

Likewise, IBM moved for partial summary judgment on its counterclaims (dkt. 394) and summary judgment on all of BMC's claims (dkt. 396). Echoing its affirmative defense, IBM argued in the former that BMC's interpretation of the 2015 OA's non-displacement provision rendered it an unenforceable restrictive covenant under New York contract law. See Dkt. 394 at 21. In the latter motion, IBM claimed that all of BMC's claims failed as a matter of law. See generally Dkt. 396.

On referral from this court, United States Magistrate Judge Christina Bryan issued a Memorandum and Recommendation (the “M&R”) adjudicating the parties' competing motions. See generally Dkt. 561. In relevant part, the Magistrate Court: (1) recommended denying summary judgment to both parties on BMC's claims for breach of sections 1.1, 5.1, and 5.4 of the 2015 OA after finding that the 2015 OA was ambiguous;² (2) recommended denying summary judgment to IBM on BMC's claim for breach of MLA section 8, fraudulent inducement, and BMC's claims for breach of TUTSA, DTSA, and common law unfair competition; (3) recommended granting summary judgment on BMC's lost profits damages model because it was barred by MLA section 10; (4) recommended granting summary judgment on certain of IBM's affirmative defenses except ratification; and (5) recommended denying summary judgment to both parties on IBM's counterclaim for breach of the 2015 OA's most favored customer provision. Id.

2 BMC only moved for summary judgment on its claims for breach of sections 1.1 and 5.4, even though in the SAC BMC also raised a breach claim related to section 5.1. See Dkts. 381, 299. Because IBM moved for summary judgment on all of BMC's claims, Judge Bryan analyzed whether IBM was entitled to summary judgment for BMC's breach of section 5.1 claim.

After reviewing the parties' objections, this court adopted the M&R only in part. See Dkt. 586. The court granted BMC summary judgment on its claim for breach of section 5.4 of the 2015 OA after finding that the 2015 OA was unambiguous and there was no dispute of material fact that IBM displaced BMC's products with its own. See Id. at 2-5. The court also determined that section 5.1 of the OA was unambiguous. Id. at 5-6. Because the court determined that sections 5.1 and 5.4 were unambiguous, it did not defer ruling on BMC's claim for breach of section 1.1. Id. at 6. Finding that section 1.1 “merely puts IBM to an election regarding how it would use BMC products in relation to its provision of IT Services at AT&T, ” the court concluded that BMC could not “proceed on an independent claim for breach of section 1.1.”³ Id. The court next adopted the M&R's recommendation that MLA section 9's consequential damages limitation barred BMC's lost profits model. Id. at 8. However, the court noted that this decision would not preclude BMC “from arguing that the damage limitation provisions are unenforceable if it succeeds on its claim for fraudulent inducement of the 2015 OA.” Id. Turning next to the MLA's “paid or payable” language, the court did not determine whether BMC's licensing fees were recoverable. See Id. at 9. Finally, with respect to IBM's counterclaim for breach of the 2015 OA's most favored customer provision, the court concluded that IBM “failed to raise a genuine issue of material fact as to the existence of a comparable competitor or its damages.” Id. at 9.

3 Importantly, the court noted that this finding would not “preclude BMC from arguing that section 1.1 provides the framework for benefit-of-the-bargain damages for IBM's breach of section 5.4.” Dkt. 589 at 6.

Following further briefing from the parties, the court clarified in a subsequent order that: (1) its finding that the 2015 OA was unambiguous constituted objective evidence that there was a meeting of the minds; (2) the determination of IBM's breach of section 5.4 did not resolve the other elements of BMC's contract claims, including its own performance; (3) whether section 5.4 is an unenforceable restrictive covenant was not squarely resolved at summary judgment because BMC did not technically move for summary judgment on IBM's unenforceability affirmative defense; (4) whether IBM displaced BMC products for the “sole purpose” of supporting AT&T- an issue critical to BMC's breach of section 5.1 claim-remained to be tried; and (5) the breach, causation, and damages elements of BMC's breach of MLA section 8 claim remained to be tried. Dkt. 603 2-10.

BMC has six remaining claims against IBM. Of those six, three are breach of contract claims: (1) IBM's breach of section 5.4 of the 2015 OA; (2) IBM's breach of section 5.1 of the 2015 OA; (3) IBM's breach of section 8 of the MLA. BMC also claims that IBM fraudulently induced it into signing the 2015 OA. Finally, BMC claims that IBM misappropriated its trade secrets under TUTSA and DTSA, and, for any information that does not qualify as a trade secret, BMC asserts a claim for common law unfair competition by misappropriation. See Dkts. 295, 561, 586, 603.

IBM denies it breached the MLA or the 2015 OA, fraudulently induced BMC into the 2015 OA, and misappropriated BMC's trade secrets. In addition, IBM asserts that even if BMC can prove some or all its claims, it cannot show any damages. See Dkt. 299.

As a result of the parties' waiver of a jury trial, this case was tried to the court from March 14 through March 24, 2022. Both parties moved for judgment on partial findings. See Dkts. 665, 751, 752. Both parties filed proposed findings of fact and conclusions of law, as well as multiple post-trial briefs on the remaining issues. See Dkts. 668 (BMC Software's Trial Brief on Standard for Fraudulent-Inducement Claim), 687 (Defendant IBM's Post-Trial Brief Addressing Damages Issues Raised at Trial), 689 (IBM's Post-Trial Proposed Findings of Fact and Conclusions of Law), 694 (Defendant IBM's Posttrial Brief and Response to BMC's Motion for Judgment on Partial Findings), 723 (BMC Software, Inc.'s Corrected Post-Trial Proposed Findings of Fact and Conclusions of Law).

The parties agreed to the following facts:

a) BMC is a Delaware corporation and a citizen of Texas, with its principal place of business located in Texas.

b) IBM is a New York corporation and a citizen of New York, with its principal place of business located in New York. IBM also does business in Texas.

c) AT&T and BMC entered into a Master Purchase Agreement in 2007.

d) IBM was the IT Outsourcer for AT&T's mainframe environment for many years.

e) The current AT&T is the result of various mergers, including the acquisition by SBC of two other “Baby Bell” companies.

f) BMC and IBM entered into the MLA in March 2008.

g) The MLA is a valid contract in existence between BMC and IBM.

h) BMC and IBM entered into a 2008 Outsourcing Attachment to the MLA.

i) BMC and IBM entered into a 2013 Outsourcing Attachment.

j) IBM and AT&T entered into an agreement to perform Project Swallowtail on June 26, 2015.

k) BMC and IBM entered into the 2015 OA on September 30, 2015.

l) The 2015 OA is a valid contract in existence between BMC and IBM.

m) Project Swallowtail is complete. It resulted in the replacement of fourteen BMC mainframe software products with IBM's mainframe software products, the replacement of five BMC mainframe software products with third-party products, and the retirement of one additional BMC mainframe software product.

n) BMC is not pursuing any claims for equitable relief.

The matter is fully briefed and tried. After thoroughly reviewing this case's extensive record, including the applicable law, the testimony at trial, designated depositions, and admitted exhibits, the court makes the following Findings of Fact and Conclusions of Law.⁴

4 If any finding of fact is more properly considered a conclusion of law, it shall be so deemed. Likewise, if any conclusion of law is more properly deemed a finding of fact, it shall be so deemed.

II. Findings of Fact

A. The Parties and Their Relationships

1. BMC is a private, Houston-based software company that, among other things, develops and licenses proprietary mainframe software products that help customers manage and automate business operations. Dkt. 561 at 1; Trial Tr. 137:17-23, 149:25-150:6, Mar. 14, 2022 (Ah Chu); Trial Tr. 102:9-103:5, Mar. 15, 2022 (Jones). A “mainframe” is a high-performance computer-a piece of hardware made of integrated circuits-that can process massive amounts of information at once. Trial Tr. 140:23-25, Mar. 14, 2022 (Ah Chu); Trial Tr. 57:01-59:25, Mar. 16, 2022 (Román). When first produced, mainframes were the size of refrigerators. Trial Tr. 58:12-13, Mar. 16, 2022 (Román). Today, mainframes are smaller pieces of hardware, though still significantly larger than the laptops familiar to most consumers. See id.

2. Because mainframes exceed the needs of most computer users, they are used primarily by large organizations for critical applications that require high volumes of data processing. Trial Tr. 144:17-24, Mar. 14, 2022 (Ah Chu); Trial Tr. 103:12-16, Mar. 15, 2022 (Jones); Trial Tr. 57:25-58:16, Mar. 17, 2022 (Skinner). A mainframe's capacity is measured by the millions of instructions per second (“MIPS”) that it can process. Trial Tr. 140:15-20, Mar. 14, 2022 (Ah Chu). Depending on what a software product does, its efficiency can be more-or-less gauged by the number of MIPS it uses, with more efficient software products using fewer MIPS. Trial Tr. 138:06-08, Mar. 15, 2022 (Ah Chu).

3. BMC's mainframe software products, and the accompanying services that BMC provides, are used by its customers to run, manage, and secure operations on their mainframe computers. Trial Tr. 137:17-23, Mar. 14, 2022 (Ah Chu). BMC's software products are like the “system utilities” familiar to some PC users: they help customers manage computer memory usage, backup data, automate tasks, and the like. Trial Tr. 60:13-63:20, Mar. 16, 2022 (Román). BMC's mainframe business unit is responsible for selling BMC's mainframe software products to “end user” customers-the business entities that are using the software on their mainframe computers. See Appleby Dep. 8:08-9:10.⁵

5 The parties designated twenty-four depositions in lieu of presenting those witnesses at trial.

4. BMC's entire business model is predicated on the creation and licensing of its software. Trial Tr. 140:01-04, Mar. 14, 2022 (Ah Chu). During trial, Raul Ah Chu, BMC's Vice President of Global Outsourcers and Systems Integrated Group, testified that “software [intellectual property] is the entire company. So, every dollar…[of]…revenue that BMC generates is directly related to licensing that IP.” Id.

5. IBM is a public, New York-based information technology company. Dkt. 561 at 1. With more than 345, 000 employees and $73 billion in annual revenue as of 2020, IBM is one of the world's largest companies and is a “heavyweight” puncher in the IT industry. Trial Tr. 62:23-25, Mar. 15, 2022 (Román). In relevant part, IBM manufactures mainframe computers, creates mainframe software, and provides IT outsourcing services, including to BMC customers. Dkt. 561 at 1; Trial Tr. 140:14-16, Mar. 14, 2022 (Ah Chu). “IT outsourcing” refers to a business's use of an external service provider to operate and manage its information technology (IT)-enabled business processes, application services, and infrastructure solutions. Trial Tr. 138:17-20, Mar. 14, 2022 (Ah Chu); Trial Tr. 26:23-27:07, Mar. 16, 2022 (McGuinn); Trial Tr. 65:13-65:19 Mar. 16, 2022 (Román); Trial Tr. 50:20-51:07, Mar. 17, 2022 (Skinner). In short, IT outsourcers are tasked with keeping the client's computer systems up and running. Trial Tr. 62:10-63:11, Mar. 16, 2022 (Román).

6. As a mainframe software developer, IBM directly competes with BMC “almost one for one.” Trial Tr. 145:16-17, Mar. 14, 2022 (Ah Chu). In its role as an IT outsourcer, IBM is paid to operate and maintain its customers' mainframe IT services, including customers who employ BMC software on their mainframes. Dkt. 561 at 1; see Bergdoll Dep. at 58:08-13 (“IBM [is] . . . in essence, a monopoly player in the mainframe business because they provide the hardware, they provide the system software, they provide the database software, and they have a host of offerings that other [Independent Software Vendors] might have . . . .”); Trial Tr. 138:17- 20, Mar. 14, 2022 (Ah Chu); Trial Tr. 117:07-19, Mar. 15, 2022 (Jones); Trial Tr. 26:17-27:07, Mar. 16, 2022 (McGuinn); Trial Tr. 65:05-19, Mar. 16, 2022 (Román); Trial Tr. 50:20-51:07, Mar. 17, 2022 (Skinner).

7. IBM provides IT outsourcing services to its customers through its Global Technology Services (“GTS”) business, also known as its Strategic Outsourcing Division. Shell Dep. 48:01-05; Cattanach Dep. 14:13-20; Stafford Dep. 5:11-25; Trial Tr. 145:22-146:03, Mar. 14, 2022 (Ah Chu); Trial Tr. 105:01-25, 106:18-24, Mar. 15, 2022 (Jones); Trial Tr. 50:16-51:05, 50:08-12, Mar. 17, 2022 (Skinner). GTS' functions are now housed in a separate spin-off business named Kendryl. Trial Tr. 51:10-12, Mar. 16, 2022 (Román). IBM's outsourcing division generates approximately $20 billion in revenue per year, and IBM functions as the largest outsourcer of mainframe services in the world. Trial Tr. 100:23-24, Mar. 15, 2022 (Jones).

8. Although other companies offer IT outsourcing services that use BMC products, IBM is the only outsourcer that both uses BMC products to service its clients and offers a full suite of competing mainframe software products large enough to replace a client's entire portfolio of BMC products.⁶ Bergdoll Dep. 187:08-24; see Trial Tr. 145:17-21, Mar. 14, 2022 (Ah Chu). BMC does not offer competing IT outsourcing services, though it “supports and manages the relationship between [itself] and . . . global outsourcers, ” like IBM, through its Global Outsourcers and Systems Integrators (“GOSI”) team. Trial Tr. 138:08-15, Mar. 14, 2022 (Ah Chu). By virtue of these two complementary roles, IBM-as an outsourcer-can acquire unique knowledge about how a competitor's software operates on a mutual customer's mainframe system. Trial Tr. 63:21- 25, Mar. 16, 2022 (Román).

6 Computer Associates (“CA”), another software development company, directly competes with BMC's mainframe software portfolio but does not manufacture mainframes or provide IT outsourcing services. Trial Tr. 100:12-101:15, Mar. 15, 2022 (Jones).

9. Hundreds of BMC's software customers also use IBM's IT outsourcing services to manage their mainframe computers' operations. Dkt. 561 at 2; PX250 at BMC-000070461 (list sent by IBM to BMC identifying certain mutual customers); PX479 at IBM00000390 (same); see Stafford Dep. 5:16-6:22; Trial Tr. 186:24-187:04, Mar. 14, 2022 (Ah Chu). As a general matter, when this occurs, the mutual customer contracts with BMC to own licenses to the BMC mainframe software products, but separately contracts with IBM in its outsourcing capacity to operate BMC products in the customer's mainframe environment and interact with BMC's technical support when issues arise. Shell Dep. 34:12-20; Trial Tr. 103:06-09, 105:04-25, Mar. 15, 2022 (Jones); Trial Tr. 26:13-27:07, Mar. 16, 2022 (McGuinn); Trial Tr. 65:05-19, Mar. 16, 2022 (Román).

10. Because they are direct competitors in the software development space, but partners when a mutual customer hires IBM as an outsourcer to operate mainframes running BMC software, BMC and IBM have “a complex, multifaceted relationship.” Trial Tr. 145:24-25, Mar. 14, 2022 (Ah Chu); see also Bergdoll Dep. at 53:12-21; id. at 54:05-11 (noting that IBM is different from other outsourcers because “IBM acts in multiple behalves in part for their customers”); Stafford Dep. 221:14-20 (noting that BMC is one of IBM's top software suppliers). Augmenting this complexity further, IBM was also, at the time, a “significant customer” of BMC's insofar as it purchased software licenses or certain licensing rights from BMC. Trial Tr. 101:24-102:02, Mar. 15, 2022 (Jones).

B. AT&T

11. One of BMC and IBM's mutual customers is AT&T. AT&T was one of BMC's biggest mainframe software clients, and “was one of the most strategic and tenured customers that [BMC] had.” Trial Tr. 187:22-23, Mar. 14, 2022 (Ah Chu). BMC considered AT&T to be a “showcase” account. DX128 (describing “showcase” accounts as “key strategic clients” to BMC); see also Appleby Dep. 51:25-52:12; PX205 at BMC-000010535 (“AT&T is [BMC's] largest commercial account.”); Poole Dep. 186:23-187:02 (noting that AT&T was a valued client to BMC).

12. Prior to the events giving rise to this case, BMC provided mainframe software to AT&T. Conway Dep. 21:18-20. For this software, AT&T held a perpetual license granting it and its IT outsourcers the right to use BMC's software. Trial Tr. 195:01-11, Mar. 15, 2022 (Jones) (testifying that AT&T purchased a perpetual license to use BMC's software). AT&T also held a separate “software maintenance and support” contract from BMC that entitled AT&T (and IBM and other AT&T services providers) to contact BMC customer service for assistance when problems arose relating to AT&T's use of BMC's products. See DX018 (2014 Mainframe Order).

13. Since 1999, AT&T has used IBM as its mainframe IT outsourcer. Shell Dep. 23:18-21; Ulaszek Dep. 9:23-25; see Conway Dep. 11:08-24, 13:14-23. IBM's revenue from AT&T is substantial; as of 2017, IBM received more than $100 million per month from AT&T. Ulaszek Dep. 17:12-18:02 (testifying that IBM received “in th[e] neighborhood” of over $100 million in revenue a month from AT&T); see also Brickhaus Dep. 83:03-84:05 (IBM monthly bills to AT&T for mainframe services alone are “approximately $10 million a month”); Knight Dep. 66:10-21 (noting the importance of AT&T to IBM). Over the last seven years, AT&T accounted for at least $1 billion of IBM's outsourcing division's revenue. See PX222 at IBM00007098-99 (internal slideshow discussing “contract value” figures); PX216 at IBM00007356-57 (same). So, understandably, AT&T is also an important client for IBM. See Knight Dep. 66:10-21 (noting the importance of AT&T to IBM).

14. To perform as an IT outsourcer when a client used BMC mainframe software, IBM required “a contractual vehicle . . . to conduct . . . business” with “BMC and any clients [in] an outsourcing environment.” Craig Dep. 26:01-07. BMC provided IBM that ability under the 2015 OA's “no fee” Access and Use option on the condition that it not then displace BMC's products with its own products. See PX4; Schulman Dep. 145:10-147:05 (explaining that the parties' agreements prevented IBM from “leveraging its position and access to displace BMC”).

15. As AT&T's IT outsourcer, IBM manages and operates AT&T's mainframe operations-“all the application jobs, ” database support requirements, “hardware refreshes, ” and other “basic IT stuff”-and the software products on the AT&T mainframe computers, including the BMC products AT&T previously used. Shell Dep. 34:12-20. IBM's responsibilities also include resolving technical issues as they arise by opening technical support cases directly with any Independent Software Vendor (“ISV”), including BMC, that licenses software installed in AT&T's mainframe environment. See Conway Dep. 11:25-12:06. For example, when IBM had a question about the way a BMC product functioned in the AT&T environment, IBM would contact BMC's support services, and BMC would diagnose the problem and provide instructions to IBM on how to resolve the issue. See Sessarego Dep. 5:25-7:11; Trial Tr. 22:18-26:01, Mar. 16, 2022 (McGuinn); Trial Tr. 96:09-97:04, Mar. 16, 2022 (Román).

16. The close access that IBM personnel had to AT&T's mainframe environment and experience with how BMC products operate in that environment gave IBM exclusive insights into how the software products AT&T used, including BMC products, worked under the operational demands of AT&T's computing environment. See Shell Dep. 35:23-37:01; see also Bergdoll Dep. 187:08-24 (noting the “unique nature” of IBM as an IT outsourcer that also sells competing products, as “[n]o other outsourcer has competing products”).

C. The Parties' Contractual Relationship

17. This case, however, principally involves contract disputes between BMC and IBM. Relevant to those disputes are several contracts: the MLA; the 2015 OA; BMC's standard End User License Agreement (the “EULA”); and the BMC-AT&T End User License Agreement (the “AT&T EULA”).

1.The MLA

18. The contractual structure governing the business relationship between BMC and IBM includes the MLA. PX1. BMC and IBM executed the MLA on March 31, 2008, to streamline the way they do business by agreeing to a contractual framework that governs both IBM's licensing of BMC's software products and the parties' business relationship worldwide. PX1; see also Craig Dep. 25:14-26:7;⁷ Trial Tr. 141:25-142:05, Mar. 14, 2022 (Ah Chu) (Q: “So what generally governs BMC's relationships with companies that provide outsourcing services using BMC products? What agreements are issue?” A: “It's-the agreements [are] typically the MLA…and then the OA.”).

7 Robert Craig, a Procurement Project Manager for IBM, has been responsible for managing IBM's relationship with BMC at the global level since approximately 2006. See Craig Dep. 4:09- 4:23.

19. In recognition of IBM's preeminent role as an IT outsourcer, and “[s]ubject to the terms of an executed Outsourcing Attachment, ” the MLA “grant[s]” IBM a “license” for the “provision of services to a third party, ” among other rights. PX1 at 1. But the use of this license is subject to some restrictions.

a. MLA Section 8: Proprietary Rights and Confidentiality

20. Section 8: Proprietary Rights and Confidentiality, limits IBM's ability to use confidential information about BMC's products that it might receive during the parties' business relationship:

“Confidential Information” means all proprietary or confidential information that is disclosed to the recipient (“Recipient”) by the discloser (“Discloser”), and includes, among other things (i) any and all information relating to products or services provided by a Discloser, its financial information, software code, flow charts, techniques, specifications, development and marketing plans, strategies, and forecasts; [and](ii) as to BMC, and its licensors, the Product and any third party software provided with the Product…Confidential Information does not include information that Recipient can show: (a) was rightfully in Recipient's possession without any obligation of confidentiality before receipt from the Discloser; [or] (b) is or becomes a matter of public knowledge through no fault of Recipient…All materials containing Confidential Information must have a restrictive marking of the Discloser at the time of disclosure…Recipient may not disclose Confidential Information of Discloser to any third party, however, the Recipient may disclose Confidential Information to: i) its employees and employees of its parent and majority owned affiliates who have a need to know; and ii) any other party with the Discloser's prior written consent.

PX1 at 3 (emphasis added). The MLA defines “Product” as “the object code of the software and all accompanying Documentation delivered to [IBM], including all items delivered by BMC to [IBM] under [the support services program].” PX1 §1. “Documentation” is defined to include “the technical publications relating to the software, such as release notes, reference, user, installation, system administrator and technical guidelines, included with the Product.” Id.

b. MLA Section 9: Disclaimer of Damages

21. Just as the MLA limits the scope of acceptable behavior between the parties, so too does it limit their remedies in the event of breach in two important respects. First, section 9: Disclaimer of Damages provides that:

Except for violation of proprietary rights and confidentiality (section 8) and infringement claims (section 12), neither party, its affiliates or BMC's licensors are liable for any special, indirect, incidental, punitive or consequential damages relating to or arising out of this agreement, support, the product, or any third party code or software provided with the product (including, without limitation, lost profits, lost computer usage times, and damage to or loss of use of data), even if advised of the possibility of such damages, and irrespective of negligence of a party or whether such damages result from a claim arising under tort or contract law. The foregoing limitation of liability will apply unless otherwise required by the local law of the country where the products are ordered.

Id. at 3.

c. MLA Section 10: Limits on Liability

22. The very next section-section 10: Limits of Liability-limits “[e]ach party's liability arising out of or related to [the MLA] agreement, the product, or the use of the product . . . . to the greater of $5,000,000 or the amount paid or payable by the customer for the license to the applicable product giving rising to the claim.” Id. at 10.

2. The 2015 OA

23. As contemplated by the MLA, the parties signed several outsourcing attachments to define the terms of their business relationship. Most relevant among them is the 2015 OA. PX4. Titled as an “Attachment” to the “Master License Agreement, ” the 2015 OA “addresses the terms under which BMC grants to [IBM] the right to use the Products in its IT Services business.”⁸ Id. Acknowledging the parties' mutual “intent . . . to operate under a common set of terms and conditions on a worldwide basis, ” the 2015 OA, “along with the [MLA], ” was meant to “serve as the master worldwide document[] for all applicable” software and software licensing orders. Id. Were there any doubt, though, the 2015 OA makes clear that “[a]ll Products provided under the OA are governed by the OA, the [MLA], any Participation Agreement, and any applicable Order.” Id. (“The parties agree that . . . this OA will govern all IT Services engagements by Customer and that all prior Orders, Access and Use Agreements . . . entered into under the Prior Agreements will now be governed by this OA and the [MLA].”). In short, the 2015 OA functions as the “single point of control for the business terms related to the subject matter” contained in its four corners. Id.

8 As Craig testified in his deposition, the outsourcing attachments “typically cover[ed] operational matters” and guided IBM in managing the day-to-day relationship with BMC. See Craig Dep. 39:24-40:03.

24. The kinds of “IT Services” that IBM could pick from were detailed in section 1.1 of the 2015 OA, “IT Services Options.” That section required IBM to “elect one of . . . five options regarding its use” of BMC's products: “(a) Access and Use - (sections 5.1, 5.2, 5.3, and 5.4), (b) License Suspension - (section 5.5), (c) Customer owned Products - (sections 6, 7, and 8), (d) Mirror Order - (section 11), or (e) Shared Hosting - (section 12).” Id. §1.1.

25. Pursuant to section 3 of the OA, IBM “may exercise its rights to access and use the Products (including BMC Customer Licenses) through its Authorized Users.”⁹ Id. §3.

9 “Authorized Users” is defined in section 2 of the OA as IBM's “employees, agents, contractors and third party service providers. PX4.

26. Section 3.1 supplements section 3 and ensures that IBM “may allow its Clients [i.e., AT&T in this case] and Clients' employees, agents, contractors and third party service providers . . . to have worldwide access to and use of the Products.” Id.

27. Section 5 of the 2015 OA provides that it “will apply to [IBM]'s access and use of the BMC Customer Licenses and apply retroactively to the point of first access by [IBM].” Id. at § 5.

28. IBM availed itself of BMC's “Access and Use” option. Section 5.1 provided that “BMC will allow [IBM] to use, access, install and have operational responsibility of the BMC Customer Licenses . . . under the terms of the BMC Customer's license agreement with BMC for no fee . . . provided that the BMC Customer Licenses are used solely for the purposes of supporting the BMC Customer who owns such licenses.”¹⁰ PX4 §5.1 (emphasis added). While this language limited how IBM could use AT&T's license, it required BMC to make certain disclosures to IBM:

10 “BMC Customer” is defined as “a third party which licenses the Products and which becomes a Client at commencement of or during delivery of IT Services.” PX4 at 1. Here, the “BMC Customer” is AT&T, for whom IBM served as an IT outsourcer.

Except as set forth herein, the BMC Customer Licenses will continue to be governed by the terms, conditions and discounts of the BMC license agreement between BMC and the BMC Customer; notwithstanding the terms of the Agreement and the OA, Customer shall be bound by such terms of such license agreement. BMC will make available to Customer a representative copy of its typical BMC Customer license agreement (“BMC EULA”) and will inform Customer in writing of any material differences between the BMC EULA and BMC Customer's agreement after Customer provides written notice of Access and Use in accordance with the process detailed in section 5.3.

Id. Elsewhere, in section 5.4, the 2015 OA also limited IBM's ability to displace BMC's products. Specifically, the contract's non-displacement provision

applies only to [IBM's] Access and Use of BMC Customer Licenses by [IBM's] strategic outsourcing division (or its successor) for the BMC Customers listed on Exhibit K (the “Exhibit K Customers”). Subject to the foregoing, [IBM] agrees that, while [IBM] cannot displace any BMC Customer Licenses with [IBM] products, [IBM] may discontinue use of BMC Customer Licenses for other valid business reasons. All terms in sections 5.1, 5.2, and 5.3 apply to [IBM's] use of BMC Customer licenses belonging to any Exhibit K Customers.

Id. §5.4.

29. These restrictions and obligations were absent under other election options outlined in section 1.1, namely “(c) Customer owned Products - (sections 6, 7, and 8).” Id. For a price, IBM could purchase its own license unencumbered by the restrictions attached to its “no fee” use of AT&T's license under section 5. Should IBM want its own license, section 8.1 stated that it “will be entitled to a global minimum discount of 72% off the Listed Price in Exhibit H for all purchases of new licenses.” Id. ¹¹ Exhibit H (Systems Management - Mainframe Products), in turn, lists the standard pricing available for each of BMC's products. Id.

11 The minimum discount to which IBM would be “entitled” for BMC products would downgrade to 35% off the list price in the event the parties failed to reach a renewal agreement following the OA's termination date. PX4 at 10.

30. While the MLA contained various liability and damages restrictions, the 2015 OA also included a full-scale release for IBM for “claims of any nature whatsoever arising from or related to [IBM's] performance or failure of performance under” prior outsourcing attachments and the MLA, including “any and all Claims related to Customer displacements of BMC Customer Licenses to date.” Id.

3.The BMC-AT&T Licensing Agreement

31. AT&T's mainframe utility software-related agreements with BMC include a 2007 Master Purchase Agreement (“MPA”). DX10. The MPA “grants to AT&T a nonexclusive, irrevocable (except as provided), perpetual . . . Enterprise-Wide license to use the Standard Software.” Id. at 44. Section 3.3 of the MPA gave AT&T certain assignment and delegation rights. Specifically,

[n]either Party may assign, delegate, or otherwise transfer its rights or obligations under this Agreement . . . without the prior written consent of the other Party, except as follows with notice: (a) Without securing the consent of the other Party, either Party may assign its rights, or delegate its duties, or both, in whole or in part, (i) to any present or future Affiliate of the assigning Party; or (ii) to any third party that

assumes the operation of or otherwise acquires any substantial portion of the business of the assigning Party affected by this Agreement or an Order provided that in either (a)(i) or (ii) above, the assignee assumes all obligations under this Agreement or Order and if the assignment is for Materials the assigning entity no longer uses the assigned Material; and (b) Supplier may subcontract its performance…Any assignment, delegation or transfer for which consent is required hereby and which is made without such consent given in writing will be void.

The provisions of the foregoing paragraph are not applicable in the event a third party (“Outsourcer”) which by purchase, lease, outsourcing or otherwise, assumes the operation, administration and/or management of any substantial portion of the business of AT&T . . . for the purpose of providing data processing services to AT&T or an Affiliate (“Outsourcing”). In the event of an Outsourcing, [BMC's] consent shall be required and shall not be unreasonably withheld provided that the Outsourcer agrees in writing with [BMC] to be bound by the terms of this Agreement and the applicable Order and only uses the Material for the sole benefit of AT&T or the applicable Affiliate by providing data processing services to such entity.

DX10 at 10-11.

32. Under the MPA, “services” meant “any and all labor or service provided in connection with this Agreement and an applicable Order, including but not limited to, consultation, engineering, installation, removal, maintenance, training, technical support, repair, programming, ” among others. Id. at 8.

4.The Standard BMC EULA

33. BMC's standard EULA was considerably shorter than the company's agreement with AT&T. Compare DX10 (sixty-six pages), with DX7 (twelve pages). It did not contain the same assignment and delegation rights as the BMC-AT&T licensing agreement and, in fact, (subject to exceptions not relevant here) stated that the customer “may not assign or transfer a Product separate from the applicable Agreement and License, and may not assign or transfer an Agreement or a License.” DX7 at 4.

D. IBM's History of Displacement and the Displacement Concerns Between the Parties

34. Because BMC claimed that IBM fraudulently induced it into signing the 2015 OA, the parties devoted substantial attention to their prior course of dealings and each other's subjective understandings of the contract language, specifically section 5.4's non-displacement language. “Non-displacement has always been a contentious issue between BMC and IBM since 2008, ” Craig Dep. 49:15-49:16, in no small part because IBM had a general desire to replace BMC's software with its own at outsourced accounts. See Stanton Dep. 57:01-57:08. Thus, in addition to reviewing the at-issue contracts, the court reviews the parties' interpretative evidence, including IBM's prior displacement conduct, the negotiations leading up to the 2013 OA, IBM's displacement at AT&T through Project Swallowtail, the 2015 OA negotiations, and how IBM talked about displacement, both internally and externally.

1.IBM Displaces BMC at Bank of Ireland

35. Before the instant dispute, BMC and IBM ran into issues over contractual non-displacement restrictions over a decade ago involving a different mutual customer: Bank of Ireland (“BoI”). See Trial Tr. 206:02-06 Mar. 16, 2022 (Jones); Trial Tr. 14:21-15:08 Mar. 21, 2022 (Sweetman). As IBM's John Sweetman tells it, BoI wanted to transition away from using BMC software on its mainframe systems and toward IBM software. See Trial Tr. 15:01-17 Mar. 21, 2022 (Sweetman). At the time, IBM worked as BoI's IT outsourcer; the MLA and a 2008-era Outsourcing Attachment containing the same non-displacement language present in the 2015 OA governed IBM's relationship with BMC. Id. 15:16-17 (Sweetman). See generally PX2. Regarding that language, IBM's John Stafford, in an internal e-mail, explained that the language was “very clear” in limiting “what [IBM] [is] permitted or not permitted to do in situations where we take over a client's footprint.” PX460.

36. In late 2011 or early 2012, BMC caught wind that IBM, in its capacity as BoI's IT outsourcer, might be displacing BMC's products with IBM's own. See DX152 at 3. BMC executives e-mailed their counterparts at IBM raising concerns that IBM's conduct contravened the 2008 OA. Id. More than a year later, on March 19, 2013, BMC's Chris Alexander, who served as the IBM Strategic Account Manager, e-mailed BMC executives about IBM's position on the BoI displacement project. DX149 at 1. In his summation, “IBM asked BoI to uninstall all BMC [software] before IBM came on . . . and took over, ” a task that BoI only partially completed. Id. Nevertheless, IBM “then argued that BoI solely and independently removed BMC products.” Id. IBM characterizes this e-mail as evidence of its “consistent[]” advisement of BMC that it believed the non-displacement provision authorized customer-directed displacements, i.e., where the mutual customer requests IBM to displace BMC's products with its own products.¹² See Dkt. 612-8 at 13-14; Trial Tr. 19:23-20:01, Mar. 21, 2022 (Sweetman) (Q: “Based on [these e-mails], in IBM's view, who was causing the displacement at Bank of Ireland?” A: “The Bank of Ireland.”). In any case, IBM conducted the BoI displacement without buying a license. Trial Tr. 36:21-37:10, 30:25-31:14, Mar. 21, 2022 (Sweetman).

12 The court notes that the e-mail does not squarely support IBM's reading in two important respects. First, BMC's understanding that IBM “asked BoI to uninstall” its software-which IBM's quoted language neglected to acknowledge-suggests that IBM might have requested the displacement, undermining IBM's contention that BoI directed the displacement. Second, and more importantly, the e-mail can also be read as evidence that IBM believed that the non-displacement provision limited what it could do, but not what mutual, third-party customers may do, which is consistent with both parties' contemporaneous understanding of the non-displacement language. DX149 at 1-3. In that sense, if IBM believed that it could serve as BoI's agent in conducting the displacement-the position it has adopted in this litigation with respect to BMC's fraudulent inducement claim-it would not have asked BoI to remove BMC's products, itself.

2.The 2013 OA Negotiations

37. As Sweetman testified, BMC raised some of its concerns about IBM's displacement activities at BoI in the lead up to, and during, the negotiations surrounding the 2013 OA. Trial Tr. 20:05-07, Mar. 21, 2022 (Sweetman). Two weeks after Alexander's e-mail, on March 31, 2013, the parties executed the 2013 OA. PX3. Like the 2008 OA, the 2013 OA provided IBM several ways to use BMC's products when serving as an IT outsourcer. For the option to “access and use” a mutual customer's BMC products for “no fee, ” the 2013 OA included a nearly identical non-displacement provision that prohibited IBM from displacing BMC's products. Compare PX3 at 2 (“Customer agrees that while Customer cannot displace any BMC Customer Licenses with Customer products, Customer may discontinue use of BMC Customer Licenses for other valid business reasons . . . .”), with PX2 at 2 (“IBM agrees that while IBM cannot displace any Products with IBM products, IBM may discontinue use of Products for other valid business reasons.”). The 2013 OA also included the same alternative path whereby IBM could purchase its own licenses- which did not contain any displacement restrictions-for a discounted price instead of operating under the BMC customer's licenses for free. PX3 at 1-3; PX48 at IBM00078246 (IBM internal guidance on 2013 OA explaining that “[w]here displacement is desired by the Client, ” the “[c]ostly [a]pproach” is to “[n]egotiate with BMC to acquire equivalent BMC licenses to be held by IBM that do not include the restrictive language” contained in “client-licensed BMC products”); PX94 at IBM00081140 (IBM document explaining that if IBM wanted to displace a joint-customer's license then it needed to purchase its own license).

38. During the negotiations, IBM unsuccessfully sought to “remov[e] . . . all migration restrictions, ” i.e., the OA's non-displacement provision. PX22 at 5; PX28 at 6 (noting that BMC has “aggressively sought to enforce [the non-displacement] provisions” and that “executive direction for renegotiation is to focus on removing the [non-displacement] restrictions”); PX30 at 1 (IBM employee internally stating, prior to the 2013 OA's execution, that she “didn't think [IBM was] agreeing to another non-displacement clause - in fact I thought we agreed to negotiate out of that”); PX34 at 1-3 (Sweetman e-mail noting that an “access and consent agreement . . . allows IBM to access the software without a fee, but also bars IBM from displacing the BMC products with IBM products for the duration of our services agreement” and that this displacement restriction “is being discussed in our worldwide negotiations with BMC but there is no change at present to current arrangements.”).

39. While negotiating the 2013 OA, IBM even proposed adding contractual language specifying that IBM was “not restrict[ed] . . . from executing on IBM customer requests with regard to any decision to discontinue or displace as determined by the IBM customer.” PX475 at IBM00061376. BMC, however, refused and “despite very senior level discussions amongst [the] companies, ” there was “no change on the non-displace.” PX37 at 3; PX33 (IBM internal guidance noting that “BMC is not willing to remove current non-displacement language”); see also DX167 (e-mail from BMC's Brian Jones stating “I have also updated the non-displacement language to make sure it is the same as the 2008 agreement as agreed” by BMC's and IBM's then-negotiators); DX164 at 2 (in response to IBM's request to “remove or amend . . . the current non-displacement language, ” BMC responded that its “position is unchanged as it related to non-displacement”); Craig Dep. 168:22-169:13 (IBM wanted to get rid of the non-displacement provision, but BMC would not agree to it without “additional consideration” which “would have cost [IBM] a hell of a lot more money.”).

40. Within four days of executing the 2013 OA, IBM began strategizing internally about how it “may cho[o]se to manage the non-displacement language overall as a business” despite there being “no change on the non-displace.” PX37 at IBM00062682, 684; see also PX38 (internal IBM e-mail wherein IBM account executives discussed “put[ting] BMC to the test on the additional non-displace restrictions); PX52 at IBM00063460 (“As migration projects get on, the ‘other valid business reason' text from non displacement clause will need to be leveraged in discussion [with] BMC.”); PX462 (e-mail from Bruno Hibert, the Vice President of Facilities and Software Infrastructures in IBM's Strategic Outsourcing division, noting that “this reads as a great ‘test case' for ‘other valid business reasons.'”). Two months after the 2013 OA was signed, IBM began carrying out its strategy to “manage” the non-displacement language, informing BMC that it believed the non-displacement provision did not “clearly align” with “the intent and the principles of U.S. competition law.” See PX465 at IBM00077478-79. BMC immediately rebuffed this claim, and BMC's Brian Jones, an outsourcing executive, reminded IBM of the other options it had under the OA to displace products. Id. at IBM00077476-78. IBM lawyers also drafted what IBM employees referred to as the “BMC White Paper” regarding the “[p]lanning, communication, and strategy relating to guiding IBM employees in the appropriate management of specific vendor/competitor issues.” See PX80 at IBM00078423.

3. IBM Displaces at National Australia Bank

41. No later than August of 2013, and after the parties executed the 2013 OA, BMC learned that IBM was involved in another displacement project involving a mutual (but longtime BMC) customer, National Australia Bank (“NAB”). DX192 at 4-6; DX193 at 1 (noting that NAB informed BMC that IBM was “actively displacing the majority” of BMC's software). On August 1, 2013, Andrew Wiltshire, BMC's Sales Director in its GOSI division, e-mailed IBM personnel asserting that IBM's upcoming displacement of “up to 19 of the BMC tools at the NAB” directly contravened the outsourcing attachment both parties had signed. Id. Though Wiltshire noted his preference “to work to a commercial solution over a contractual one, ” IBM evidently did not respond within a week, and BMC's Jones escalated the matter to his counterparts at IBM. Id.

42. As with BoI, IBM did not buy product licenses to perform the NAB displacement project. Trial Tr. 36:21-37:10, 30:25-31:14, Mar. 21, 2022 (Sweetman); Trial Tr. 254:02-04, Mar. 14, 2022 (Ah Chu). Instead, it appears that the parties were able to come to alternative business resolution over the dispute, as evidenced by the 2015 OA's release of claims against IBM. Trial Tr. 254:25-255:01, Mar. 14, 2022 (testifying that the NAB dispute “ended up being part of the release . . . negotiated as part of the 2015” OA) (Ah Chu).

4. Project Swallowtail

43. IBM's displacement opportunities did not end with NAB. In April 2013, immediately after signing the 2013 OA-and unbeknownst to BMC-IBM began communicating with AT&T regarding a secret AT&T project, codenamed “Project Swallowtail.” Cattanach Dep. 56:18-57:04; PX39 at IBM00000794. The leadup to Project Swallowtail functionally advanced in two stages. In the first, in 2014, AT&T solicited IBM's involvement in a displacement project but ultimately pulled out of negotiations. In the second, in 2015, AT&T expressed renewed interest in integrating IBM's software onto its mainframe computers. IBM's eventual execution of a contract with AT&T to displace BMC's products with its own and its subsequent signing of the 2015 OA with BMC expressly prohibiting it from displacing BMC's products with its own gave way to the claims raised in this case.

a. Project Swallowtail: Stage 1

44. The purpose of Project Swallowtail was “for IBM to migrate the BMC software products currently installed in the AT&T environment to [alternative] software products, ” primarily IBM's competing software products. PX18 at IBM00000139; PX41 at IBM00061937; PX39 at IBM00000794; PX68 at IBM00079049 (“I believe the intent at AT&T is to remove 100% of the existing BMC [software] estate and replace with IBM tools. This is being done at the direction of AT&T with the intent that we standardize on the platform AT&T has chosen which includes the IBM products in scope.”); see also Shell Dep. 19:14-20:23 (“Q. The general scope of the project was to replace BMC software products with other products for AT&T, right? A. Yes. Q. And some of the products that would be replacing the BMC products were IBM products, right? A. Yes.”); Brickhaus Dep. 95:05-95:15 (agreeing that Project Swallowtail was a displacement of BMC software products with “IBM products and . . . some CA [Technologies] products”). Indeed, Project Swallowtail explicitly called for IBM to displace AT&T's BMC products. See PX57 at IBM00063804 (“scope” of Swallowtail included “migration of all BMC Mainframe Software to other products across the 3 AT&T [mainframes]”); PX56 at IBM00063957 (IBM internal Swallowtail description: “Software takeout of BMC across the [AT&T] enterprise”).

45. Due to the Project's “sensitive nature, ” IBM signed a non-disclosure agreement with AT&T and instructed its employees to refrain from acknowledging the Project or its purpose outside of the Project Swallowtail Team. See PX41 at IBM00061934 (IBM giving AT&T confirmation on April 15, 2013, it “shall comply with the [confidentiality obligations] . . . with respect to [IBM's] participation in AT&T's Swallowtail Project”); PX55 at IBM00066638 (“Please remember we should not discuss the Swallowtail project with anyone outside of those involved with preliminary findings . . . .”).

46. Initially, IBM performed “an assessment . . . and analy[sis] [of] [AT&T's mainframe] software products in order to determine the operational feasibility and financial cost to remove or replace such products.” PX41. Subsequently, on September 4, 2013, at AT&T's request, IBM e-mailed AT&T its “proposal for the AT&T Swallowtail project.” PX67. That plan promoted IBM's mainframe products and explained why AT&T should choose IBM's products over BMC's. Id. at IBM00063703 (“Why IBM for BMC Replacement Software Group Value.”).

47. Though IBM thought that participating in Project Swallowtail could further its long-term growth and expansion goals, there is no indication that it initiated the partnership with AT&T. See PX72 at IBM00076969; PX68 at IBM00079048-09; PX476 at IBM00076329; PX82 at IBM00068683 (“Clearly last year's big play was our contract extension(s) coupled with the BMC take-out opportunity we are now targeting 1Q'14.” (emphasis added)); PX38 at IBM00063091; PX248 at IBM00090399 (“The software part of [Project Cirrus], is around $60M+ in additional revenue.”); PX72 at IBM00076969 (“[I]t is clearly the right move to remove a competitor from this environment, ” and its “best scenario is that our plans include displacement of 100% of the current BMC footprint.”); see also PX67 IBM00063692 (proposing to AT&T that “all 20 BMC Products . . . be migrated utilizing a common migration approach”); PX476 at IBM00076329 (listing benefits of “putting IBM first” and using IBM software in migration projects); Shell Dep. 75:05-22 (IBM wanted the work associated with Project Swallowtail because it needed the revenue); id. at 120:14-122:10 (stating that he “wanted to get the business” from AT&T rather than having AT&T renew with BMC). IBM also described Swallowtail as a “project[] to drive growth” at IBM by “displac[ing] BMC in [AT&T's] Mainframe environment.” PX82 at IBM00068683-84; see also PX38 at IBM00063091 (internal IBM e-mail discussing efforts to try to get a mutual IBM-BMC customer “to use IBM [software] in place of the BMC products which they are currently using” because it is “an additional source of revenue for [IBM]”).¹³

13 “Project Cirrus” was the code name for an AT&T initiative to restructure IBM's mainframe IT services contract with AT&T. PX97 at ATT00000967-68. AT&T's working strategy for Cirrus was to transition IBM's mainframe IT services contract to a “consumption model that includes all 3rd party software, new labor and hardware options.” PX97 at ATT00000971.

48. Despite the excitement from IBM's AT&T team about this “BMC Takeout Opportunity, ” certain IBM executives and its BMC-facing team were “concerned” because of the “very specific verbiage around software replace” in the 2013 OA. PX72 at IBM00076967-69 (IBM executives agreed Swallowtail “would be an Excellent BMC Replace Opty-but, are also concerned”); id. at IBM00076968 (“example of Excellent Oppty - that is riddled with concern over Legal inhibitors/etc ..... ”). Still, IBM continued to negotiate with AT&T to further a deal on Project Swallowtail. See Shell Dep. 238:10-239:1 (IBM offered various incentives to make the price of Project Swallowtail more attractive to AT&T to encourage AT&T to close on the deal by the end of the year); PX064 at 3 (“Project Swallowtail - Additional Consideration for 3Q'13 Contract Signature”). Notwithstanding IBM's apparent enthusiasm, in March 2014, AT&T decided not to proceed with Project Swallowtail, opting instead to renew its agreement with BMC. See DX-245; PX85; PX84 at IBM00068614-616; PX12 at 1-19.

49. Upon hearing that AT&T would not proceed with Project Swallowtail, IBM voiced its frustrations about losing the project and immediately started strategizing ways to “flip [AT&T's] decision back to IBM.” PX84 at IBM00068613; see Id. (“[W]e need to fight to try and get back in this or make the decision that it is not financially viable.”); see also PX104 at IBM00110239-40 (“About a year ago, there was a[n] Engagement Project called “Swallowtail” that was specifically intended to displace BMC software on the [AT&T] account (i.e. replace with IBM software solutions). The client ended up signing a new deal with BMC and we lost the new business opportunity.”). IBM's desire to perform Project Swallowtail was so great that it offered to perform Project Swallowtail for free. PX84 at IBM0068612 (noting that IBM offered its migration services to AT&T “at no charge (free)”).

a. Project Swallowtail: Stage 2

50. Around February 2015, AT&T changed its mind and decided to restructure its mainframe environment after all, at which point it approached IBM about a new initiative code named Project Cirrus. PX97 at ATT00000967-968. AT&T's working strategy for Cirrus was to transition IBM's mainframe IT services contract to a “consumption model that includes all 3rd party software, new labor and hardware options, ” reaping “drastic[]” savings on “mainframe total costs” in the process. PX97 at ATT00000969. And it wanted IBM to lead the project because “IBM ha[d] skills and expertise” necessary “to successfully execute the project.” See Conway Dep. 11:11-11:17. AT&T specifically wanted to replace BMC's products with other vendors' products primarily out of cost considerations. See Id. 18:03-19:25 (“What triggered the development of the Swallowtail plan to transition from the BMC products was negotiations . . . in 2013” wherein AT&T “could get the financial terms . . . that they wanted in the contract, so finance asked us to look at alternatives to BMC.”).

51. Shortly thereafter, on March 10, 2015, Joe Dzaluk, a Vice President in IBM's Global Technology Services (GTS) unit, contacted various strategic outsourcing account teams for the purpose of “assessing [IBM's] ongoing relationship with BMC.” PX104 at IBM00110241. Dzaluk requested that IBM's account teams help “explor[e] instances in which clients have expressed an interest in alternative technologies that do not include BMC products . . . includ[ing] replacing BMC software with other products or services performing similar functionality.” Id. Dzaluk further stated that “[IBM] need[ed] [IBM's account teams] to begin working with your client to form an assessment of the client's interest in considering alternatives and the size, scope, timeline, and impacts a project like this might have on your day-to-day operations.” Id. Guy Skinner, IBM's Director and Sr. Delivery Project Executive for the AT&T account, responded to Dzaluk's request by informing him that “[a]bout a year ago, there was a[n] Engagement Project called ‘Swallowtail' that was specifically intended to displace BMC software on the account (i.e., replace with IBM software solutions).” PX104 at IBM00110239-40. Skinner then told Dzaluk that “[t]he client ended up signing a new deal with BMC and [IBM] lost the new business opportunity.”¹⁴ Id.

14 Skinner's e-mail also noted that “the BMC software is retained by the AT&T client and they have a software license agreement directly with BMC.” PX104 at IBM00110240.

52. On March 26, 2015, Jason Cattanach, an Associate Project Executive for IBM's GTS division, sent an e-mail with a PowerPoint presentation titled “GTS Savings for AT&T Challenge.” PX113. That slide show listed “transfer architectural control and software to IBM, ” resulting in “savings and simplification via Project Cirrus methodology” under the heading “Mainframe.” Id. at IBM00104420; see also Stafford Dep. 206:24-209:12 (confirming replacing BMC software with other products, including IBM products, at IBM strategic outsourcing accounts was part of IBM's broader effort to identify possible displacement projects). Shortly after Skinner informed Dzaluk that IBM lost Project Swallowtail when AT&T signed a new deal with BMC, IBM resumed discussing Project Swallowtail with AT&T executives as part of the larger Project Cirrus. PX114 (e-mail from AT&T noting that it “plan[ned] to displace all of the. . . BMC products before their next renewal date”); see Conway Dep. 25:24-26:03 (describing Project Cirrus as a “transition plan” included in “the agreement that we signed with IBM in 2015”).

53. To be sure, like before, IBM wanted to participate in Project Swallowtail's second iteration. Project Swallowtail furthered IBM's strategic plan to “[a]ggressively [p]ursue migrations to competing [IBM] solutions” as part of IBM's strategic outsourcing division's “Top Spend Suppliers Initiative.” PX89 at IBM00051438-442; see Id. (noting that “IBM offers alternatives for most BMC products”); see also PX91 (Craig stating “I have a concern from these charts in that there is an entry ‘IBM offers alternatives for most BMC products.' While this is true there should have been a caveat that where IBM does not own the BMC license then we cannot displace their products.”). IBM had a “dedicated team . . . co-sponsored by GTS” that aggressively pursued these opportunities at outsourced clients with the goal of displacing third-party software products licensed from other software providers with IBM products. PX219 at IBM00088635-636 (“IBM teams are committed to ensure that the best ISV SW to IBM SW migration case pricing is available to your deal.”).

54. Further, IBM's senior management for its strategic outsourcing division implemented a strategy to “pursue exclusion of all BMC [software] from [IBM's] engagement solutions.” PX111 at IBM00050586. And, by participating in Project Swallowtail, IBM thought it would financially benefit.¹⁵ See Shell Dep. 284:03-285:08; PX216 at IBM00007356-57 (internal IBM resource showing that IBM valued the outsourcing contract between it and AT&T, including Project Swallowtail, at approximately $800 million between July 2015 and December 2020, with IBM expected to earn profits nearing $300 million.); PX17 at IBM00000326 (milestone payment); Conway Dep. 40:20-41:03 (discussing AT&T's payments to IBM); PX125 at IBM00105494 (IBM internal resource illustrating AT&T revenue projections); Shell Dep. 284:03- 285:08 (the plan to execute Project Swallowtail was back on as part of Project Cirrus, which was a $860 million contract); PX248 at IBM00090399 (“$60M+ in additional revenue” for software).

15 The evidence shows that, as the “project manager” for Project Swallowtail, IBM functioned as “the overall coordinator for conversion activities.” Brickhaus Dep. 107:06-18; 119:20-23. BMC claims that this, coupled with IBM's desire to participate in Project Swallowtail, “makes clear that IBM directly influenced . . . the decision to implement Project Swallowtail.” See Dkt. 723 at 24. Contrary to this interpretation, the court reads this evidence to, at most, show that IBM was directly involved in Project Swallowtail's implementation.

55. Ultimately, on June 26, 2015, AT&T and IBM formed a partnership over the BMC displacement, and incorporated Project Swallowtail into the larger Project Cirrus framework. See PX16; PX216 at IBM00007356; Cattanach Dep. 117:13-18:04 (testifying that AT&T started paying IBM more after execution of Project Swallowtail). The executed agreement for Project Swallowtail explicitly stated that “[t]he purpose of this project is for IBM to migrate the BMC software products currently installed in the AT&T environment to the software products listed in section 4.” PX18 at IBM00000139.¹⁶

16 Section 4 of the Project Cirrus (or Swallowtail) agreement is comprised mostly of IBM products and sub-products that were expressly identified as replacements for the BMC products. PX18 at IBM00000142-45; see also Trial Tr. 70:12-71:02, Mar. 16, 2022 (Román).

5.The 2015 OA Negotiations

56. While IBM pursued negotiations with AT&T regarding Project Swallowtail in 2015, at that same time, IBM pursued negotiations with BMC to renegotiate the 2013 OA. IBM executed its Project Swallowtail agreement with AT&T in June 2015. But, by that point, IBM and BMC had failed to reach an agreement on the renegotiated OA. So, before IBM executed what would become the 2015 OA with BMC, it knew that it was going to displace BMC's products with its own at AT&T. This, in turn, influenced IBM's negotiation goals with BMC.

a. IBM's Goals in the 2015 OA Negotiations: Remove or Restrict the Scope of the Non-Displacement Clause

57. Rick Clyne, a professional negotiator, led negotiations for IBM. According to Clyne, the re-negotiation of the 2013 OA turned, in part, on the non-displacement language.¹⁷ See PX98 at IBM00124943 (IBM internally noting that buying IBM licenses to displace was an “Option[] on the Table” but one negative was the “High Cost to Duplicate Clients environment as IBM owned”); PX121 at IBM00038707; Stafford Dep. 224:16-25; see also PX196 at IBM0012448 (IBM presentation: because AT&T remained on the list of accounts for which displacement was prohibited, IBM risked incurring fees to perform displacement). As evidenced by internal e-mails with IBM executives, including Dzaluk, Clyne was aware of the non-compliance complaints BMC had raised. In one e-mail, Dzaluk wrote that “[o]ver the last 5 years, we have had numerous global claims from BMC, Microsoft and others claiming hundreds of millions and have always settled for a small percent of the claim.” PX109 at IBM00038867. In response, Clyne wrote:

17 Other issues were also negotiated, including IBM's shared hosting rights, which took on greater urgency than the displacement concerns. Clyne Dep. 230:05-230:15; Calo Dep. 10:22- 11:3, 55:9-14; PX106; PX108. With respect to those rights, Clyne testified that IBM “[f]ar, far exceeded” its goals. Clyne Dep. 139:04-139:08.

Trust that I would expect settlement payments for alleged, or even real, past-non-compliance issues to always be for pennies on the dollar. (A $12M+ non-compliance claim by BMC at AMEX last year was settled for $0.) It's the future frameworks for shared environments as well as non-displacement terms, such that we finally get out of the business of writing checks to these guys, that makes this a really steep climb. But we'll see where we land. (For clarity: this week is to address shared environments only.)

PX109 at IBM00038867.

58. Clyne's e-mail reflected IBM's goal to preempt non-compliance issues resulting in settlements at the outset by removing displacement restrictions entirely. See Id. IBM's own internal guidance documents-which were labeled as “IBM Confidential/Prepared for IBM Attorney”-reflected this negotiation need following IBM's execution of the Project Swallowtail contract with AT&T. Those documents noted that the 2015 OA “need[ed] to ensure that IBM may displace a BMC product with an IBM product, if client directed.” PX170 at IBM00124599. See PX196 at IBM0012448 (“Proposal will still limit IBM's ability to offer Displacement at up to fifty-four named customers. . . . Potential financial impact of paying $140m fee for planned displacement [at] AT&T not in pricing case [for Project Swallowtail].”).

59. Though the non-displacement language from the 2013 OA covered approximately 700 mutual customers, Clyne understood that BMC was “agreeable to . . . reducing the . . . list [to] 50” because it was “just fed up with this crap about, you know, IBM's interpretation [of] the language and whatnot” and did not “want to have to spend any more attorney time” on displacement issues. See Clyne Dep. 55:13-55:19, 148:02-148:23 (explaining that BMC offered to provide a list of accounts for whom “IBM cannot displace BMC software . . . under any circumstances, period.”). Thus, Clyne understood that, with respect to a handful of customers, BMC insisted that “under no uncertain circumstances c[ould] IBM ever replace BMC software at any of th[ose] accounts for any reason.”¹⁸ Id. 55:21-55:23, 249:19-250:12 (explaining that BMC “wanted [the contract] to be as clear as it possibly could be that . . . IBM was prohibited from displacing BMC at any of the . . . accounts, regardless even if the customer asked for it”). To that end, BMC proposed language that would contractually neuter IBM's “interpretation” of the existing language that IBM argued enabled customer-directed displacements. See id.

18 BMC initially sought stronger non-displacement language than that included in section 5.4 of the 2015 OA but settled for the existing language of the 2013 OA. See Clyne Dep. 203:01- 203:13.

60. For its part, IBM initially sought “the complete elimination of the [non-displacement] clause.” See Id. 36:17-36:23, 54:01-54:03, 54:23-55:02; PX156 at IBM00097161 (“IBM wants the non-displacement language in the current OA fully removed from our agreement with BMC.”); PX163 at IBM00032987-92 (negotiating whether “non-displacement [will be] completely removed”).¹⁹ When BMC rejected that, IBM suggested alternative language that would permit it to displace BMC's products with IBM's products at the customer's direction. Clyne Dep. 287:07-287:14. Even though IBM “beat that drum constantly” during the negotiations, BMC was unwavering in its rejection of the customer-directed allowance. See Id. 287:14-287:21, 320:09-320:16 (explaining that IBM sought language that would enable it “to displace BMC with IBM products” if a “customer asks” but that “BMC would not agree [to that proposal] and it never made it into the final agreement”).

19 IBM pursued the complete elimination of the non-displacement clause in the days leading up to the execution of Project Swallowtail. See PX162 at BMC-000008614-8616; PX163 at IBM00032995 (e-mail from Clyne stating, “Just a heads-up regarding the criticalness of obtaining agreement to have all reference to non-displacement removed and the importance of getting the confirmation to me tonight . . . however late it might be.”); PX163 at IBM00032993 (e-mail from Clyne, three days prior to Swallowtail's execution: “[I]f you can deliver where we landed on Friday, . . . I am now 90% certain it will be accepted by IBM and we'll be done. It's EXTREMELY urgent, however, that I receive e-mail confirmation asap . . . .”) (emphasis in original).

61. Citing the parties' troublesome “shared history, ” BMC rejected IBM's proposal. PX163 at IBM00032991-92. Given the impasse between them with respect to changes to the contract language, BMC and IBM compromised by retaining the existing non-displacement language and narrowing down the number of accounts to which it applied. Clyne Dep. 287:25- 288:15, 291:04-291:10 (explaining that part of the compromise was “that the parties agreed to keep the original language”); see also PX163 at IBM00032988; Trial Tr. 208:21-209:5, Mar. 14, 2022 (Ah Chu).

62. As BMC confirmed in its negotiations with Clyne, limiting the non-displacement language would still require IBM to “pay BMC the cost of the license fees and one year's support at a price discounted . . . per the OA” when IBM sought to “replace BMC software at IBM outsourcing accounts.” PX132. But, if that were the case, IBM would have to purchase the licenses directly from BMC to fulfill its Project Swallowtail obligations and comply with the BMC contract. At the direction of IBM's executives, Clyne proposed the removal of AT&T and three other accounts from the narrow list of mutual customers covered by the non-displacement language. BMC wanted to know why:

Q. [W]hen you asked for the [accounts] to be removed, what as BMC's response?

A. Why.

Q. And what did you tell them?

A. As I had foretold IBM, that would be their very first question.

Q. And what did you tell them?

A. I diverted. I said the business wants those [accounts] out, a nonresponsive response. You know, I wasn't going to lie and so I - that was the response I gave.

Id. 61:17-16:20. Clyne may not have lied, but he did obfuscate. Clyne's strategy in including the other three accounts was to conceal the importance of AT&T to IBM's overall business goals:

[I]f I was to ask for just AT&T then I would highlight it even more than I wished to as negotiator for something that my client wanted removed. So, I informed the executives that I was going to add three other accounts just to have AT&T in the mix so that the focus wouldn't be just on AT&T. And I, as a negotiating tactic, added [the three other accounts.]

See Id. 73:06-73:13.; see also PX200; PX205. But as Clyne acknowledged, his correspondence with BMC omitted “the fact that IBM was intending to displace BMC's products at AT&T . . . even though [he] knew that at the time.”²⁰ Clyne Dep. 196:21-197:03, 101:01-101:05. Eventually-in August 2015, approximately two months after IBM executed the Project Swallowtail contract-IBM's displacement concerns could be distilled to one account: AT&T. See Id. 95:21-95:23, 289:09-289:14 (explaining that IBM told him not to worry about the other accounts because they “just need[ed] AT&T”); PX203 (removal of AT&T from list of protected accounts a “[m]ust have”); PX205 (removal of AT&T “an absolute must”); PX174 (“Finalize displacement verbiage and agreement on 50 clients (ATT to not be on the list).”); PX183 at IBM00124379 (“In all cases, IBM must ensure that AT&T is not included on BMC's list of clients subject to non-displacement clause, when negotiating the current list of fifty-four down to fifty.”). Importantly, IBM's efforts to remove AT&T from the list of fifty-four protected accounts significantly increased after June 26, 2016, when IBM and AT&T executed Project Swallowtail. PX18 at IBM00000139. Compare PX155 at IBM00124233 (June 16, 2015 IBM presentation stating non-displacement provision will be reduced to 50 accounts), with PX169 at IBM00124317 (June 26, 2015, updated draft of the same presentation emphasizing need to “ensure that AT&T is off BMC's list when bringing down the fifty-four currently on the list down to fifty”).

20 Clyne was in contact with IBM executives regarding the negotiations “[a]lmost everyday, sometimes on Saturdays and Sundays, too.” Clyne Dep. 198:24-199:03.

63. IBM's efforts to remove AT&T from the list of customers on Exhibit K subject to the non-displacement language failed because BMC was unwilling to contractually jeopardize its business relationship with AT&T. Trial Tr. 195:16-24, Mar. 14, 2022 (Ah Chu) (“AT&T was just too strategic of a customer for BMC; and there were several attempts, I recall, from Rick Clyne to remove AT&T. And each and every time he tried to remove them, it was flat out denied by BMC.”); PX213 (“The net is that BMC will not agree to remove AT[&]T . . . from the protected list, under any circumstances.”); PX217 (IBM “would not relent on [its] insistence to discuss [BMC's] refusal to delete AT[&]T and Kaiser from the protected list”); DX-390 at IBM00028229 (BMC informing IBM that AT&T “is an unrealistic account to relinquish”); Bergdoll 52:16-54:20 (Exhibit K was “nonnegotiable” and “if there wasn't going to be a list” of protected accounts, BMC “wasn't interested” in executing the 2015 OA); Trial Tr. 195:16-19, Mar. 14, 2022 (“Q. To your knowledge, would . . . BMC have signed the 2015 OA if AT&T were not on Exhibit K, the list of customers covered by the non-displacement clause? A. Absolutely not.”) (Ah Chu).

64. In any event, Clyne acknowledged in his deposition testimony that IBM did not plan on conforming its conduct to the contractual language:

Q. In your negotiations with BMC, you never told BMC that IBM was not going to abide by the non-displacement provision in the 2015 OA?

A. No, never told them that. Never said that.

Q. But in fact, IBM's plan was not to abide by it, wasn't it?

A. That was the position that legal had taken.²¹

Clyne Dep. 111:10-111:19; see Id. 113:11-113:21 (explaining that “IBM's conduct throughout the negotiations and the time spent on non-displacement claims by BMC” led Clyne to know that IBM was not going to comply with the 2015 agreement). In that sense, Clyne did not believe that IBM negotiated the 2015 OA contract with BMC in good faith. Id. 117:19-118:12.

21 Clyne also explained that “IBM had gone on the record with Mr. Hagen's letter that they . . . didn't agree with the language that they had executed a couple months before in the 2013 agreement and went through with two more years of more claims [and] . . . more compliance issues” before agreeing to the same language in the 2015 OA. Clyne Dep. 146:02-146:13. Though IBM's legal department communicated that the non-displacement clause was unenforceable and others in IBM claimed it was vague, Clyne acknowledged that IBM “very much” wanted the clause removed from the 2015 OA. Id. 151:15-151:20. Likewise, Clyne found IBM's position on the non-displacement language contradictory. See Id. 231:21-232:03 (“However, there were contradictions to IBM's position [regarding the non-displacement language] that IBM made in the course of the negotiations regarding that position.”). “In my experience as a business person and professional that if the agreement has been executed, then it's totally inappropriate to go back a month later, two months later and basically say, you know, the agreement we just signed all - you know, all bets are off. We're not going to honor it.” Clyne Dep. 296:12-296:20; see also Id. 103:10-19 (testifying the IBM “knew what [it was] signing”). Moreover, in Craig's deposition, he testified that he was unaware of the interpretation promulgated to BMC by IBM Legal was ever published in IBM's own internal guidance documents. Craig Dep. 202:04-202:16 (“I don't believe it was ever issued as a general guidance document.”).

65. Asked about the conflicting contractual obligations IBM might encounter if it signed one contract agreeing to displace BMC's products and another prohibiting it from displacing BMC's products, Clyne acknowledged that his “job [was] to negotiate and obtain what [his] client wants to achieve.” See Id. 103:10-103:11. Compliance issues came “after the fact, after the agreement [was] signed.” Id. 103:12-103:13. Because he had “no control over” how IBM would perform its conflicting contractual obligations under Project Swallowtail and what became the 2015 OA, Clyne did not occupy himself with the issue: “throughout the seven months there were things that I discussed, advised against doing. Some were listened to. Some were not. But at the end of the day, they know what they're signing and it's up to them to comply.” See Id. 103:10-103:19. In Clyne's view, those tasked with complying with the contract included IBM Vice President Peter Lynt, Yvonne Calo, and John Stafford, among others. Id. 103:20-104:01, 204:20-204:24.

66. Internally, IBM was considering proposing a modification to the non-displacement language that would allow displacement for a flat percentage fee. Id. 62:04-62:08. That idea originated with Stafford, but Clyne cautioned that “the penalty per the contract per BMC is that if . . . IBM is found to have actually breached the contract . . . the penalty was that IBM had to pay for a license and one year's maintenance . . . I said, John, you know . . . that's awfully steep.” Id. 62:18-62:24.

b. IBM's Execution of the 2015 OA

67. Ultimately, the parties significantly reduced the provision's scope from hundreds of mutual customers to a defined list of fifty-four BMC customers, which are identified in Exhibit K of the 2015 OA and include AT&T. PX4 § 5.4; see Clyne Dep. 43:21-45:05 (“Q. In the 2015 OA [the non-displacement provision] was limited to a list of fifty-four? A. Correct. Q. Was that limitation in scope valuable to IBM? A. Very much so.”). When the contract was executed, only eighteen of those customers were also IBM outsourcing customers. See PX228 at IBM00124048; Trial Tr. 208:21-209:05, Mar. 14, 2022 (Ah Chu).

6.How IBM Talked About Displacement

68. How IBM talks about the non-displacement language found in the 2008, 2013, and 2015 OAs depends on whether IBM is talking to itself or to BMC.

a. How IBM Talked About the Non-Displacement Language to BMC

69. On May 23, 2013, IBM Senior Counsel Thomas Hagen sent an e-mail to BMC's legal counsel shortly after the parties signed the 2013 OA. See PX53 at IBM00081348. In it, Hagen explained that IBM “interpret[ed] the intent of [the non-displacement] clause . . . to prevent IBM from unfairly competing with BMC by abusing information and access to BMC software afforded to IBM as an outsourcer providing IT Services.” Id. Citing IBM's prior suggestion that the clause did not “align . . . this intent [to] the principles of U.S. competition law, ” Hagen declared that IBM did “not believe that this language restricts [it] from competing openly and fairly with BMC for customers who are interested in licensing IBM products.” Id. at IBM0081348-49. Hagen continued:

The right, in fact, the obligation, of horizontal competitors to engage in fair competition is a fundamental principle of competition . . . .We believe that any other reading-for instance, a reading that would prevent an IBM Software seller from approaching any customer-BMC licensee or otherwise-would be per se illegal.

PX53 at IBM00081348. Apart from IBM's supposed concerns about U.S. competition law, Hagen offered, as a matter of contract interpretation, that IBM did “not believe the language restricts [BMC's] licensees from directing IBM to implement competing solutions, whether of IBM or any other BMC competitor.”²² Id. at IBM0081349. BMC's legal counsel responded that BMC “believe[d] the provisions in the Outsourcing Attachment speak for themselves.” Id. at IBM0081348.

22 This carefully worded interpretation is correct. The outsourcing attachments governing the conduct of BMC and IBM did not extend to the parties' mutual customers. Therefore, the non-displacement language did not “restrict” the parties' mutual customers from “directing” IBM to do anything, including displacing BMC's products with its own, though it did restrict IBM's ability to follow their direction lawfully.

70. By the close of October 2013-and as the dispute around the NAB displacement unfolded-IBM's Sweetman echoed some of these same interpretations in an e-mail he sent BMC's Jones. DX233 at 1-2. Sweetman explained that the OA did not govern where, as with NAB, “the licensee has independently secured” IT outsourcing rights. Id. at 2. And, while denying that IBM had displaced BMC software at NAB, Sweetman claimed that “even if IBM were somehow deemed to be displacing software . . . acting at the customer's direction would constitute the most obvious example of a ‘valid business reason.'” Id. at 2. Jones disagreed with both points, claiming that the OA alone governed IBM's access and use of BMC products and that the OA's “valid business reason” language applies only “to the discontinuation of BMC products.” Id.

71. BMC continued to raise displacement concerns during the 2015 OA negotiations. Clyne, the negotiator, sent BMC executives an e-mail:

As I am sure you know, BMC has continued to claim compliance issues by IBM in respect to the 2008 and now 2013 OA agreements between the two companies. I'm equally confident that you are aware of IBM's very different view of its compliance conduct having to do with the provisions and terms of those same agreements as well as the behavior of our mutual customers having to do with their own respective license terms, too.

Clyne Dep. 237:14-237:24. IBM's communications with BMC regarding the differing interpretations of the non-displacement language were part of an overall strategy to secure leverage during the negotiations. See, e.g., Clyne Dep. 239:18-239:22 (“Q: you were suggesting to Mr. Dzaluk when he talks to Mr. Ah-Chu, Mr. Dzaluk reinforce the notion that IBM and BMC are worlds apart, correct? A. Yes.”). As Clyne further testified:

[T]he bottom line is I'm just regurgitating what IBM's positions are internally to posture for what I knew would be very difficult negotiations such that I had to have leverage. So, the starting point was to reiterate exactly what IBM's position was vis-à-vis BMC, including what IBM included in its internal presentations and culling these out as potential exposures. I did not share that but that's what's driving this dialogue. It's posture. It doesn't make it fact.

Id. 243:23-244:07 (emphasis added).

b. How IBM Talked About the Non-Displacement Language Internally

72. Internally, IBM's conversations regarding the OA's non-displacement language reflected a more measured interpretation. Mere months before telling Jones that IBM could effectuate a customer-directed displacement, Sweetman sent an e-mail to other IBM employees explaining that IBM's access and consent rights “ba[r] IBM from displacing the BMC products with IBM products for the duration of our services agreement.” PX34 (e-mail dated March 19, 2013). Elsewhere, shortly after the 2013 OA was signed, confidential IBM internal guidance documents advised account teams that the 2013 OA retained the “2008 non-displacement provision, ” which “IBM w[ould] be bound by” under the “current Access and Consent terms.” PX79 at IBM00116122-25. That same guidance noted that the “licenses” that “are IBM owned” were “not subject to displacement restrictions, ” which in turn gave “IBM . . . the flexibility to explore IBM product migration opportunities to reduce capacity over time.”²³ Id. at IBM00116124.

23 IBM's internal guidance also instructed employees not to “assume that the Client's agreement with BMC grants the rights necessary for IBM to provide services” or that the “Client's agreement with BMC's obligation to implement an Access Consent Agreement and/or comply with restrictions on displacement of BMC products with IBM products.” PX79 at IBM00116120.

73. Some IBM employees expressed frustration amongst themselves that the non-displacement provision made “IBM responsib[le] to maintain BMC's revenue” and that “[w]ithout the threat of displacement” they would not be able to ensure that clients received the best software pricing. PX34 at 1; PX24 at 7 (“Current agreement restricts IBM's ability to offer clients technology choice in Outsourcing situations - including when clients demand flexibility. . . . IBM expressed verbally in last proposal our interest in an option to accommodate our clients' demands that was financially viable.”); PX33 at 5 (noting that removing “non-displacement” would enable “greater flexibility, ” an “IBM Need[]” for the 2013 OA). Yet some elements within IBM were mindful that that compliance with the “rules” of its contract with BMC was important, as “not cost[ing] appropriately” could result in “a significant hit to [its] accounts.” PX28 at 2. Compliance issues could, after all, threaten IBM's “profit margins.” PX111 at IBM00050586; see also PX111 at IBM00050583 (“The non-displacement clause is included in the [access and consent agreement] and BMC always tr[ies] to enforce it.”).

74. Notwithstanding these concerns, IBM's Hibert sent an e-mail to IBM outsourcing staff “in the midst of negotiations with BMC” regarding the 2013 OA, “looking to explore all opportunities for potential displacement of BMC products in [IBM's] installed environment.” PX25 at 3. Hibert framed this solicitation as a “[c]all to action, ” stating that the company's software group's “product portfolio has many solutions that displace BMC products” and that the company would be “launch[ing] a focused program . . . identifying migration opportunities and driving them to successful completion.” Id. at 4.

75. Thus, though IBM's efforts to remove the non-displacement language from the 2013 OA were unsuccessful, IBM software and outsourcing personnel began “internal[]” discussions “about how . . . to manage the non-displacement language overall as a business” within four days of executing the contract, even while acknowledging the “general rule” that “all client-owned BMC [software] is now subject to non-displace” restrictions. PX37 at 1; see Findings of Fact, supra ¶ 40. In one e-mail, Craig, who was responsible for managing the global IBM-BMC relationship for IBM, cautioned that “[w]here IBM [is] using the client's license [IBM] cannot displace BMC's product with IBM product. If IBM hold[s] the license then we can do whatever we want regarding replacing BMC's product(s).” PX88 at IBM00079734. According to Craig, he saw the “potential danger” of displacement, and he “issu[ed] th[at] e-mail as a cautionary and conservative note to be aware of that and the potential breaches of any agreement between [IBM and BMC].” Craig Dep. 69:14-69:20;²⁴ see also PX87 at IBM00076667 (“IBM has an agreement with BMC that we do not displace any of its tools in favour of IBM tools.”); PX208 at IBM00058136; PX461 at IBM00078823-824; PX462 at IBM00063122; Trial Tr. 182:10-186:16, Mar. 15, 2022 (Jones); Trial Tr. 69:20-24, Mar. 21, 2022 (Sweetman). IBM's internal guidance- which Craig vetted-acknowledged the same: the 2013 OA restricted IBM from displacing BMC products “[w]here IBM requires the ability to access BMC licenses held by the Client.” PX40 at IBM00063333; see Id. (“No right to displace the BMC products with IBM products”); Trial Tr. 69:03-13, Mar. 21, 2022 (Sweetman); see also PX208 at 13. Beyond e-mails, IBM's internal educational documents-which Craig and others within IBM also reviewed-noted that there were some IT contexts in which buying the BMC license made sense and that license purchases would free IBM from the displacement restrictions. Craig Dep. 77:07-77:19.

24 Craig's own understanding of the 2013 OA was that if IBM owned the BMC product licenses, the non-displacement and discontinuation restrictions did not apply. See Craig Dep. 73:23-74:09. Craig also testified that he could determine the price for the licenses by referring to “the discounts, et cetera contained within the Outsourcing Attachment and applied against the list price exhibits, which are also in the Outsourcing Agreement” that IBM agreed to. Id. 74:18-74:23. However, Craig also explained that the guidance materials would not necessarily reflect IBM's interpretation of the other “valid business reasons” provision because they towed a “conservative, cautionary line” that was intended to prevent “circumstances that might be problematic in the future.” See Id. 93:03-93:16.

76. A little over one month prior to Sweetman's October e-mail to Jones, on September 20, 2013, Mary Quigley, an IBM executive, noted that the agreement with BMC contained “very specific verbiage around software replace.” DX214 at 2; see Findings of Fact, supra ¶ 48. Though IBM executives found certain “BMC software takeout opportunit[ies]” attractive, they were “also concerned.” DX214. Earlier that year, on February 14, 2013-as the parties negotiated the 2013 OA-Hibert advised IBM personnel that IBM could not “migrate, ” or perform a displacement, if IBM operated a license otherwise owned by a mutual BMC customer. PX25 at 1. However, Hibert also acknowledged that IBM could “migrate when [it] own[ed] the licenses.” Id.

77. Even the guidance within IBM that eschewed bright line rules around displacement reflected the more measured discussions IBM employees had internally. For example, a 2014-era IBM internal guidance document illustrating a decision tree for IBM employees included a displacement scenario. PX94 at IBM00081122-23. Though noting that exceptions could apply, that decision tree asked if the “Client already commenced activities to pursue such displacement” and “[i]f so” whether IBM could “document that this activity pre-dated the outsourcing engagement?” Id. at IBM00081123. Far from suggesting an unqualified right to effectuate a client-directed displacement, the internal guidance indicates that IBM entertained the notion that it could displace if the mutual client already “commenced” the displacement. See Id. at IBM00081123.

78. But, by and large, IBM internally understood that the non-displacement provision significantly curbed its ability to effectuate a displacement. Thus, around August 2014, IBM managers advised employees confronted with displacement opportunities that “[w]here we are using the client's license we cannot displace BMC's product with IBM product. If IBM hold[s] the license then we can do whatever we want regarding replacing BMC's product(s).” PX88 at 4. IBM's internal guidance similarly acknowledged that there was “[n]o right to displace the BMC products with IBM products” under the 2013 OA's access requirements. PX40 at 14. That same internal guidance recognized that “[i]n some IT Services contexts, it may make economic sense for IBM to acquire licenses from BMC for the purpose of providing IT Services to a named Client, ” such as when the “Client License Access rights afforded [to] IBM are insufficient.” Id. at 10.

79. By 2015, some of IBM's internal communications reflected an understanding of the non-displacement provision's plain meaning. See Craig Dep. 129:06-130:25 (discussing an internal e-mail in which an accounts team failed to adhere to the agreement with BMC and testifying that IBM may have had “a potential problem” with “replacement” compliance issues) id. 144:03-144:05 (testifying as to the 2015 OA's “release from past indiscretions”); PX226 (with respect to the anticipated 2015 OA, “IBM is free to compete at any time at all accounts that are not [protected accounts] . . . [t]he net is that the restrictions have been removed . . . except for . . . the fifty-four account protected list. Further, the BMC accounts on the list cannot be changed . . . without our specific agreement (potential leverage item in the future FOR U.S. to be sure!).”); PX238 (internal guidance document noting that non-displacement language in the 2015 OA applies to accounts listed on Exhibit K); PX132 at BMC-000165729-730 (BMC confirming that “IBM still has the option to replace BMC software at IBM outsourcing accounts, but to do so, it would be required to pay BMC the cost of the license fee and one year's support at a price discounted by 45 or 72% per the OA”).

80. Except for the reduction in the number of accounts subject to non-displacement, the non-displacement language in the 2015 OA tracks that of the 2013 OA. Compare PX3 §5.1, with PX4 §5.4. Because IBM failed in its efforts to remove both the non-displacement clause in its entirety and then, later, AT&T from the list of accounts subject to it, the company faced significant business “risks.” See PX270 at IBM00000524; see also PX196 at IBM00124445 (internal guidance document noting “Agreement to Allow Displacement at all Accounts less a List of fifty-four Named Accounts”); Stafford Dep. 229:14-21 (testifying as to conversations within IBM regarding the “risk” of agreeing to unchanged non-displacement language).

81. Two days prior to executing the 2015 OA, IBM internally acknowledged in an internal guidance document that its planned displacement of AT&T's BMC products with IBM products was one of the deal's “cons.” PX228 at IBM00124048. Tracking the distinction between the 2015 OA's use of “displacement” and “discontinue, ” that internal guidance document notes that the “non-displace clause [was] eliminated for IBM portfolio minus a list of fifty-four protected accounts, ” including “AT&T [and two other companies] where displacements are currently occurring/planned.” Id. at IBM00124046. For those fifty-four accounts, including AT&T, “existing language with rights to ‘discontinue for other valid business reasons' will be retained.” Id. Displacement concerns, the guidance went on, were “to be addressed by Legal.” Id. at IBM00124048. But as a general matter, IBM understood that under the 2015 OA, it could not “displace BMC products with IBM in any of the fifty-four clients on Schedule K but can displace with other ISV products (third parties).” PX240 at IBM00050260; PX282 at IBM00050628 (internal guidance noting that “[t]here are no limitations on IBM's ability to displace BMC Customer Licenses with IBM products for any client other than those listed amongst the fifty-four Accounts in Exhibit K . . . and only if IBM is providing strategic outsourcing services.” But see PX282 at IBM00050623 (“[A]s a reminder to everyone, the old OA did not allow IBM to displace any BMC products with IBM products, other than for valid business reasons.”) IBM likewise understood that the “[c]urrent displacement plans at AT&T” would likely raise a “future dispute with BMC on [the] terms” of the contract.” PX258 at IBM00049965.

82. Also, just prior to the 2015 OA's execution, IBM's legal department prepared a PowerPoint presentation on the outstanding negotiation disputes.²⁵ See PX150. In a slide titled “Areas of Dispute within Agreement, ” IBM noted that it and BMC had differing views on “Displacement of Client owned Licenses with IBM Products (BMC Claim $769m).” Id. at IBM 00124242. IBM noted that BMC took the position that “IBM is specifically forbidden from offering displacement of BMC products with IBM products under the terms of the OA . . . and can only discontinue use of products for other valid business reasons.” Id. In contrast, IBM asserted its position was that “IBM must compete in the marketplace and will meet client requirements for . . . solutions, ” and that the “basis and intent of [the] OA was/is that IBM will not use any BMC confidential information to compete.” Id. “Valid business reasons, ” it claimed, “include[d] reasons to displace.” Id. Still, in noting the “cons” to the proposed 2015 OA, IBM observed that the non-displacement provision would “limit IBM's ability to offer Displacement at up to 50 named customers, ” with “other concerns to be addressed by Legal.” Id. at IBM 00124244.

25 BMC created its own slideshow in the leadup to the 2015 OA's execution wherein it noted that IBM's “core strategy [was] to push back on core terms of the 2013 OA to the legal maximum.” PX152 at BMC-000118299.

83. IBM knew it could have complied with its obligations under both Project Swallowtail and the 2015 OA by purchasing its own licenses to the products it had agreed to displace. IBM's internal guidance documents even specified the steps IBM should take when displacing a customer's products with IBM products under the OA, stating that IBM could “[n]egotiate with BMC to acquire equivalent BMC licenses to be held by IBM that do not include the restrictive [non-displacement] language.” PX79 at IBM00116135; PX282 at IBM00050626 (“When does it make sense for IBM to buy a BMC license?”). But IBM's behavior and its internal statements establish that it never had any intention to purchase the licenses that would have allowed it to comply with both agreements. See, e.g., PX228 (three days before execution of 2015 OA, IBM internally notes “potential risks for planned displacement [at] AT&T not in pricing case”).

E. BMC's Performance Under the 2015 OA

84. Section 5.1 of the 2015 OA requires BMC to “make available to [IBM] a representative copy of its typical BMC Customer license agreement [BMC EULA] and [to] inform IBM] in writing of any material differences between the BMC EULA and BMC Customer's agreement after [IBM] provides written notice of Access and Use.” PX4 section 5.1. According to Sweetman, IBM sought a copy of BMC's agreement with the mutual customer both to understand what additional conditions might restrict their conduct and “understand any rights the customer might have.” Trial Tr. 47:08-15, Mar. 21, 2022 (Sweetman). Despite this, BMC never advised IBM that AT&T negotiated for outsourcer use rights in certain circumstances, which are not contained in the standard EULA.²⁶ IBM argued at trial that, if it had known this, it could have proceeded as an outsourcer under the AT&T-BMC license agreement, rather than proceeding under the 2015 OA and abiding by the non-displacement provision and other OA restrictions. In short, IBM argued that the AT&T EULA is materially different than the standard EULA and BMC's non-performance under the 2015 OA forecloses its breach of contract claims. See Id. 49:01-49:05 (Sweetman) (testifying that he believed the AT&T-BMC license agreement to materially differ from the standard BMC EULA).

26 The standard EULA provides that BMC's “Customer will not.. . . . allow the products to be used by an outsourcing or service bureau provider on Customer's behalf.” DX007 (EULA) section 4(e).

F. BMC's Claims for Misuse of Confidential, Trade-Secret, and Other Information Against IBM

85. BMC claimed that IBM violated section 8 of the MLA, federal and state trade secrets law, and common law constraints by misusing confidential and trade-secret information. See Dkt. 295 at 30-31, 37-46. To understand BMC's claims, the court reviews how outsourcers like IBM interact with end users, like AT&T, and software vendors, like BMC. It then discusses the evidence relevant to each of BMC's misuse-of-information claims.

1.How IBM Generally Worked with BMC to Assist AT&T

86. Outsourcers function as agents and contractors for the customers they serve. See generally Trial Tr. 26:23-27:07, Mar. 16, 2022 (“An outsourcer is contracted to take over, basically, the runnings of the customer's environment.”) (McGuinn); Trial Tr. 65:13-19, Mar. 16, 2022 (“[T]he outsourcer's role is to manage the operation of the computer system. They would set it up. They would keep it running. They would monitor it and try to determine if their resources, for example, were running out so that they could proactively avoid anything that would bring the system down.”) (Román). By supporting customer environments, outsourcers develop a “very deep knowledge” of that environment and the product tools, features, and functions that the customer uses. Pachnos Dep. 63:21-64:21. IBM has provided AT&T with IT outsourcing services for decades; the current master IT services agreement between AT&T and IBM has existed since 1999. DX572 (Skinner Decl.). IBM, therefore, has intimate familiarity with AT&T's mainframe environment and how BMC software operates within that environment. Conway Dep. 35:13-20.

87. As AT&T's outsourcer, IBM is responsible for making sure that the software that AT&T has chosen works correctly and functions with AT&T's business applications. DX572 (Skinner Decl.) ¶ 4. It is also IBM's job to maintain AT&T's existing environment-which, during Project Swallowtail, included BMC products. Trial Tr. 69:11-70:01, Mar. 17, 2022 (Skinner). To maintain a customer's existing environment, outsourcers like IBM regularly step into a customer's shoes to deal with ISVs like BMC. Trial Tr. 21:23-22:5, 26:23-27:70, Mar. 16, 2022 (McGuinn).

88. AT&T-and IBM as AT&T's agent-had access to BMC technical support to facilitate the maintenance of AT&T's environment. AT&T contracted with BMC for this support. See 2014 Mainframe Order sections 2.2, 2.6. AT&T relied on IBM to communicate with BMC and obtain the necessary support that AT&T is entitled to under its contract with BMC. Conway Dep. 67:23-68:11, 166:24-167:19.

89. BMC's technical support is accessible on the restricted portion of BMC's website behind a login firewall that is only accessible with a current customer ID number and valid login credentials.²⁷ Trial Tr. 16:12-18:02, 21:10-15, Mar. 16, 2022 (McGuinn); PX399 (BMC “Create BMC/Account Developer” webpage); PX400 (BMC login portal to access BMC technical support); Trial Tr. 204:20-205:12, Mar. 16, 2022 (Román).

27 To be an authorized user for a customer who has licensed a BMC software product, a party must register with BMC and provide a valid code associated with the purchased maintenance and support to obtain the necessary login credentials to access BMC's online platform for technical resources and information. Trial Tr. 16:12-18:02, Mar. 16, 2022 (McGuinn); see also PX379. BMC's “Web Site Terms of Use” further specify that the information beyond the login firewall is “restricted product information” and that “[t]his restricted information is considered confidential and proprietary information of BMC.” Trial Tr. 20:08-21:15, Mar. 16, 2022 (McGuinn); PX402 (“BMC Web Site Terms of Use”); PX383 (same). This prohibition extends to any information provided by BMC's technical personnel via the support platform. Trial Tr. 20:08-21:15, Mar. 16, 2022 (McGuinn). Prior to disclosing support, BMC confirms that a current BMC customer with a valid support agreement has made the request. Trial Tr. 17:22-18:02, 20:21-21:15, Mar. 16, 2022 (McGuinn).

90. In response to IBM's requests, BMC provided IBM with the data analysis, procedures, and code fixes to resolve the issues raised in five technical support cases. Trial Tr. 97:8-132:16, Mar. 16, 2022 (Román); see also Trial Tr. 12:11-15:18, Mar. 16, 2022 (McGuinn). These code fixes came in the form of program temporary fixes (“PTFs”), which resolve errors or abnormal endings (“ABENDs”) that may or may not have appeared because of tasks associated with ongoing software version upgrades, with Project Swallowtail, or independently of either.

91. PTFs provide a temporary change to a BMC product to fix a problem or correct a defect encountered when operating BMC software in a customer's environment and to ensure that BMC software is functioning as intended. DX572 (Skinner Decl.) ¶ 37; DX568 (Sessarego Decl.) ¶ 8; see Trial Tr. 41:22-24, Mar. 16, 2022 (McGuinn). BMC provides PTFs only when it has identified an issue with a BMC product. See Sessarego Dep. 76:01-07; Pachnos Dep. 54:24- 55:14. BMC would not provide a PTF unless there was a confirmed problem with a BMC product. Trial Tr. 42:08-10, Mar. 16, 2022 (McGuinn).

92. BMC created some PTFs before IBM opened a support case for AT&T, prompted by similar problems encountered by other customers. Sessarego Dep. 71:16-73:20. There are indexes of PTFs that BMC previously created that BMC makes available on its website for customers to utilize as needed. Pachnos Dep. 59:15-60:10; see also Trial Tr. 48:11-16, Mar. 16, 2022 (McGuinn). BMC created other, newer PTFs because of a problem with the software identified by AT&T or IBM. Sessarego Dep. 73:21-25. After a PTF is created, BMC makes the PTF available to all customers who contract for BMC support. Pachnos Dep. 59:03-60:10.

93. BMC provides its software in executable or object code format. Pachnos Dep. 40:11-13, 42:8-10; Trial Tr. 94:10-12, Mar. 17, 2022 (Skinner). Object code and executable code are not trade secret information. See Pachnos Dep. 42:14-16, 43:18-24. IBM never received source code for BMC products in its role as an outsourcer for AT&T. See DX572 (Skinner Decl.) ¶¶ 8, 33; Pachnos Dep. 40:11-15, 41:8-10.

2. Specific Incidents

94. BMC claimed that its trade secrets were disclosed in the following five technical support cases: Case No. 332327 (opened on May 9, 2017), PX370, and Case No. 343507 (opened on May 29, 2017), PX371, which concern errors encountered as a result of Project Swallowtail related to BMC's System Performance for IMS (MainView for IMS) product; Case No. 198877 (opened on August 17, 2016), PX311, and Case No. 210489 (opened on September 9, 2016), PX330, which concern errors encountered as a result of Project Swallowtail related to BMC's Recovery Manager products; and Case No. 120397 (opened on April 6, 2016), PX320, which concerns errors encountered as a result of Project Swallowtail related to BMC's System Performance for IMS (MainView AutoOperator) and BMC's MainView SRM Allocation products. Trial Tr. 97:8-132:16, Mar. 16, 2022 (Román). But BMC's expert, Kendyl Román, offered no opinion about trade secrets. Id. at 136:09-14 (Román). And Román conceded that the PTFs and ABEND fixes could sometimes be found in publicly available manuals. See Id. at 205:19-206:01 (Román).

a. BMC Technical Case No. 00120397

95. As described by BMC's expert, this case involved the replacement of a subset of BMC's MainView products with non-IBM products. Trial Tr. 203:12-17, Mar. 16, 2022 (Román). An IBM employee sought information regarding errors connected to a BMC product. DX656.001; Román Rpt. ¶ 204. BMC analyzed dumped data and determined that DTS software-not BMC software-caused one of the ABENDs. DX656.009; Román Rpt. ¶ 204. BMC also discussed a configuration setting regarding the code that had an ABEND. DX656.006; Román Rpt. ¶¶ 205, 278. The standard configuration setting discussed by BMC support is not a trade secret and is also publicly available on BMC's website. DX667; DX668; DX669; DX684.

96. Within AT&T's dump was a “string that's in English that says ‘Copyright detail software 1991, license material proprietary of DTS software.'” Trial Tr. 129:15-18, Mar. 16, 2022 (Román); DX656.009. Mr. Román described this as “the eye catcher, ” i.e., “a string that's human readable in the middle of all this gobbledygook, [which] catches your eye.” Trial Tr. 129:10-21, Mar. 16, 2022 (Román); DX656.009. The “eye catcher” is what identified that the first ABEND was related to DTS software. See Trial Tr. 129:19-21, Mar. 16, 2022 (Román). Thus, the “[r]oot cause, ” which Mr. Román identified as “DTS Software, ” see Id. 130:22-24, was identified in plain English in AT&T's data dump. DX656.009.

b. BMC Technical Case No. 00198877

97. An IBM employee experienced a problem during a disaster recovery drill for AT&T. DX658.001; Román Rpt. ¶¶ 214-15. BMC explained that the data set being used was created by an IBM product, not a BMC product, and thus required a different format to be processed by BMC's Recovery Manager. DX658.004. Accordingly, BMC disclosed “already-existing functionality and configuration settings” for a product. Hartley Rpt. ¶ 89. The information necessary to identify the problem experienced in this support case-that the data set was not in the proper format for use with BMC's Recovery Manager-is available on BMC's website. Id. ¶ 89 & n.63; DX673; DX694.

c. BMC Technical Case No. 00210489

98. An IBM employee contacted BMC support after determining that a crash occurred because the authorization module was not found for BMC's product, Recovery Manager, which was an important disaster-recovery tool. DX659.001; Román Rpt. ¶¶ 220, 283; Trial Tr. 125:23- 25, March 16, 2022 (Román) (testifying that Recovery Manager crashed because of a user ABEND and that an “authorization module was not found for Recovery Manager”). An authorization module is like a password that enables the program to run. Trial Tr. 126:03-09, March 16, 2022 (Román). The information provided by BMC regarding which authorization modules to use and where they are located is available on BMC's public website. See DX670; DX671; DX724. Indeed, during cross-examination, Mr. Román testified:

Q. So, the information that was provided by BMC in this case was information that, if IBM had looked hard enough, they would have found on . . . BMC's publicly available world wide web website?

A. Well, I can't say what was available at that time, but just to speed things along, I'll say if it's publicly-if it was publicly available at the time, there are certain pieces of information that are publicly known, yes.

Trial Tr. 202:01-202:08, Mar.16, 2022 (Román).

d. BMC Technical Case No. 00332327

99. IBM first discovered the ABEND that led to Technical Case No. 00332327 in February 2017 while it was solving another technical issue. DX554.012. The ABEND arose as IBM was displacing BMC's product, MainView, with its own product, Omegamon. Trial Tr. 185:15-187:15, Mar. 16, 2022 (Román). Specifically, IBM found an ABEND with code U3042. DX554.012. After further analysis, in May 2017 IBM determined that an 0CE ABEND immediately preceded every U3042 ABEND, and that a BMC MainView for IMS module appeared in connection with that ABEND. DX554.026, .035.

100. Once IBM's technical support team had isolated the S0CE and U3042 ABENDs and tied them to BMC's MainView product, IBM (acting as AT&T's IT outsourcer) contacted BMC on May 9, 2017, to open Technical Case No. 00332327. DX663.001-002. The BMC help desk reviewed the AT&T dumps that IBM had isolated and concluded that a bit was set to “on.” DX663.022. On May 10, 2017, BMC provided PTFs to address the S0CE and U3042 ABENDs. DX663.010. BMC would not have provided PTFs if there was not a confirmed problem with its product. See Trial Tr. 42:08-10, Mar. 16, 2022 (McGuinn).

101. At the time BMC was responding to this support request, BMC was aware that an IBM replacement product-Omegamon-was being used in AT&T's environment. Trial Tr. 188:08-21, 190:05-09, Mar. 16, 2022 (Román). At no point in responding to this support request did BMC suggest that IBM was doing anything improper with BMC's information. Id. at 189:05- 14, (Román).

102. In his report, Mr. Román called out four purported trade secrets disclosed in this support case. First, “two custom PTFs for Mainview, and the accompanying PTF descriptions” were provided by BMC support. Román Rpt. ¶ 244. The description accompanying the PTF provides no technical information-it “merely provides an overview of ‘what' the PTF does, not ‘how' the PTF works internally.” Hartley Rpt. ¶ 145.

103. Second, “a description of the internal functions performed by Mainview's code related to the accumulation of event timing fields and the PSW PGM MASK” was provided by BMC support. Román Rpt. ¶ 244. BMC support provided no substantive information regarding the accumulation of event timing fields-merely the fact that event timing fields were accumulating. Hartley Rpt. ¶ 126.

104. Third, “the changes that the PTFs made” were explained by BMC support. Román Rpt. ¶ 244. As explained by IBM's expert, “BMC's PTFs did nothing more than apply basic and well-understood principles of coding”-clearing a hardware register before use. Hartley Rpt. ¶ 128. BMC was describing “a simple ‘housekeeping' function.” Id. Similar information is also provided in publicly available IBM documentation. Id. For instance, a z/OS Programmer's Guide, freely available on IBM's public website, instructs that “[a] program should save the values contained in the general registers when it receives control and, on completion, restore these same values to the general registers.” DX746.001.

105. Fourth, “the analysis performed by BMC's product development group related to the root cause of errors occurring in the mainframe environment” was identified by Mr. Román as trade secret information. Román Rpt. ¶ 244. BMC analyzed information dumps from AT&T's mainframe environment to try to identify the root cause. Hartley Rpt. ¶ 129. This information, if proprietary at all, is proprietary to AT&T, not BMC. Id.

106. At trial, Mr. Román testified that “finding the problem” is what BMC contributed to IBM. Trial Tr. 205:19-206:7, Mar. 16, 2022 (Román). BMC's “identification of the problem” in this support case, according to Mr. Román's report, was the following:

PSW Key mask set on for Significance Exceptions; What happened here is that we have fields in our TRN record that are zero and we are accumulating these various event timing fields and when this particular PSW Key mask is on and the field is zero, you get the S0CE during floating point ADD. Did the setting for the PSW Key Mask recently change on the systems encountering the problem and what is the setting compared to other systems that haven't experienced the problem?

DX663.012; Román Rpt. ¶ 237.

107. However, Mr. Román's testimony ignores that this was not new information to IBM. On March 6, 2017, IBM referenced the “Systems Codes manual, ” which listed possible causes of the ABENDs, including that “[a]t least one reserved bit in the ESPIE parmlist is incorrectly set on.” DX554.019. After eliminating other possibilities, IBM stated: “That leaves a reserved bit being set on.” DX554.020. IBM also identified that “[t]his is a Significance exception.” DX554.022. This is the same information conveyed by BMC months later, on May 9, 2017, when BMC stated that the “PSW Key mask set on for Significance Exceptions” and that the ABEND was occurring because the “PSW Key mask is on.” DX663.012. Thus, BMC's purported identification of the problem was not even new or unique.

108. Further, while Mr. Román testified that IBM asked for more information from BMC after BMC provided the PTFs because “they were still concerned about what the root cause was, ” Trial Tr. 107:01-07, Mar. 16, 2022 (Román), IBM explained that it asked “more questions” because it “want[ed] to make sure that BMC PTF will not introduce any other issue[s] to the system, ” DX554.041.

109. Mr. Román also testified about his understanding that there was a purported seven-month period before IBM contacted BMC for support concerning the problem. See, e.g., Trial Tr. 104:04-105:21, March 16, 2022 (Román). However, Mr. Román simply ignored that the first several months of this seven-month period consisted of IBM investigating an unrelated S46D ABEND. Hartley Rpt. ¶¶ 106-07. This unrelated S46D ABEND was ultimately determined to be related to a product from a different software vendor, Informatica, and was resolved separately by IBM. Id. ¶ 120. IBM did not encounter the ABEND connected to BMC's MainView product until February 2017. Id. ¶ 108. Between February 2017 and when BMC was contacted in May 2017, IBM was investigating that ABEND. Id. ¶¶ 108-16. This investigation eventually led to IBM connecting the ABEND to BMC's MainView product. Id. ¶ 116.

e. BMC Technical Case No. 00343507

110. BMC support was contacted after a PTF it had developed kept failing. DX664.001; Román Rpt. ¶ 250. After reviewing AT&T system dumps that were provided, BMC “identified a problem that was exposed with BQ12263 PTF, ” DX664.028, i.e., one of the PTFs that was applied in connection with Case No. 00332327, DX664.030. As IBM's expert explained, “it appears that the original PTF provided by BMC may simply have reset configuration settings that had been customized for the AT&T environment back to standard BMC default values, thus causing . . . abends. After the problem was identified, BMC recommended certain configuration settings.” DX664.023; Román Rpt. ¶ 251. BMC's recommendations are publicly disclosed on its website. See, e.g., DX665; DX676; DX677. BMC also provided two PTFs, which are not trade secret information. DX664.010-011.

G. BMC's Damages Models

111. BMC argued that license fees for the BMC products which IBM displaced with IBM products at AT&T in Project Swallowtail are “payable” under the MLA's plain text. But even if they were not, BMC presented a license fee damage model that measures the value-based on the contract's text-of the licensing rights that IBM used without payment. Separately, BMC presented a damage model measuring its lost profits and the alleged misuse of its confidential information. Finally, arguing that the MLA's damage limitations do not apply because of IBM's wrongful, fraudulent conduct, BMC seeks exemplary damages. The court reviews each in turn.

1. The License Fees Are Payable

112. The court finds that BMC is entitled to the full amount of the lost license fees for its claim for breach of section 5.4 of the 2015 OA. The court previously ruled that pursuant to MLA section 10, any “amount paid or payable by [IBM] for the license to the applicable product giving rise to the claim” is not subject to the $5 million-per-product limitation. Dkt. 586 at 8-9. Because these “license” amounts are “greater” than $5 million per product-and because their value can be ascertained within the four corners of the 2015 OA-the court finds that they reflect “payable” sums for the bundle of rights IBM exercised in this case, including the right to displace BMC products with IBM products.²⁸ See Conclusions of Law, infra ¶¶ 166-76.

28 Even if MLA sections 9-10 are enforceable, they only apply to BMC's OA breach claims if the MLA and OA are one integrated agreement. This court previously ruled that they are one integrated agreement; BMC has reserved the right to challenge this ruling.

2. License Fee Models

113. BMC seeks to recover damages based on the product-specific licensing and interrelated support fees prescribed by the 2015 OA for the bundle of rights, including the right to displace an Exhibit K customer's mainframe software, that IBM used but did not pay for.²⁹ See generally PX4. As explained above, under OA section 1.1, IBM had to choose from multiple options when deciding how to use BMC products while providing IT outsourcing services for a mutual customer. PX4 section 1.1. IBM elected to access and use AT&T's BMC license for no additional fee but that entailed IBM's agreement not to displace AT&T's BMC products with IBM products and to adhere to the other limitations in OA section 5. PX4 §§5.1, 5.4.

29 IBM is one of the world's largest and most sophisticated information technology companies and understood that it was contracting for a bundle of rights. Its own internal guidance reflected as much. See PX282 at IBM00050629 (listing the rights and restrictions associated with “client license access, ” including the “right to displace with IBM products”).

114. IBM could have elected to purchase its own BMC licenses under OA section 1.1, which would have given it the right to displace BMC's products on AT&T's mainframes.³⁰ Trial Tr. 176:20-177:12, 194:08-195:10, Mar. 14, 2022 (Ah Chu). BMC's core damages models calculate BMC's damages based on the licenses and support that IBM failed to pay for under the OA.

30 At summary judgment, the court rejected BMC's argument that 2015 OA section 1.1 required IBM to make a different election than the one it made and purchase its own licenses. See Dkt. 586 at 6-7.

115. Specifically, BMC seeks recovery of $717 million in unpaid license and support fees for the fourteen BMC products that were displaced with IBM products at AT&T in violation of OA section 5.4 and $791 million for the nineteen BMC products that BMC claims IBM improperly accessed and used while performing Project Swallowtail in violation of OA section 5.1. Trial Tr. 220:13-221:10, Mar. 16, 2022 (Ratliff). BMC seeks the $717 million in unpaid license and support fees not only for its claim for breach of OA section 5.4, but also for IBM's fraudulent inducement of BMC into the 2015 OA.³¹ Trial Tr. 222:22-223:4, Mar. 16, 2022 (Ratliff).

31 BMC asserted several claims and associated remedies in this lawsuit. Once the court resolves BMC's claims, if the court finds and concludes that BMC is entitled to recover on multiple claims, the court will require BMC to make an election of remedies (if necessary) as part of entering a final judgment in this case. For example, if the court finds and concludes that BMC can recover the $717 million in license-fee damages on either its OA section 5.4 breach claim or its fraud claim, BMC's election of that remedy under the OA section 5.4 breach claim would preclude BMC from also recovering the AT&T lost profits under that claim (discussed below). But BMC's election of that remedy under the fraud claim would not foreclose recovery of certain other remedies, such as exemplary damages based on fraud.

116. BMC also seeks $109 million in unpaid license and support fees for the three BMC products that were the basis for IBM's breach of MLA section 8, misappropriation of trade secrets under TUTSA and DTSA, and common-law unfair competition. Trial Tr. 221:11-17, Mar. 16, 2022 (Ratliff). These amounts represent unpaid licensing revenue recoverable for each of those claims. Trial Tr. 257:14-22, 14:24-15:12, Mar. 17, 2022 (Ratliff) (“[I]ts the actual loss to BMC of license fees from IBM related to IBM's use of the products and support during the displacement.”) (Ratliff). For trade-secret misappropriation, BMC asserts that the same calculation represents a “reasonable royalty” for IBM to obtain the right to use BMC's trade secrets for the purpose of displacement as it did for Project Swallowtail.

117. The basis for each of these calculations is the sum amount the parties agreed upon in sections 6 through 8 of the 2015 OA for IBM to purchase its own BMC licenses, including having the right to displace BMC products. PX4. The licensing fees for BMC's products are stated in section 8 of the 2015 OA, which details the price for each product (in Exhibit H to the 2015 OA), the applicable discounts, and the contractually required “support” fees associated with any payment of licensing fees. PX4; Trial Tr. 216:05-219:22, Mar. 16, 2022 (Ratliff); see also Stanton Dep. 102:16-102:20 (“Q: Is it consistent with your understanding that if IBM elected to acquire licenses from BMC, that those prices were set forth in the . . . outsourcing attachment? A: Yes.”). The products at issue in this lawsuit are “System Management Products” set forth in section 8, and IBM received a 72% global minimum discount off of the price listed in Exhibit H for those products. PX4; Stanton Dep. 207:02-207:07 (“Q: What did you mean by ‘IBM has good pricing with BMC?' A: As per the OA, we had a 72 percent discount.”). The interrelated support fees were “calculated at 20% of the net price, where the net price is the List price set forth on Exhibit H, less the [72%] discount.” Id. at Exhibit H (IBM00049713-IBM00049717); see PX112 (“The calculations on displacement values actually reflect a license price and one-year of support for the accounts they identify using IBM's existing discounts (i.e., 72% on the license with maintenance at 20% of the license cost).”); Trial Tr. 217:07-218:04, Mar. 16, 2022 (“[T]he basic math on the support was the years times 20 percent times the net of discount license fee.”) (Ratliff).

118. IBM rejected BMC's license-fee damage model. Putting aside its general disagreement, IBM contended that it would have been entitled to more than a 72% discount even if it were liable for the license fees. See, e.g., Trial Tr. 86:02-17, 87:15-20, 88:02-10, 89:03-19, 92:17-23, Mar. 22, 2022 (Gerardi). But see Findings of Fact, supra ¶ 79. Although the parties agreed to a 72% global minimum discount in the OA itself, IBM argued that the parties have sometimes negotiated larger discounts and that IBM may have been able to obtain a greater discount if it had negotiated to purchase the licenses. See Trial Tr. 108:01-16, Mar. 21, 2022 (Sweetman) (testifying that the 72% discount is a “starting point for negotiations”).

119. The court finds and concludes it is proper to use 72% as the discount in calculating licensing-fee damages. The parties agreed to 72% in the OA, and that negotiated, contractually specified figure provides a reliable, non-speculative basis for ascertaining damages. See PX4 at 4 (OA section 8 provides IBM “will be entitled to a global minimum discount of 72% off the Listed Price in Exhibit H for all purchases of new licenses”). BMC has a valid contractual basis for insisting on that discount rate.

120. Notwithstanding the contract's clear, express terms, BMC used the 72% discount when showing IBM how to calculate the license prices under the OA, and IBM never contested that and, indeed, used that in its own analysis during negotiations. See, e.g., PX440 at BMC-000012586; PX132; PX463. Moreover, IBM's internal Guidance Documents consistently refer to the 72% discount rate for acquiring licenses. See, e.g., PX282.10 (2015); PX208.13 (2013). IBM cannot undercut the OA-specified figure through its own speculation that the parties hypothetically might have negotiated a different discount than the 72% rate stated in the OA. Indeed, parties to any contract could always choose to re-negotiate the terms of their contract at any time, but that hypothetical possibility cannot be used to undermine the legal effect of a negotiated, contractually set term such as the 72% discount rate. IBM cannot render BMC's damages model speculative by itself speculating that the parties might have decided on different terms. Further, it was IBM that chose to forgo its ability to negotiate for a larger discount to purchase the licenses when it instead exercised displacement rights without paying for them. Importantly, again, IBM does not dispute any other variable or mathematical step in Ratliff's calculation of license-fee damages. IBM's damages expert Gerardi walked through the other variables and confirmed he had no reason to disagree with any of the other variables or inputs used by Ratliff. Trial Tr. 86:02-17, 87:15-20, 88:02-10, 89:03-19, 92:17-23, Mar. 22, 2022 (Gerardi).

121. The court finds, based on the text of the contract, that the formula, variables, and inputs used by Ratliff in PX446 are the proper basis for calculating the license fees under the OA. The court finds that the calculation by Ratliff is correct and that all the numbers in it are correct.³²

32 Gerardi admitted he had no disagreement with Ratliff's MIPS input. A discussion arose at trial over this input, and IBM agreed it “would not dispute what Gerardi testified to” on this and assented when the Court stated the view that there is not “any evidence to dispute [the MIPS number] in some other way when . . . [IBM's] expert witness has testified that's the right number.” Trial Tr. 137:19-142:17, Mar. 22, 2022. Relying on IBM's representations, the court determined that no further rebuttal testimony was necessary. Id.

122. Using the license-fee model, the amount BMC is entitled to recover for each claim differs depending on the number of impacted products for each particular claim. As BMC's expert Ratliff explained, it would have cost over $717 million in license and support fees for IBM to purchase its own set of licenses to the fourteen products it displaced with IBM products. Relatedly, it would have cost approximately $791 million for the nineteen products that IBM improperly accessed and used while performing Project Swallowtail (these nineteen products include the fourteen products displaced by IBM products). BMC also argued that the $717 million for the fourteen products at issue in BMC's breach of section 5.4 claim applies to its fraud claim. Finally, BMC argued that it accrued $109 million in damages for the three products at issue in BMC's claims for misuse of confidential information, misappropriation of trade secrets, and unfair competition by misappropriation. Trial Tr. 222:22-223:04, Mar.16, 2022 (Ratliff).

123. The court finds that BMC provided sufficient evidence, under the applicable legal standards, of both the existence (or “fact”) and the amount of damages in the form of its lost licensing fees and associated support services totaling approximately $717 million for IBM's breach of OA section 5.4 and $717 million for IBM's fraud. See Conclusions of Law, infra ¶¶ 166-76. Furthermore, BMC demonstrated these damages were foreseeable and within the contemplation of both parties as evidenced by the contract's text, and that BMC established the amounts of its licensing-fee damages with reasonable certainty.

3. Lost Profits

124. As another type of recovery, BMC seeks damages for $104.5 million in lost profits from AT&T. Trial Tr. 249:24-250:11, Mar. 16, 2022 (Ratliff). This number consists of the fees that AT&T would have continued to pay BMC if IBM had not caused AT&T to terminate its relationship with BMC through IBM's misconduct, including by displacing BMC's AT&T products with IBM's own products. Trial Tr. 243:07-10, 247:3-06, Mar. 16, 2022 (Ratliff). BMC's expert Ratliff calculated these damages using a baseline from BMC's historical core mainframe revenues from AT&T, based on a variety of considerations he outlined, and then forecasted BMC's lost revenues from the loss of AT&T as a mainframe customer. Trial Tr. 226:18-228:05, Mar. 16, 2022 (Ratliff).

125. But the evidence at trial does not support these damages. Though BMC had a long, successful relationship with AT&T, AT&T independently decided to displace BMC software. Trial Tr. 187:22-25, Mar. 14, 2022 (Ah Chu); Trial Tr. 223:7-11, Mar. 16, 2022 (Ratliff); DX019.029; Trial Tr. 63:24-64:05, Mar. 17, 2022 (Skinner); DX460.30 (Ridge notes); Brickhaus Dep. 40:03-12. Moreover, the evidence showed that AT&T was planning on transitioning away from mainframe software. Findings of Fact, supra ¶¶ 43-45. So, AT&T's decisions and conduct-not IBM's-are most consequentially tied to BMC's lost profits from AT&T's .³³

33 BMC's expert did not quantify the acceleration of the Project Swallowtail as a consequence of IBM's participation, even as he asserted it would have taken AT&T longer to complete Project Swallowtail without IBM's help. Trial Tr. 154:01-20, March 16, 2022 (Román).

126. A more credible lost profits model would have based its estimation on evidence showing how IBM's participation in Project Swallowtail accelerated the completion of AT&T's goals and the resultant loss of revenue for BMC. BMC's model, however, does not do this. See Ratliff Dep. 208:13-23. Instead, BMC's model is predicated on a counterfactual assumption that AT&T would have chosen to use the BMC software forever but for IBM's alleged breach of section 5.4.

127. Therefore, the court finds that BMC's lost profits model is too causally attenuated from IBM's own conduct and that BMC has not proven lost profits for IBM's breach of section 5.4 of the 2015 OA.

4. Damage Limitations in the MLA

128. IBM has asserted that MLA sections 9 and 10 limit BMC's potential recovery. The court sets forth the applicable law below in the Conclusions of Law. See Conclusions of Law, infra ¶¶ 207-10. As explained further below, the court finds and concludes that MLA sections 9 and 10 are not enforceable against any of BMC's claims in this case because of IBM's intentional wrongdoing.

129. Specifically, based on all the evidence recounted above concerning IBM's conduct, the court finds that IBM's conduct constitutes “intentional wrongdoing” as that term has been defined under New York law governing the enforceability of exculpatory damages clauses. As explained above, IBM negotiated and entered into the 2015 OA with no intention whatsoever of performing thereunder, having previously signed a secret, conflicting agreement with AT&T to perform displacements pursuant to Project Swallowtail. IBM never intended to (and never did) pay for the right to displace. IBM's scheme to defeat BMC's contractual rights cheated BMC-a software company wholly dependent on the licensing of its intellectual property-out of hundreds of millions of dollars it was entitled to receive under the contract in exchange for the rights IBM exercised. Based on all the foregoing facts, the court finds that IBM's conduct in this case was both fraudulent and malicious. See Id. The court also finds that, through its misconduct in this case, IBM acted in bad faith. See Id. The court further finds that, through its dealings with BMC and its performance of Project Swallowtail, IBM acted with reckless indifference to the rights of others, namely BMC. See Id. As a result, these contractual limitations provisions are unenforceable.

5. Punitive Damages

130. BMC also seeks punitive damages based on both IBM's fraud and IBM's “willful and malicious” misappropriation of trade secrets. Because the court finds and concludes elsewhere in these findings and conclusions that MLA sections 9 and 10 are unenforceable under New York law governing enforceability of contractual damages limitations, those provisions do not limit BMC's ability to recover punitive damages for these claims. See Conclusions of Law, infra ¶¶ 207-10.

131. The court finds and concludes elsewhere in these findings and conclusions that IBM committed fraudulent inducement. The court finds by clear and convincing evidence that the harm with respect to which BMC seeks recovery of punitive damages resulted from fraud, and as to each and every element of the fraudulent-inducement claim found for liability and damages, the court hereby finds by clear and convincing evidence that BMC has established all elements of fraudulent inducement. See Conclusions of Law, infra ¶¶ 185-205.

132. The court finds that awarding punitive damages is proper here and that $717,739,615.00 is an appropriate award in this case. In making that determination, the court has considered all evidence relating to the nature of the wrong, the character of the conduct involved, the degree of culpability of IBM, the situation and sensibilities of the parties concerned, the extent to which such conduct offends a public sense of justice and propriety, and IBM's capacity to pay. Here, the evidence of IBM's deliberate plan to defraud BMC out of hundreds of millions of dollars to enrich itself justifies an award of exemplary damages that would be meaningful for a company of IBM's size and based on IBM's behavior.

III. Jurisdiction

133. The parties agree that the court has jurisdiction pursuant to 28 U.S.C. § 1332(a)(1), original jurisdiction pursuant to 28 U.S.C. § 1331, and supplemental jurisdiction pursuant to 28 U.S.C. § 1367. See Dkt. 612 at 22. The parties likewise agree that venue in this court is proper pursuant to 28 U.S.C. §§ 1391(a), (b), and (d). Id.

IV. Standard of Review

134. The parties agree that New York law governs the contract claims in this case and Texas law applies to BMC's common law and statutory trade secrets claims. Id. To prevail on its breach of contract claims, BMC must present evidence in support of the allegations set forth in the SAC and prove those allegations by a preponderance of the evidence. See Raymond v. Marks, 116 F.3d 466, 1 (2d Cir. 1997) (unpublished). “The burden of showing something by a preponderance of the evidence . . .simply requires the trier of fact to believe that the existence of a fact is more probable than its nonexistence.” Metro. Stevedore Co. v. Rambo, 521 U.S. 121, 137 n.9, 117 S.Ct. 1953 (1997) (internal quotation marks omitted). As the fact finder, the court is entitled to make credibility findings of the witnesses and testimony and to draw reasonable inferences from the evidence presented. See Merck Eprova AG v. Gnosis S.p.A., 901 F.Supp.2d 436, 448 (S.D.N.Y. 2012), aff'd, 760 F.3d 247 (2d Cir. 2014).

135. The preponderance standard also applies to BMC's fraudulent inducement and misuse of information claims. See Champion v. Robinson, 392 S.W.3d 118, 127 (Tex. App.- Texarkana 2012, pet. denied) (reviewing a fraudulent inducement claim under the preponderance standard); Malone v. PLH Grp., Inc., No. 01-19-00016-CV, 2020 WL 1680058, at *5 (Tex. App.- Houston [1st Dist.] Apr. 7, 2020, pet. denied) (reviewing trade secret claims under the preponderance standard). However, the court may award punitive damages only if BMC proves by clear and convincing evidence that the harm at issue resulted from fraud. See Tex. Civ. Prac. & Rem.Code Ann. § 41.003(a) (Supp.).

V. Conclusions of Law

136. Based on the factual findings, the court concludes that BMC failed to prove its breach of contract claim involving section 8 of the MLA. BMC failed to show by the preponderance that the information IBM purportedly misused fell within the contract's definition of confidential information and that, even if it did, IBM misused the information. For similar reasons, the court also concludes that BMC failed to prove its DTSA, TUTSA, and common law unfair competition claims by the preponderance.

137. Turning next to BMC's breach of contract claims involving the 2015 OA, the court concludes that BMC failed to demonstrate that IBM breached section 5.1. Though IBM benefitted from Project Swallowtail, BMC did not prove that IBM used BMC's licenses for a purpose other than to help AT&T complete its software migration. The court previously determined at summary judgment that IBM breached section 5.4. Dkt. 586. Now, the court concludes that IBM's breach resulted in direct damages, in the form of unpaid license fees, under the contract's express terms. BMC also proved that it substantially performed under the contract and persuasively argued that section 5.4 is not an unenforceable restrictive covenant. Accordingly, the court finds that BMC is entitled to direct damages for the licensing rights that IBM used but for which it did not pay. However, the court concludes that BMC's consequential damages model is too speculative. For that reason, BMC is unable to recover lost profits associated with AT&T's migration to IBM's software.

138. The court further concludes that BMC proved by clear and convincing evidence that IBM fraudulently induced it into signing the 2015 OA. As a result of IBM's fraudulent conduct, the court finds that the MLA's damages disclaimers are unenforceable and that BMC is entitled to punitive damages.

A. BMC's Contract Claim Involving Section 8 of the MLA

139. BMC claims that IBM violated section 8 of the MLA, which governs IBM's use of confidential information. The parties agree that New York law requires the following four elements to sustain a breach of contract claim: (1) the existence of a contract; (2) the plaintiff's performance pursuant to the contract; (3) the defendant's breach; and (4) damages resulting from, or caused by, that breach. Riccio v. Genworth Fin., 124 N.Y.S.3d 370, 372, 184 A.D.3d 590 (N.Y.App.Div. 2020); Canzona v. Atanasio, 989 N.Y.S.2d 44, 47, 118 A.D.3d 837 (N.Y.App.Div. 2014); see also Dkt. 612 at 15. The court concludes that BMC failed to prove that IBM breached section 8.

140. Per the MLA, “Confidential Information does not include information that [IBM] can show . . .is or becomes a matter of public knowledge through no fault of [IBM].” Findings of Fact, supra ¶ 20; PX1 at IBM00035620. At trial, BMC pointed to specific instances where it said IBM violated this confidentiality restriction. But BMC failed to show that this information was confidential under the contract's own definitions or that IBM misused it. Findings of Fact, supra ¶¶ 85-110. BMC's expert did not identify which information was public and which information was confidential. See id.; Trial Tr. 202:9-16, Mar. 16, 2022 (Román). On the contrary, IBM demonstrated that much of the information it allegedly misused was publicly available. Findings of Fact, supra ¶¶ 85-110. Accordingly, BMC did not present legally sufficient evidence that IBM contravened MLA section 8 through its use or disclosure of any confidential BMC information.

141. Therefore, the court DISMISSES BMC's breach of contract claim as to MLA section 8 with prejudice.

B. BMC's DTSA, TUTSA, and Unfair Competition Claims 1.DTSA and TUTSA Claims

142. The parties agree that the elements of a claim for trade secret misappropriation under TUTSA and DTSA are: (1) the ownership of a trade secret; (2) the misappropriation of a trade secret; and (3) an injury to the plaintiff or unjust enrichment to the defendant. See Morris-Shea Bridge Co. v. Cajun Indus., LLC, No. 3:20-cv-00342, 2021 WL 4084516, at *6-7 (S.D. Tex. Feb. 22, 2021); Silverthorne Seismic, LLC v. Sterling Seismic Servs., Ltd., No. H-20-2543, 2021 WL 4710813, at *4 (S.D. Tex. Oct. 7, 2021) (Miller, J.); see also Tex. Civ. Prac. & Rem. Code § 134A.004; 18 U.S.C. § 1836; Dkt. 612 at 23.

143. A claim under DTSA also requires a showing that the trade secrets were used in interstate commerce. 18 U.S.C. § 1836(b)(1); Silverthorne, 2021 WL 4710813, at *4. “Both the DTSA and TUTSA define ‘trade secret' to include scientific and technical information, such as ‘any formula [or] design,' that (1) the owner has taken reasonable measures to keep secret and (2) derives independent economic value from not being generally known or readily ascertainable through proper means.” DBG Grp. Invs., LLC v. Puradigm, LLC, No. 3:21-CV-678-S, 2022 WL 313435, at *3 (N.D. Tex. Feb. 2, 2022) (citing 18 U.S.C. § 1839(3); Tex. Civ. Prac. & Rem. Code § 134A.002(6)).

144. BMC did not present legally sufficient evidence that IBM misappropriated a BMC-owned trade secret. See DBG Grp. Invs., 2022 WL 313435, at *3. Specifically, BMC failed to demonstrate that any of the five technical support cases at issue contains trade secrets because: (a) each only contains customer support provided by low-level BMC help desk employees about off-the-shelf software products; and (b) the information at issue was publicly available.

145. “‘[M]isappropriation' under both statutes includes (1) ‘acquisition of a trade secret of another by a person who knows or has reason to know that the trade secret was acquired by improper means,' and (2) ‘disclosure or use of a trade secret of another without express or implied consent by a person who used improper means to acquire knowledge of the trade secret,' or had reason to know that the trade secret was derived using improper means.” Id. (citing 18 U.S.C. § 1839(5); Tex. Civ. Prac. & Rem. Code § 134A.002(3)).

146. BMC's expert did not testify to any trade secrets. Findings of Fact, supra ¶ 94. Nor did BMC present legally sufficient evidence that it owned a trade secret. Id. ¶¶ 97-122. But even assuming BMC showed that it owned a trade secret, BMC did not present legally sufficient evidence that IBM misappropriated that trade secret. For example, IBM used BMC customer support for which AT&T paid $9.5 million annually to keep AT&T's mainframe environment running properly while AT&T continued to use BMC software to help run its business. Sections 3 and 3.1 of the 2015 OA authorized IBM employees and third-party contractors to use BMC customer support information. Findings of Fact, supra ¶¶ 25-26.

147. In addition, BMC did not present legally sufficient evidence that it sustained any recoverable damages from any misappropriation of a trade secret by IBM, in part because its damages model did not quantify the value of any alleged trade secret. See Select Interior Concepts, Inc. v. Pental, No. 3:20-CV-295-L, 2021 WL 961690, at *3 (N.D. Tex. Mar. 14, 2021); Tex. Civ. Prac. & Rem. Code § 134A.004; 18 U.S.C. § 1836(b)(1).

148. Therefore, BMC did not prove its claims under DTSA and TUTSA, and the court DISMISSES those claims with prejudice.

2. BMC's Common Law Unfair Competition by Misappropriation Claim

149. BMC claimed that IBM “solicited confidential and trade secret information from BMC to facilitate the replacement of BMC products, ” thereby engaging in unfair competition via misappropriation. See Dkt. 295 at 37-38, ¶ 148. This claim likewise fails.

150. Absent a contractual commitment, Texas law recognizes a claim for “misappropriation” of confidential information only where that information is “either secret or, at least substantially secret.” Stewart & Stevenson Servs., Inc. v. Serv-Tech, Inc., 879 S.W.2d 89, 99 (Tex. App.-Houston [14th Dist.] 1994, writ denied); SP Midtown, Ltd. v. Urban Storage, L.P., No. 14-07-00717-CV, 2008 WL 1991747, at *5 n.5 (Tex. App.-Houston [14th Dist.] May 8, 2008, pet. denied) (mem. op.).

151. BMC did not present legally sufficient evidence that any information in the five technical support cases was actually “secret.” See Findings of Fact, supra ¶¶ 85-110. Even if BMC could show that any information in the five technical support cases was “secret, ” for the same reasons set forth above, BMC did not present legally sufficient evidence to show that IBM misappropriated any confidential information of BMC contained in the technical support cases.

152. Therefore, BMC did not prove its common law misappropriation claim, and the court DISMISSES the claim with prejudice.

To summarize, BMC failed to prove by the preponderance of the evidence that IBM breached section 8 of the MLA or misappropriated trade secrets under DTSA, TUTSA, or unfair competition. The court DISMISSES those claims with prejudice. Finally, the court concludes that BMC did not bring its trade secrets claims in bad faith.

C. BMC's Contract Claims Involving the 2015 OA

153. BMC claims that IBM breached sections 5.1 and 5.4 of the 2015 OA. See Dkt. 295 at 23-27, ¶¶ 71-89. New York law requires the following four elements to prove a breach of contract claim: (1) the existence of a contract; (2) the plaintiff's performance pursuant to the contract; (3) the defendant's breach; and (4) damages resulting from, or caused by, that breach. Riccio, 989 N.Y.S.2d at 47 . While BMC failed to prove that IBM breached section 5.1, the court concludes that BMC satisfied each element for its breach of section 5.4 claim.

1. BMC Failed to Prove that IBM Breached Section 5.1

154. BMC argued that IBM breached section 5.1 of the 2015 OA by using its products to effectuate the displacement at AT&T.

Section 5.1 provides:

BMC will allow [IBM] to use, access, install, and have operational responsibility of the BMC Customer Licenses (together, “Access and Use”) under the terms of the BMC Customer's license agreement with BMC for no fee, including on Computers owned or leased by BMC Customer and BMC Customer's facility, provided that the BMC Customer Licenses are used solely for the purposes of supporting the BMC Customer who owns such licenses.

PX4 § 5.1. Thus, to establish a breach of OA section 5.1, BMC had the burden to prove that IBM used BMC products at AT&T in a manner that was not “for the sole purpose of supporting AT&T.” Dkt. 603 at 9 (quotations, brackets and internal citations omitted); see also PX4 § 5.1.

155. BMC did not meet its burden of proving that IBM's use of the licenses was for any other purpose than to support AT&T in its Project Swallowtail efforts. To be sure, IBM benefitted from Project Swallowtail and, by proxy, derived ancillary benefits from the using the AT&T BMC licenses. See Findings of Fact, supra ¶¶ 32, 48-54. And BMC put forth ample evidence that IBM wanted to participate in Project Swallowtail. Id. But whether IBM benefitted from, or wanted, Project Swallowtail is not contractually relevant: the contract did not condition IBM's “no fee” use of the licenses on it not benefitting from the use of the licenses. See PX4 § 5.1. The record showed that IBM used the licenses to achieve what AT&T wanted done-and nothing more.

156. Accordingly, the court DISMISSES BMC's breach of contract claim as to section 5.1 with prejudice.

2. BMC's Non-Performance Under Section 5.1 Was Immaterial

157. IBM argued that it cannot be held liable for any breach because BMC failed to perform its own obligations under the 2015 OA, namely disclosing material differences between the AT&T customer license agreement and its standard EULA agreement. See Dkt. 588 at 5-6; Findings of Fact, supra ¶ 84. The parties agree that “[u]nder New York law, a party's performance under a contract is excused where the other party has substantially failed to perform its side of the bargain or, synonymously, where that party has committed a material breach.” Merrill Lynch & Co. v. Allegheny Energy, Inc., 500 F.3d 171, 186 (2d Cir. 2007); see Restatement (Second) of Contracts § 237 (1981) (“material failure”). It is a “fundamental principle of contract law” that “the material breach of a contract by one party discharges the contractual obligations of the non-breaching party.” Bear, Stearns Funding, Inc. v. Interface Grp.-Nevada, Inc., 361 F.Supp.2d 283, 291 (S.D.N.Y. 2005); see Medinol Ltd. v. Bos. Sci. Corp., 346 F.Supp.2d 575, 618 (S.D.N.Y. 2004) (“As a general principle of contract law, a material breach excuses the other party's nonperformance.”) (collecting cases); see Dkt. 612 at 22-23.

158. A breach is material under New York law if it “go[es] to the root of the agreement between the parties.” Frank Felix Assocs., Ltd. v. Austin Drugs, Inc., 111 F.3d 284, 289 (2d Cir. 1997) (internal quotation omitted). A party's contractual obligation to perform will be excused if “the other party's breach of the contract is so substantial that it defeats the object of the parties in making the contract.” Id. at 289; see Restatement (Second) of Contracts § 241 (1981) (listing circumstances that are significant for determining the materiality of a breach, including “the extent to which the injured party will be deprived of the benefit which he reasonably expected” and “the extent to which the behavior of the party failing to perform or to offer to perform comports with standards of good faith and fair dealing”).

159. “There is no simple test for determining whether substantial performance has been rendered and several factors must be considered.” Hadden v. Consol. Edison Co. of N.Y., 312 N.E.2d 445, 449, 34 N.Y.2d 88, 356 N.Y.S.2d 249 (N.Y. 1974). These factors include the ratio of the performance already rendered to that unperformed, the quantitative character of the default, the degree to which the purpose behind the contract has been frustrated, the willfulness of the default, and the extent to which the aggrieved party has already received the substantial benefit of the promised performance. Id.; see also Wechsler v. Hunt Health Sys., Ltd., 330 F.Supp.2d 383, 421-22 (S.D.N.Y. 2004); id. at 414, 419, 422 (breach is material only if it goes “to the root of the agreement between the parties, ” and alleged non-material breach “does not show that [plaintiff] did not substantially perform its contractual obligations”).

160. BMC did not disclose AT&T's customer license to IBM. Findings of Fact, supra ¶ 84. However, based on the facts presented, the court finds that BMC substantially complied with the 2015 OA because there are no “material differences” between the standard BMC EULA and the BMC-AT&T license agreement that section 5.1 required it to disclose. See Findings of Fact, supra ¶¶ 28-33. For example, IBM could not act as AT&T's outsourcer without BMC's consent under either agreement, and BMC could (and indeed, did) reasonably condition its consent on IBM agreeing to the same non-displacement protection as in the OA. See Id. The AT&T-BMC license agreement also would limit the outsourcing to only providing “data processing services” and acting for the “sole benefit” of AT&T. See id.

161. Most importantly, the OA represents the “single point of control” between IBM and BMC and does not afford IBM any “opt-out” rights that would enable it to operate directly under AT&T's license. Findings of Fact, supra ¶ 23; PX4. It further states that the OA provides “the terms under which BMC grants to [IBM] the right to use the [BMC] products in its IT Services business.” Findings of Fact, supra ¶ 23; PX4. Similarly, the OA states that it is the document that “sets forth [IBM's] rights” regarding “Access and Use” of BMC customer licenses for IT Services. Findings of Fact, supra ¶ 23; PX4. The parties agreed in the OA that “[i]t is the intent of the parties to operate under a common set of terms and conditions on a worldwide basis that addresses both [IBM's] and BMC's interests.” Findings of Fact, supra ¶ 27; PX4. Section 5 likewise provides that the OA controls IBM's provision of outsourcing services, regardless of any rights available under other contracts. Findings of Fact, supra ¶ 27; PX4 (“Section 5 “will apply to [IBM]'s access and use of the BMC Customer Licenses and apply retroactively to the point of first access by [IBM].”). No matter what terms exist in BMC's agreement with a third-party customer, nothing in the OA qualifies its applicability such that IBM could operate outside of its limits.

162. Thus, IBM's argument that would have “chosen to proceed directly under AT&T's license, rather than electing Access and Use through the OA” presents epistemological difficulties. See Dkts. 428 at 31-32, 597 at 8. BMC's disclosure obligation did not arise until after the parties executed the OA, and once IBM signed the OA, it could not proceed exclusively under AT&T's own rights. To the extent IBM suggests that BMC breached the 2013 OA's EULA provision, IBM has not explained how BMC's alleged non-compliance with a past contract is relevant, given that BMC asserted breach of only the 2015 OA. Regardless, any such argument with respect to the 2013 OA would fail for the same reasons described for the 2015 OA. Therefore, the obligation to inform IBM of any “material differences” never arose, disposing of this “EULA” argument as a factual matter.

163. Even if IBM could proceed directly under AT&T's license rather than abiding by the 2015 OA-and, as discussed above, it could not-the AT&T-BMC license agreement still would not have permitted IBM's Swallowtail actions, rendering any non-disclosure about that agreement legally immaterial in this case. Under section 3.3 of the AT&T-BMC license agreement, outsourcers were not permitted to use BMC products unless BMC gave its written consent, and BMC had the right to “reasonably” withhold that consent. PX6. Given IBM's status as a mainframe software competitor and the importance of AT&T as a customer, it would hardly be unreasonable for BMC to withhold its consent to IBM serving as AT&T's outsourcer unless IBM agreed to protections for BMC similar to OA section 5.4. See also Trial Tr. 203:13-204:2, 205:4-206:3, Mar. 14, 2022 (Ah Chu). For this additional reason, the AT&T-BMC agreement would not have allowed IBM's planned Project Swallowtail activities, which means this alleged EULA non-disclosure is legally immaterial to this case involving BMC's claim for breach of the OA. Accordingly, for both factual and legal reasons, the court concludes that, to the extent BMC breached, its nonperformance was not material and, therefore, it substantially performed under the contract. See Merrill Lynch, 500 F.3d at 187.

3. IBM's Breach of Section 5.4 Caused BMC's Damages

164. The court already determined that IBM breached section 5.4 of the 2015 OA by displacing BMC's software products with IBM's own at AT&T. See Dkt. 586. In light of the court's finding that BMC substantially performed under the contract, the court must now determine whether IBM's breach of section 5.4 damaged BMC. Looking to the MLA and 2015 OA's express, plain language, the court concludes that IBM's breach directly damaged BMC.

165. Damages are generally bifurcated into two categories: direct and consequential. BMC seeks both in connection with IBM's breach of section 5.4. Under either, BMC must show that IBM's breach caused its damages. As the Court of Appeals of New York explained, “[i]t is axiomatic that damages for breach of contract are not recoverable where they were not actually caused by the breach.” Pesa v. Yoma Dev. Grp., Inc., 965 N.E.2d 228, 230, 18 N.Y.3d 527, 942 N.Y.S.2d 1 (N.Y. 2012).³⁴ The court concludes that BMC established that its license fees represent the direct damages expressly contemplated under the contract. As a result, BMC may recover $717,739,615.00 in unpaid license fees. With respect to BMC's lost profits model, however, the court concludes that BMC failed to prove that IBM's breach of section 5.4 through its participation in Project Swallowtail caused it to lose profits from AT&T. Accordingly, BMC recovers nothing on its lost profits claim.

34 Because damages must “be directly traceable to the breach, ” IBM was permitted to produce evidence disputing the causal nexus between its breach and BMC's harm. See Nat'l Mkt. Share, Inc. v. Sterling Nat'l Bank, 392 F.3d 520, 526 (2d Cir. 2004). See also Dkt. 603 at 4.

a. BMC's Damages are Direct Damages Under the Contract's Express Terms

166. A buyer seeks direct damages “when he tries to recover the value of the very performance promised.” Schonfeld v. Hilliard, 218 F.3d 164, 175 (2d Cir. 2000) (internal quotations omitted); Versatile Housewares & Gardening Sys., Inc. v. Thill Logistics, Inc., 819 F.Supp.2d 230, 237 (S.D.N.Y. 2011). “Direct damages are typically expectation damages, measured by what it would take to put the non-breaching party in the same position that it would be in had the breaching party performed as promised under the contract.” Latham Land I, LLC v. TGI Friday's, Inc., 948 N.Y.S.2d 147, 151-52, 96 A.D.3d 1327 (N.Y.App.Div. 2012) (where the plaintiff sought to recover “the loss of the benefit of the bargain, ” the damages sought were not barred by a contractual provision excluding consequential damages); see also Bi-Econ. Mkt., Inc. v. Harleysville Ins. Co. of N.Y., 886 N.E.2d 127, 130, 10 N.Y.3d 187, 856 N.Y.S.2d 505 (N.Y. 2008) (“It is well settled that in breach of contract actions ‘the nonbreaching party may recover general damages which are the natural and probable consequence of the breach.'”); In re CCT Commc'ns, Inc., 464 B.R. 97, 116 (Bankr. S.D.N.Y. 2011) (direct damages “provide the aggrieved party with the difference between the price he agreed to pay and the value he was to receive through performance”); Kenford Co. v. County of Erie, 537 N.E.2d 176, 178, 73 N.Y.2d 312, 540 N.Y.S.2d 1 (N.Y. 1989) (direct damages are those that “are the natural and probable consequence of the breach”).

167. IBM argued that BMC cannot prove causation because “[n]on-breach of OA section 5.4 would have meant that IBM did not execute Project Swallowtail, not that IBM would have purchased hundreds of millions of BMC software licenses.” Dkt. 735 ¶ 431. Instead, IBM claimed that BMC's “license fee” damage model is “based on a hypothetical and speculative transaction in which BMC asserts that IBM would have agreed to pay more than $700 million to buy its own BMC software licenses.” Id. ¶ 433.

168. The problem for IBM, however, is that it agreed to do just that. The MLA is clear: “[e]ach party's liability arising out of or related to this agreement, the product, or the use of the product shall be limited to the greater of $5,000,000 or the amount paid or payable by customer for the license to the applicable product giving rise to the claim.”³⁵ PX1 at IBM00035620. “Payable” refers to the “sum of money . . . that is to be paid.” Payable, Black's Law Dictionary (11th ed. 2019); Payable, Merriam-Webster Dictionary, https://www.merriam-webster.com/dictionary/payable (last visited May 9, 2022) (the amount that “must be paid”); see also Carr v. Maryland Cas. Co., 88 Misc.2d 424, 427, 388 N.Y.S.2d 196 (N.Y. Civ. Ct. 1976) (finding “paid or payable” language unambiguous). The court concludes that IBM's liability is limited to the greater of $5 million or the amount [that must be paid] for the license to the applicable product giving rise to the claim. See PX1 at IBM00035620.

35 Section 10 of the MLA does not contain a liquidated damages provision and the “paid or payable” language does not reflect a contractual penalty.

169. IBM displaced fourteen BMC products with respect to BMC's section 5.4 claim. The 2015 OA provided only one avenue through which IBM could exercise the displacement rights it did in performing Project Swallowtail. See Id. ¶¶ 24, 29. To secure those rights, IBM would have had to purchase the licenses and related support under section 8 and Exhibit H, which contained agreed-upon fee amounts for each specific product, including pre-negotiated minimum discounts. See Id. ¶ 29; PX4 §8.1.

170. Notwithstanding this textual support for BMC's license fee model, IBM appeared to argue that the 2015 OA's “minimum discount” language renders the contract materially incomplete. See Trianco, LLC v. IBM, 466 F.Supp.2d 600, 605 (E.D. Pa. 2006) (“Every material term must be agreed upon in order to form a binding contract, and the court will not enforce a contract if price negotiations are left to a future date.”), aff'd in part, vacated in part, 271 Fed.Appx. 198 (3d Cir. 2008); see also Major League Baseball Props., Inc. v. Opening Day Prod., Inc., 385 F.Supp.2d 256, 271 (S.D.N.Y. 2005) (“Price or compensation are material terms in a contract requiring definiteness.”).

171. In Trianco, IBM and a subcontractor entered into an agreement wherein the subcontractor would assist IBM with preparing a government bid. 466 F.Supp.2d at 603. Their agreement did not include a specific price term, though it did provide a method to determine the price ceiling: the subcontractor would “supply the equipment and/or services . . . at prices that do not exceed” a set amount. Id. Only after IBM won the contract were the parties to identify a fixed price. See Id. Looking to the contract's plain language, the district court found that the parties failed to negotiate a subcontract price. Id. at 606.

172. IBM also relies on United Press v. New York Press Company, a turn-of-the-century Court of Appeals of New York case, for the proposition that the 2015 OA failed to set a contract price as a matter of law. See 56 N.E. 527, 164 N.Y. 406 (N.Y. Ct. App. 1900); Dkt. 735 ¶ 434. In that case, the defendant agreed to “pay to [the plaintiff] . . . a sum not exceeding three hundred dollars during each and every week” that the plaintiff delivered a daily news report “without interruption from and after the first day of January, in the year 1900.” United Press, 164 N.Y. at 408 (emphasis added). For some time, the defendant paid United Press $300 per week for the news reports. Id. at 409. But when the defendant sent United Press written notice that it could not continue paying the $300 sum, United Press filed suit and demanded judgment for damages in the sum of $93,000, based on its assertion that it was owed $300 a week from January 1, 1894, through January 1, 1900. Id. The trial court ruled in favor of United Press; the Court of Appeals of New York reversed because the contracted-for $300 limit set a price ceiling that the parties could not exceed, not a fixed, definite rate of compensation. Id. at 411. Because the contract was “silent as to the price which is to be paid to the plaintiff during its term, ” the court concluded it did not possess “binding force.” Id. at 413-14.

173. But unlike Trianco and United Press, the pre-negotiated minimum discounts in this case do not represent a “starting point” for price negotiations between the parties. See Dkt. 735 ¶ 434. Rather, they reflect definite, bargained for prices: “[IBM] will be entitled to a global minimum discount of 72% off the listed price in Exhibit H for all purchases of new licenses.” PX4 § 8.1. The price IBM must pay for the licenses is thus easy to discern. See Id. Had IBM sought to purchase the BMC licenses, it might have successfully negotiated a steeper discount. Indeed, sophisticated business entities often bargain for more beneficial prices than those set by contract. But IBM did not attempt to purchase the licenses, and it never sought to negotiate a steeper discount. So, in measuring BMC's direct damages, this court declines IBM's invitation to go beyond the contract's text and imagine a scenario in which IBM negotiated a steeper discount for BMC's licenses while displacing BMC's products at AT&T.

174. In addition to being factually distinguishable from both cases, the dictum in United Press is unhelpful to IBM's argument that it owed BMC nothing for exercising rights for which it did not pay. See 164 N.Y. at 411 (“Whether in all cases of an executory contract of purchase and sale, where the parties are altogether silent as to the price, the law will supply the want of any agreement as to price, by inferring that the parties must have intended to sell and to buy at a reasonable price, may be a question of some difficulty. Undoubtedly, the law makes that inference where the contract is executed by the acceptance of the goods by the defendant, in order to prevent the injustice of the defendant taking the goods without paying for them.”); id. at 412 (“[W]here work has been done, or articles have been furnished, a recovery may be based upon quantum meruit or quantum valebant.”); see also Quantum valebant, Black's Law Dictionary (11th ed. 2019) (“At common law, a count in an assumpsit action to recover payment for goods sold and delivered to another.”).

175. The license and support fees sought by BMC are the amounts IBM owes under the OA for the set of rights it utilized. Those fees are thus the “amount . . . payable” under MLA section 10 because they are what “must be paid” under OA section 8 and Exhibit H to obtain the rights IBM exercised. Thus, under the contract's plain text, the license fees represent direct damages that are “the natural and probable consequence of the breach.” See Bi-Econ., 886 N.E.2d at 130.

176. Therefore, the court finds that BMC is entitled to $717,739,615.00 in direct contract damages for IBM's breach of section 5.4 of the 2015 OA.³⁶

36 The parties agree that, under New York law, a plaintiff is entitled to recover prejudgment interest at the rate of 9% “as a matter of right” on a breach of contract claim. New England Ins. Co. v. Healthcare Underwriters Mut. Ins. Co., 352 F.3d 599, 602-03, 606 (2d Cir. 2003); N.Y. C.P.L.R. §§ 5001(a), 5004. Had BMC elected to recover only its contract damages, New York law would have entitled it to prejudgment interest at a rate of 9% from January 1, 2016, until the date of judgment.

b. BMC's Consequential Damages Model Is Too Speculative

177. BMC also seeks to recover $104.5 million in lost profits from its relationship with AT&T. See Trial Tr. 249:24-250:11, Mar. 16, 2022 (Ratliff). BMC argues that this figure represents the amount in fees that AT&T would have continued to pay BMC had IBM not “caused” AT&T to terminate its relationship with BMC. Dkt. 723 ¶ 153. To arrive at the $104 million figure, BMC's damages expert used a terminal value approach-also known as a perpetuity valuation model-to value what he had determined to be about $5.6 million in annual lost profits from the loss of the AT&T business. Trial Tr. 230:05-10, 235:03-09, Mar. 16, 2022 (Ratliff). Mr. Ratliff testified that he used a perpetuity approach because he thought there was “uncertainty” about what might have happened to BMC's mainframe business with AT&T “in the absence of the accused acts.” Id. 226:18-228:05. Basic arithmetic confirms that BMC's lost profits model assumes that AT&T would have stayed a BMC mainframe software customer for a total of nineteen years. See Id. 247:20-248:18 (Ratliff). BMC did not present an alternative model that measured how IBM's participation in Project Swallowtail may have accelerated the loss of BMC's mainframe revenues from AT&T. See Id. 245:11-15 (Ratliff); see also Trial Tr. Mar. 16, 2022 154:01-20 (Román).

178. A plaintiff need only demonstrate “a stable foundation for a reasonable estimate of the damage incurred as a result of the breach.” Tractebel Energy Mktg., Inc. v. AEP Power Mktg., Inc., 487 F.3d 89, 110-11 (2d Cir. 2007) (citation and internal quotation marks omitted). Damages, however, “must be not merely speculative, possible, and imaginary.” Id. Rather, a plaintiff must prove that they are a “reasonably certain” consequence of the defendant's breach. Id.; see also E.J. Brooks Co. v. Cambridge Sec. Seals, 105 N.E.3d 301, 307, 80 N.Y.S.3d 162 (N.Y. 2018) (explaining that lost future profits that are “remote, contingent or speculative” and are not “reasonably traced to the event” cannot satisfy the requisite element of causation); Kenford Co. v. County of Erie., 67 N.Y.2d 257, 261, 493 N.E.2d 234, 502 N.Y.S.2d 131 (N.Y. 1986) (“[I]t must be demonstrated with certainty that such damages have been caused by the breach and, second, the alleged loss must be capable of proof with reasonable certainty. In other words, the damages may not be merely speculative, possible or imaginary, but must be reasonably certain and directly traceable to the breach, not remote or the result of other intervening causes.”).

179. BMC did not present legally sufficient evidence of causation of damages under a “lost profits” theory because it failed to show that AT&T would not have removed BMC's software even if IBM had complied with OA section 5.4, and thus any breach did not “directly and proximately cause[]” BMC's damages. Dkt. 603 at 4 (quoting Nat'l Mkt. Share, Inc. v. Sterling Nat'l Bank, 392 F.3d 520, 525 (2d Cir. 2004)). BMC's lost profits damages model assumes that BMC would have retained AT&T as a mainframe customer for nineteen years-and at the same level of revenue and profitability-if IBM had declined to assist AT&T with Project Swallowtail.

180. But the factual record belies this assumption. AT&T initiated Project Swallowtail in April 2013 to migrate away from BMC's products because its use of BMC's software inhibited its larger standardization goals. Findings of Fact, supra ¶¶ 44-55. As BMC has acknowledged, AT&T could have accomplished these strategic objectives by having a different company perform the displacement. See Dkt. 454 at 20; Dkt. 416 at 49 (“[O]ther outsourcers could have performed the displacement for AT&T; in fact, six such outsourcers did perform parts of the displacement.”). Accordingly, BMC recovers nothing on its lost profits claim.

4. Section 5.4 is not an Unenforceable Restrictive Covenant

181. IBM pleaded an affirmative defense that section 5.4 of the 2015 OA is an unenforceable restrictive covenant. See Dkts. 299 at 22, 694 at 24-25. The court adopted Judge Bryan's conclusion that BMC's interpretation of section 5.4 did not render it an unenforceable restrictive covenant. See Dkt. 561 at 12-13. Now that IBM's larger unenforceability argument is squarely before it, the court concludes that section 5.4 is not an unenforceable restrictive covenant.

182. Under New York law, non-competition restraints are reasonable only if “(1) no greater than is required for the protection of the legitimate interest of the [business], (2) does not impose undue hardship on the [subject], and (3) is not injurious to the public.” BDO Seidman v. Hirshberg, 712 N.E.2d 1220, 1223, 93 N.Y.2d 382 (N.Y. 1999). However, while courts “rigorously examine[]” these factors in the context of restrictive employment agreements, see Am. Inst. of Chem. Eng'rs v. Reber-Friel Co., 682 F.2d 382, 387 (2d Cir. 1982), “restrictive covenant[s] in an ordinary commercial contract” receive lighter scrutiny. See Calico Cottage, Inc. v. TNB, Inc., No. 11-CV-0336 (DLI) (MDG), 2014 WL 4828774, at *5 (E.D.N.Y. Sept. 29, 2014). That is because, generally, “where two business entities agree to a restrictive covenant, there is . . . no concern about the loss of [an] individual's livelihood or an imbalance of bargaining power.” Id.; see also Penske Media Corp. v. Shutterstock, Inc., 548 F.Supp.3d 370, 378 (S.D.N.Y. 2021) (“[C]ompetition concerns, if they are relevant at all, are minimized in the context of a contract between two sophisticated business parties.”). Three factors govern the court's assessment of contractual restraints agreed to between businesses: (1) the presence of a legitimate business interest; (2) the reasonableness of the restriction's scope; and (3) the hardship on the restricted party. See Calico Cottage, Inc., 2014 WL 4828774, at *5.

183. First, IBM's insistence that section 5.4 only protects BMC's revenue is misplaced. BMC demonstrated that section 5.4's displacement prohibition was designed to prevent giving IBM an unfair advantage when competing with BMC in the software business-a clearly legitimate business interest. IBM directly competes against BMC as a mainframe software developer and vendor. Findings of Fact, supra ¶¶ 6-9. But as an IT outsourcer, IBM works with clients that use BMC's mainframe software-and not IBM's-to keep their computer systems up and running. Id. By wearing both hats, IBM can acquire unique knowledge about how a competitor's software operates on a mutual customer's mainframe system. Id. Indeed, the trial record showed that, as AT&T's IT outsourcer, IBM could contact BMC's support services to diagnose software problems, gleaning first-hand insight into how BMC's product functioned in the AT&T environment. Id. ¶¶ 12, 15-16. To perform as an IT outsourcer, IBM needed “a contractual vehicle . . . to conduct . . . business” with “BMC and any clients [in] an outsourcing environment.” Id. ¶ 14; Craigsop Dep. 26:01-07. BMC provided IBM that ability under the 2015 OA's “no fee” Access and Use option on the condition that it did not then displace BMC's products with its own products. Findings of Fact ¶ 10.; Schulman Dep. 145:10-147:5 (explaining that the parties' agreements prevented IBM from “leveraging its position and access to displace BMC”). Both IBM and BMC are sophisticated business entities “capable of understanding the terms and implications of” section 5.4's non-displacement provision “and making an informed decision as to its risks and benefits.” See Nat'l Elevator Cab & Door Corp. v. H & B, Inc., No. 07CV1562(ERK)(RML), 2008 WL 207843, at *9 (E.D.N.Y. Jan. 24, 2008), order aff'd and remanded sub nom, 282 Fed.Appx. 885 (2d Cir. 2008); see also Findings of Fact, supra ¶ 5. Second, section 5.4 is reasonable in scope. The non-displacement provision applies only to the fifty-four customers listed in Exhibit K. Of those fifty-four customers, IBM serves as an IT outsourcer to eighteen. Thus, section 5.4 effectively is limited to eighteen mutual customers. Moreover, the 2015 OA was of a limited duration, governing the parties' relationship from September 30, 2015, to March 27, 2018.

184. Third, section 5.4 does not impose an undue hardship on IBM, which consented to the provision following intensive, sophisticated negotiations held at arm's length. The record demonstrates that IBM secured valuable benefits by assenting to the non-displacement provision, undercutting its hardship appeals. See Findings of Fact, supra ¶¶ 56-66.

To summarize, BMC did not prove by the preponderance that IBM breached section 5.1 of the 2015 OA. To the extent that BMC did not perform under section 5.1 of the 2015 OA, its nonperformance was immaterial. IBM breached section 5.4 when it displaced fourteen BMC products with its own at AT&T. The MLA expressly contemplates the payment of licensee fees in the event of liability. Section 5.4 is not an unenforceable restrictive covenant that would otherwise annul IBM's liability for its breach. Accordingly, BMC may recover $717,739,615.00 in direct damages in the form of the applicable license fees and associated support services for IBM's breach.

D. BMC Proved that IBM Fraudulently Induced it into Signing the 2015 OA

185. “Texas law has long imposed a duty to abstain from inducing another to enter into a contract through the use of fraudulent misrepresentations.” Formosa Plastics Corp. USA v. Presidio Eng'rs & Contractors, Inc., 960 S.W.2d 41, 46 (Tex. 1998). The court concludes that IBM breached that duty with respect to the 2015 OA.

186. Fraudulent inducement “arises only in the context of a contract.” Anderson v. Durant, 550 S.W.3d 605, 614 (Tex. 2018). “As a general rule, the failure to perform the terms of a contract is a breach of contract, not a tort.” Crim Truck & Tractor Co. v. Navistar Int'l Transp. Corp., 823 S.W.2d 591, 597 (Tex. 1992). However, “when one party enters into a contract with no intention of performing, that misrepresentation may give rise to an action in fraud.” Id.

Under Texas law, a fraudulent inducement claim requires proof that:

(1) the defendant made a material misrepresentation; (2) the defendant knew at the time that the representation was false or lacked knowledge of its truth; (3) the defendant intended that the plaintiff should rely or act on the misrepresentation; (4) the plaintiff relied on the misrepresentation; and (5) the plaintiff's reliance on the misrepresentation caused injury.

Int'l Bus. Machs. Corp. v. Lufkin Indus., LLC, 573 S.W.3d 224, 228 (Tex. 2019). Where a plaintiff “presents legally sufficient evidence on each of the elements of a fraudulent inducement claim, any damages suffered as a result of the fraud sound in tort, ” including economic losses relating “to the subject matter of the contract.” Formosa, 960 S.W.2d at 47.

187. Though these basic elements are uncontested, see Dkt. 612 at 23, the parties differ in their understandings of what IBM's intentions should be measured against: the unambiguous terms of the contract or IBM's subjective interpretation of the contract. Compare Dkt. 598 at 16 (“[T]he Court should assess whether IBM intended to perform under the OA against the plain contractual meaning as written, not based on IBM's theory of an unwritten ‘compromise.'”), and Dkt. 668 at 10 (arguing that Texas caselaw holds that “a party cannot justifiably rely on statements made during highly contentious negotiations, especially statements that conflict with the ultimate, unambiguous language of a written contract”), with Dkt. 597 at 23 (“IBM repeatedly told BMC that it interpreted the non-displacement language as allowing IBM to carry out a customer-ordered replacement of BMC software. The court will need to decide whether those repeated statements preclude any fraud claim by disproving an intent to deceive.”), and Dkt. 696 (“BMC's cases stand for the uncontroversial proposition that a plaintiff cannot rely on [extracontractual communications] where reliance is later disclaimed or contradicted by the words of the contract.”).

188. The court concludes that IBM made (1) a material misrepresentation when it promised in writing not to displace BMC's products with its own that it (2) knew at the time was false, and (3) intended that BMC to rely on the misrepresentation. The court likewise concludes that BMC relied on IBM's written promise not to displace BMC's products with its own and that this reliance caused its injury insofar as its injury reflects direct contract damages. BMC shall recover direct and punitive damages from IBM on its fraudulent inducement claim, plus prejudgment interest in the amount of 5% and post-judgment interest.

1. Material Misrepresentation

189. IBM made written “material misrepresentations” when it promised not to displace BMC products with its own products at Exhibit K customers-including AT&T-while operating under the 2015 OA's “no fee” Access and Use option. See PX4. Section 5.4's non-displacement provision was a significant source of contention during the 2015 OA's negotiation, and BMC made clear that it would not remove AT&T from the list of customers on Exhibit K subject to it. See Findings of Fact, supra ¶¶ 56-66. In short, the negotiation history makes clear that BMC would not have entered the contract with IBM but for (1) the non-displacement provision and (2) its application to the AT&T account. See Id. Thus, IBM made a material misrepresentation when it agreed not to displace BMC's products with its own.

2. IBM Knew the Representation was False, Lacked an Intention to Perform, and Sought BMC's Reliance on Its Misrepresentation

190. The parties agree that “a fraud claim can be based on a promise made with no intention of performing, irrespective of whether the promise is later subsumed within a contract.” See Formosa Plastics, 960 S.W.2d at 46; Dkt. 612 at 23; see also Spoljaric v. Percival Tours, Inc., 708 S.W.2d 432, 434 (Tex. 1986) (“A promise to do an act in the future is actionable fraud when made with the intention, design and purpose of deceiving, and with no intention of performing the act.”). In determining IBM's intent, the court looks to “the time the party made the representation, ” though it may also consider IBM's “subsequent acts” to infer its intent. See Spoljaric, 708 S.W.2d at 434.

191. IBM asserts that it “did not intend to deceive BMC . . . and entered the 2015 OA under its mistaken belief that it was permitted to assist AT&T with Project Swallowtail.” Dkt. 696 at 13. To be sure, “the mere failure to perform a contract is not evidence of fraud, ” Formosa Plastics, 960 S.W.2d at 48, because a “[f]ailure to perform, standing alone, is no evidence of the promisors' intent not to perform when the promise was made.” Spoljaric, 708 S.W.2d at 435. But IBM's defense strains credulity.³⁷ Though only “[s]light circumstantial evidence of fraud . . . is sufficient to support a finding of fraudulent intent, ” id., the court concludes that BMC persuasively showed that IBM “made representations with the intent to deceive and with no intention” of complying with the 2015 OA's non-displacement provision “as represented” in the written contract. See Formosa Plastics, 960 S.W.2d at 48.

37 IBM argues that a court cannot measure its intent on the unambiguous contractual provision because, otherwise, “every finding that a contractual provision is unambiguous would automatically require a finding that the breaching party also intended to deceive-an absurd result.” Dkt. 696 at 12-13. IBM's doctrinal concern need not apply here because this court's measure of IBM's intent is not in the unambiguous contract language but the substantial factual record evidencing its knowledge of the non-displacement provision's plain meaning-and its intentional disregard of the language to which it agreed.

192. IBM knew that its representations were false and lacked an intention to perform under the written contract. IBM's lead negotiator testified that IBM “knew what [it was] signing.” Findings of Fact, supra ¶ 64, n. 21. The record convincingly reflects that IBM knew it was signing a limitation on its ability to displace BMC's products with its own. See, e.g., Findings of Fact, supra ¶ 35; PX460 (e-mail from Stafford explaining that the non-displacement language was “very clear” in limiting “what [IBM] [is] permitted or not permitted to do in situations where we take over a client's footprint”); Findings of Fact, supra ¶ 30; PX49 at IBM00078246 (IBM internal guidance on 2013 OA explaining that “[w]here displacement is desired by the Client, ” the “[c]ostly [a]pproach” is to “[n]egotiate with BMC to acquire equivalent BMC licenses to be held by IBM that do not include the restrictive language” contained in “client-licensed BMC products”); Findings of Fact, supra ¶¶ 38, 72; PX34 (Sweetman e-mail explaining that the 2013 language “bar[red] IBM from displacing the BMC products with IBM products for the duration of our services agreement with the client”); PX30 at 1 (IBM employee internally stating, prior to the 2013 OA's execution, that she “didn't think [IBM was] agreeing to another non-displacement clause - in fact I thought we agreed to negotiate out of that”).

193. IBM's insistence on removing AT&T from the list of accounts covered by the non-displacement provision underscores that it did not believe that the 2015 OA authorized a customer-directed displacement. See Findings of Fact, supra ¶ 62; Clyne Dep. 73:06-13 (discussing IBM's desire and negotiation tactics around removing AT&T from the list of protected accounts). Had IBM truly believed that customer-directed displacements were permissible under the contract, it likely would not have devoted considerable resources during the 2015 OA negotiations to try first remove the non-displacement language entirely and then, when that failed, remove AT&T from the list of protected accounts. See Findings of Fact, supra ¶ 58; PX156 at IBM00097161 (“IBM wants the non-displacement language in the current OA fully removed from our agreement with BMC.”); Clyne Dep. 61:17-20, 73:06-13.

194. IBM had no intention of complying with the 2015 OA's non-displacement provision because it had already agreed to displace BMC's products-at AT&T's behest-when it entered into the Project Swallowtail agreement. See Findings of Fact, supra ¶¶ 50-56. Instead, IBM believed-especially in light of BMC's reluctance to engage in litigation-that it could “always settle[] for a small percent of the claim” or for “pennies on the dollar.” Id. ¶ 50; PX109

at IBM00038867; see also Findings of Fact, supra ¶ 57; Clyne Dep. 55:13-19 (explaining that

BMC did not “want to have to spend any more attorney time” on displacement issues); Findings of Fact, supra ¶¶ 40, 75; PX37 at IBM00062682 (detailing internal IBM communications immediately following the 2013 OA's execution in which IBM is shown strategizing how it “may cho[o]se to manage the non-displacement language overall as a business” despite there being “no change on the non-displace”) (emphasis added); Findings of Fact, supra ¶ 40; PX38 (internal IBM

e-mail wherein IBM account executives discussed “put[ting] BMC to the test on the additional non-displace restrictions). Indeed, Clyne, testified:

Q. …IBM's plan was not to abide [by the non-displacement provision in the 2015 OA], wasn't it?

A. That was the position that legal had taken.

Findings of Fact, supra ¶ 64; Clyne Dep. 111:10-19; id. 113:11-21 (explaining that “IBM's conduct throughout the negotiations and the time spent on non-displacement claims by BMC” led Clyne to know that IBM did not intend to comply with the contract). Due to IBM's intention not to perform under the contract, Clyne did not believe that IBM negotiated the contract in good faith. Findings of Fact, supra ¶ 57; Clyne Dep. 117:19-118:12; see also Findings of Fact, supra ¶ 65; Clyne Dep. 103:10-20 (Clyne testifying that he did not dwell on compliance issues which would come “after the fact, after the agreement [was] signed” because he had “no control” over how IBM would perform under the contract.”). Instead, internal correspondence at IBM largely reflected (1) a desire to preclude compliance-related settlements altogether (understandably so, since compliance disputes can cut into profit margins) and (2) an awareness that agreeing to the non-displacement language's extension to AT&T would put it immediately out of compliance with the contract. See generally Findings of Fact, supra ¶¶ 43-67.

195. IBM's refusal to disclose why it sought AT&T's removal from the list of protected accounts, and its inclusion of other accounts in its removal request to obfuscate AT&T's importance, evidences its intention that BMC rely on that written promise instead of on IBM's prior communications of a contract interpretation that it, internally, did not fully believe. See, e.g., Findings of Fact, supra ¶¶ 72-83. And that reliance was important to IBM because the 2015 OA, in addition to offering it a full release from past non-compliance claims, gave IBM the critical shared hosting rights necessary to perform its job as a global outsourcer. See Findings of Fact, supra ¶ 57, n. 17; Clyne Dep. 230:05-15, 139:04-08 (testifying that IBM “[f]ar, far exceeded its goals” in securing valuable shared hosting rights); Calo Dep. 10:22-11:3, 55:9-14; PX106; PX108.

3. BMC Justifiably Relied on IBM's Promise Not to Displace BMC's Products with Its Own

196. “[F]raud does not exist unless the defendant's representations induced the plaintiff to take a particular course of action. It is not necessary that the representations were the sole inducement, but the representations relied upon must have been a material factor in inducing the plaintiff's action.” Coffel v. Stryker Corp., 284 F.3d 625, 636 (5th Cir. 2002) (applying Texas law). Thus, to prevail on its fraudulent inducement claim, BMC must show that “it actually relied on the defendant's representation and, also, that such reliance was justifiable.” See JPMorgan Chase Bank, N.A. v. Orca Assets G.P., LLC, 546 S.W.3d 648, 653 (Tex. 2018); see also Mercedes-Benz USA, LLC v. Carduco, Inc., 583 S.W.3d 553, 554 (Tex. 2019).

197. Generally, justifiable reliance is a question of fact. Orca Assets, 546 S.W.3d at 654. “But the element can be negated as a matter of law when circumstances exist under which reliance cannot be justified.” Id. In determining whether justifiable reliance is negated as a matter of law, courts “must consider the nature of the [parties'] relationship and the contract.” Id. (quoting AKB Hendrick, LP v. Musgrave Enters., Inc., 380 S.W.3d 221, 232 (Tex. App.-Dallas 2012, no pet.)). “In an arm's-length transaction[, ] the defrauded party must exercise ordinary care for the protection of his own interests. . . . [A] failure to exercise reasonable diligence is not excused by mere confidence in the honesty and integrity of the other party.” Nat'l Prop. Holdings, L.P. v. Westergren, 453 S.W.3d 419, 425 (Tex. 2015) (quoting Thigpen v. Locke, 363 S.W.2d 247, 251 (Tex. 1962)). And when a party fails to exercise such diligence, it is “charged with knowledge of all facts that would have been discovered by a reasonably prudent person similarly situated.” See AKB, 380 S.W.3d at 232. To this end, a plaintiff “cannot blindly rely on a representation by a defendant where the plaintiff's knowledge, experience, and background warrant investigation into any representations before the plaintiff acts in reliance upon those representations.” See Shafipour v. Rischon Dev. Corp., No. 11-13-00212-CV, 2015 WL 3454219, at *8 (Tex. App.-Eastland May 29, 2015, pet. denied) (mem. op.).

198. IBM urges the court to hold that BMC, as a matter of law, could not rely on its written promise not to displace when it had, for years, voiced its position that a customer-requested displacement would not violate section 5.4.³⁸ See Dkt. 696 at 5. In its telling, “[s]ince BMC entered the 2015 OA fully understanding IBM's interpretation and its intentions to act accordingly, BMC cannot now argue it was entitled to rely on the ‘words of the OA' as containing a contradictory unwritten promise.” Id. Or, put differently, IBM argues that BMC could not justifiably rely on the unambiguous written contract given its extracontractual statements.

38 IBM suggested that the court's interpretation of the non-displacement provision is irrelevant because the Court issued its ruling in 2021, years after the 2015 OA was executed. Dkt. 600 at 1, 5-6. IBM argued that, if the court's recent interpretation of the non-displacement provision were considered, it would turn BMC's claim into a “retroactive” fraud claim. Id. But the court's ruling merely set forth the unambiguous meaning of the plain text of section 5.4. That text was the same in 2015 as it is now. The court's interpretation, by simply stating what the plain terms of the 2015 OA mean, identifies what IBM promised to do at the time IBM entered into the contract in 2015. And under the applicable law, the relevant question is whether IBM intended to perform as represented in the contract. See Formosa Plastics, 960 S.W.2d at 48.

199. Notwithstanding IBM's protests, Texas law recognizes that the written, binding contractual language matters-and plaintiffs who refrain from relying on a written contract's unambiguous language do so at their own detriment. The Texas Supreme Court's decision in Orca Assets illustrates the point. There, the Court considered whether Orca-a lessee of certain mineral interests-justifiably relied on extra-contractual representations made by JPMorgan's agent regarding land available for lease, despite “red flags” appearing in the latter's proposed contracts. Orca Assets, 546 S.W.3d at 650-52. JPMorgan argued that the company's claims were negated as a matter of law because the company could not have justifiably relied on statements that the land was “open” for lease considering the number of red flags present. Id. at 654-55. The Texas Supreme Court agreed, concluding that a letter of intent, which placed the responsibility on Orca to investigate title and contained a negation-of-warranty provision, directly contradicted the representations on which Orca purportedly relied. Id. at 659. The Texas Supreme Court concluded that the written contract directly contradicted the agent's verbal representations. Id. Taken together with other red flags-and both parties' sophistication-the court concluded that this direct contradiction precluded Orca's justifiable reliance on the alleged misrepresentation.³⁹ Id. at 660.

39 With regards to the seven “red flags” JPMorgan identified, the Supreme Court of Texas took pains to note that it was “not prepared to say that any single one of these factors could preclude justifiable reliance on its own and as a matter of law.” Orca Assets, 546 S.W.3d at 655.

As the Orca Court explained:

[A] party to an arm's length transaction must exercise ordinary care and reasonable diligence for the protection of his own interests . . . .Therefore, reliance upon an oral representation that is directly contradicted by the express, unambiguous terms of a written agreement between the parties is not justified as a matter of law ..... If written contracts are to serve a purpose under the law, relative to oral agreements, it is to provide greater certainty regarding what the terms of the transaction are and that those terms will be binding, thereby lessening the potential for error, misfortune, and dispute. . . .[A] party who enters into a written contract while

relying on a contrary oral agreement does so at its peril . . . .

Id. at 658 (quoting DRC Parts & Accessories, LLC v. VM Motori, S.P.A., 112 S.W.3d 854, 858- 59 (Tex. App.-Houston [14th Dist.] 2003, pet. denied)).

200. Following Orca Assets, the Texas Supreme Court explained that to hold that “reliance upon an oral representation that is directly contradicted by the express, unambiguous terms of a written agreement between the parties” is justifiable as a matter of law “would be to reward a party for signing a contract under false pretenses, promising to abide by the written terms while secretly intending to enforce the conflicting terms of an unwritten bargain.” Carduco, 583 S.W.3d at 559.

201. That principle applies here. Looking to the “circumstances” surrounding the 2015 OA's negotiation, see Orca, S.W.3d at 656, the court concludes that IBM's compliance-deflection strategies do not preclude BMC's reliance on the written contract's plain language. As one of the largest information technology companies in the world, IBM is a sophisticated business entity. So, too, is BMC. The record reflects that both companies used experienced, sophisticated, and savvy contract negotiators and business managers throughout their relationship. “Such world-savvy participants entering into a complicated, multi-million-dollar transaction should be expected to recognize ‘red flags' that the less experienced may overlook.” Orca Assets, 546 S.W.3d at 656. IBM is an intelligently run operation-and it was keen on promoting its theory that customer-directed displacements were permissible under the OA's “other valid business reasons” provision when BMC called it to account for its breach of the non-displacement clause. See, e.g., DX233 at 1-2 (Sweetman e-mail explaining that that “even if IBM were somehow deemed to be displacing software . . . acting at the customer's direction would constitute the most obvious example of a ‘valid business reason.'”). For IBM, the benefits of doing so were obvious: it could (1) weaken its principal competitor-BMC-in the mainframe software space; (2) acquire new business; and (3) minimize the risk of consequence by requiring BMC to litigate the contract language in court or pursue alternative dispute resolutions. To embrace IBM's reading of the caselaw-one in which a party's creation of “red flags” abrogated the terms of a contract-would perversely reward fraudulent parties for their bad deeds.⁴⁰ Gamesmanship may characterize business relationships between competitors, but it does not annul the legal force of a contract's unambiguous plain language, which serves as the touchstone for the court's reliance analysis in this case. See Westergren, 453 S.W.3d at 424-25 (“[A]s Texas courts have repeatedly held, a party to a written contract cannot justifiably rely on oral misrepresentations regarding the contract's unambiguous terms.”).

40 It would be nonsensical to prohibit a plaintiff from relying on a defendant's extracontractual statements contradicting the express terms of a contract but to inflexibly permit a defendant to negate a plaintiff's reliance on a written contract's express terms by identifying its own contradictory, extracontractual statements.

202. Moreover, as in Orca Assets, IBM's extracontractual representation that it could displace under section 5.4's “other valid business reasons” provision is contradicted by the contract itself, which uses the “other valid business reasons” provision to qualify IBM's ability to discontinue the use of BMC's software-not its ability to displace. See PX4. No reasonable person could “read the writing and still plausibly claim to believe [IBM's] . . . [oral] representation” that the contract permitted displacement under the “other valid business reasons” provision. See Orca Assets, 546 S.W.3d at 659. As the court has already recounted, the record undercuts IBM's assertion that it held a different interpretation of the non-displacement language in good faith. See Findings of Fact, supra ¶¶ 50-83 .

203. Thus, IBM cannot fault BMC for not relying on its extracontractual statements if BMC proves that it relied on the unambiguous terms of the contract. And, as a question of fact, the court concludes that BMC proved just that. BMC's witnesses testified at length that BMC relied on IBM's non-displacement promise in entering into the 2015 OA, as BMC explicitly refused to agree to IBM's repeated requests to change the non-displacement provision in ways that would have left AT&T unprotected and made a displacement project like Project Swallowtail permissible. Findings of Fact, supra ¶¶ 57-67. Indeed, BMC would never have agreed to the 2015 OA if it had known that IBM did not intend to honor its promise to protect AT&T's BMC products from displacement. Id. ¶ 66; Trial Tr. at 195:4-7, March 14, 2022 (Ah Chu); Trial Tr. At 174:6-16, March 15, 2022 (Jones); Bergdoll Dep. at 52:18-23. After lengthy negotiations around the scope of the non-displacement language, BMC agreed to limit the 2013 OA's existing language to a set number of accounts and a release absolving IBM of its past non-compliant conduct. Findings of Fact, supra ¶¶ 31, 61. The record and trial testimony credibly reflects that, through this exchange, BMC genuinely believed that IBM would not displace its products at AT&T. In that vein, BMC relied on IBM's written promise not to displace BMC's products with its own. BMC's reliance on the contract's unambiguous written text was justifiable under the circumstances.

4. BMC's Reliance Caused its Injury

204. “[T]ort damages are recoverable for a fraudulent inducement claim irrespective of whether the fraudulent representations are later subsumed in a contract or whether the plaintiff only suffers an economic loss related to the subject matter of the contract.” Formosa Plastics, 960 S.W.2d at 47. BMC argued that IBM's fraudulent behavior deprived it of its benefit-of-the-bargain under the 2015 OA. Benefit-of-the-bargain damages are recoverable under a fraudulent inducement claim. Bohnsack v. Varco, LP, 668 F.3d 262, 275 (5th Cir. 2012). Such damages “are appropriate in cases of fraudulent inducement when they are satisfactorily proven.” Id. at 275-76.

205. The court has already concluded that the contracts' text supports BMC's license fee model. Because BMC entered the 2015 OA with IBM, IBM was able to access and use BMC's software at AT&T for “no fee” while blatantly violating the non-displacement provision. IBM's violation of the contract resulted in direct damages-namely, the lost license fees-which BMC may recover. Thus, BMC's reliance caused its injury.

206. Because BMC has proved each element to sustain a fraudulent inducement claim under Texas law and shown that the unpaid license fees reflect direct damages, the court finds that BMC is entitled to $717,739,615.00 in direct damages associated with its fraudulent inducement claim. The court further finds that BMC is entitled to prejudgment interest in the amount of 5% on the $717,739,615.00 in direct contractual damages associated with its fraudulent inducement claim from September 22, 2017 through the date of judgment. See Meaux Surface Prot., Inc. v. Fogleman, 607 F.3d 161, 172 (5th Cir. 2010) (Texas law allows for an equitable award of prejudgment interest, which “should be granted to the prevailing party in all but exceptional circumstances”); see also Citizens Nat'l Bank v. Allen Rae Invs., Inc., 142 S.W.3d 459, 491 (Tex. App.-Fort Worth 2004, no pet.) (fraud).

E. New York Law Requires the Suspension of the Damage Limitations in MLA Sections 9 and 10.

207. Section 9 of the MLA provides that “[e]xcept for violation of proprietary rights and confidentiality . . .neither party . . . [is] liable for any special, indirect, incidental, punitive or consequential damages relating to or arising out of this agreement.” PX1 at IBM00035620. Section 10 likewise limits BMC damages “to the greater of $5,000,000 or the amount paid or payable by customer for the license to the applicable product.” Id. BMC argues that, under New York law, neither of these contractual limitations apply if the court concludes that it prevails on its fraudulent inducement claim. See Dkt. 598 at 25-26. IBM disagrees and argues that BMC “must prove something higher than ordinary fraud” for the damage waivers not to apply. Dkt. 597 at 25.

208. The Court of Appeals for New York answered the question in Kalisch-Jarcho Inc. v. City of New York, 448 N.E.2d 413, 416-17, 58 N.Y.2d 377, (N.Y. 1983). There, it explained:

[A]n exculpatory agreement, no matter how flat and unqualified its terms, will not exonerate a party from liability under all circumstances. Under announced public policy, it will not apply to exemption of willful or grossly negligent acts...More pointedly, an exculpatory clause is unenforceable when, in contravention of acceptable notions of morality, the misconduct for which it would grant immunity smacks of intentional wrongdoing. This can be explicit, as when it is fraudulent, malicious or prompted by the sinister intention of one acting in bad faith. Or, when, as in gross negligence, it betokens a reckless indifference to the rights of others, it may be implicit.

Id. at 416-17 (emphasis added). In other words, according to New York's highest court, “fraudulent, malicious, ” or “bad faith” conduct is sufficient to render exculpatory clauses unenforceable. See Id. This conclusion is not in tension with the Southern District of New York's decision in Net2Globe Int'l, Inc. v. Time Warner Telecom of New York, 273 F.Supp.2d 436 (S.D.N.Y. 2003), a case that IBM relies on for the proposition that application of the Kalisch-Jarcho doctrine requires more than a successful fraud claim.⁴¹ That court found that plaintiffs would have to make a “compelling demonstration of egregious intentional misbehavior evincing extreme culpability” to suspend a limitations-of-liability clause. Id. at 454. It follows that federal courts applying New York law would find that “fraudulent, ” “malicious, ” “bad faith, ” and “gross[ly] negligen[t]” behavior all meet the Southern District's “egregious intentional misbehavior” standard. See Kalisch-Jarcho, 448 N.E.2d at 416-17.

41 The parties agree on the basic contours of Kalisch-Jarcho's applicability. See Dkt. 612 at 23-24.

209. Indeed, in applying Kalisch-Jarcho, New York's intermediary appellate courts have explained that the “type of intentional wrongdoing that could render a limitation in [a contract] unenforceable is that which is ‘unrelated to any legitimate economic self-interest.'” Electron Trading, LLC v. Morgan Stanley & Co. LLC, 69 N.Y.S.3d 633, 636, 157 A.D.3d 579 (N.Y.App.Div. 2018). It is axiomatic that fraudulent inducement is unrelated to a legitimate economic self-interest. See CWCapital Cobalt VR Ltd. v. CWCapital Invs. LLC, 145 N.Y.S.3d 61, 68, 195 A.D.3d 12 (N.Y.App.Div. 2021) (applying Kalisch-Jarcho to unjust enrichment claims); TIAA Glob. Invs., LLC v. One Astoria Square LLC, 7 N.Y.S.3d 1, 9, 127 A.D.3d 75 (N.Y.App.Div. 2015) (“As for the limitation of liability clause, we note that the complaint alleges sufficient allegations of fraudulent conduct on the part of Seller such that, if proved, that clause would be unenforceable.”); Great N. Assocs., Inc. v. Cont'l Cas. Co., 596 N.Y.S.2d 938, 940, 192 A.D.2d 976, (N.Y.App.Div. 1993) (“Inasmuch as all of the 13 causes of action interposed by plaintiff against [defendants] allege intentional wrongdoing, willful, malicious and fraudulent acts, the release clearly is ineffective to insulate [defendants] from liability.”) Accordingly, BMC's successful fraudulent inducement claim is sufficient under New York law for the suspension of exculpatory clauses like those found in MLA sections 9 and 10, and BMC may therefore recover punitive damages against IBM.⁴²

42 Because the court concludes that the license fees relating to IBM's breach of section 5.4 are direct damages under the contract, the application of Kalisch-Jarcho does not raise the ceiling of recovery for BMC's license fee model. However, were the court incorrect that BMC's lost profits model is too speculative, see Conclusions of Law, supra ¶¶ 177-80, the suspension of MLA section 9's consequential damages disclaimer would enable BMC to recover otherwise unrecoverable lost profits.

210. Finally, the court notes that under Texas law, pre-judgment interest is not recoverable on an award of punitive damages. Tex. Civ. Prac. & Rem. Code Ann. § 41.007.

The evidence clearly and convincingly supports BMC's fraudulent inducement claim. BMC is entitled to recover the $717,739,615.00 value of the unpaid license fees related to IBM's breach of section 5.4, plus prejudgment interest. Because of IBM's fraud, the MLA's damages disclaimers and limits are unenforceable. The court now turns to whether IBM's conduct merits an award of punitive damages.

F. BMC's Punitive Damages Claim

211. The parties agree that Texas law permits the recovery of exemplary damages where a plaintiff proves by clear and convincing evidence that the harm with respect to which it seeks recovery of punitive damages resulted from the fraud. Tex. Civ. Prac. & Rem. Code § 41.003(a); Formosa Plastics Corp., 960 S.W.2d at 47; Dkt. 612 at 24. Punitive damages may be awarded against a defendant in an amount up to the greater of (1) two times the amount of economic damages, plus the amount of noneconomic damages; or (2) $200,000. Tex. Civ. Prac. & Rem. Code § 41.008(b).

212. “[T]he purpose of punitive damages is to punish a party for its outrageous, malicious, or otherwise morally culpable conduct and to deter it and others from committing the same or similar acts in the future.” Owens-Corning Fiberglas Corp. v. Malone, 972 S.W.2d 35, 40 (Tex. 1998) (internal quotations omitted). In addition, “exemplary damages must be reasonably proportioned to actual damages.” Alamo Nat'l Bank v. Kraus, 616 S.W.2d 908, 910 (Tex. 1981). Factors to consider in determining whether an award of exemplary damages is reasonable include: (1) the nature of the wrong, (2) the character of the conduct involved, (3) the degree of culpability of the wrongdoer, (4) the situation and sensibilities of the parties concerned, and (5) the extent to which such conduct offends a public sense of justice and propriety. Id. Through its fraudulent inducement claim, BMC can recover its $717,739,615.00 in direct contract damages. To determine a reasonable award of punitive damages, however, the court considers the four Kraus factors.

213. The court finds by clear and convincing evidence that IBM fraudulently induced BMC into entering the 2015 OA so that it could exercise rights without paying for them, secure other contractual benefits, and ultimately acquire one of BMC's core customers. See Findings of Fact, supra ¶¶ 43-83. IBM did this intentionally. See id.; see also Swinnea v. ERI Consulting Eng'rs, Inc., 481 S.W.3d 747, 757 (Tex. App.-Tyler 2016, no pet.) (upholding an award of punitive damages where party intentionally violated a fiduciary duty owed to longtime business partner). The court likewise determines that BMC genuinely believed that the narrowed scope of the non-displacement clause would put IBM's troubling history of non-compliance to bed. See Findings of Fact, supra ¶¶ 59-66. IBM is one of the largest technology companies in the world- and it exploited BMC's justifiable reliance for its own gain, cementing its abdication of good faith and fair dealing in the service of its own self-interest. See Findings of Fact, supra ¶¶ 5, 43-83. The degree of “reprehensibility of [IBM's] conduct is perhaps the most important indicium of [the] reasonableness of a punitive damage award.” Malone, 972 S.W.2d at 45-46 (internal quotations and alterations omitted). IBM's business practices-including the routine eschewal of rules- merit a proportional punitive damages award. See Findings of Fact, supra ¶¶ 35-58. Finally, IBM's conduct vis-à-vis BMC offends the sense of justice and propriety that the public expects from American businesses.

214. Therefore, the court finds that BMC is entitled to $717,739,615.00 in punitive damages.

The court concludes that BMC may recover under either a breach of contract theory for IBM's breach of section 5.4 of the 2015 OA or a fraudulent inducement theory. BMC has elected the theory providing the greatest recovery. See Dkt. 748 at 5. Accordingly, the court will enter judgment in BMC's favor on its fraudulent inducement claim.

G. Attorneys' Fees

215. Finally, section 18 of the MLA provides that the “prevailing party” in “any” litigation “is entitled to recover reasonable and customary attorney's fees and costs from the other party.” PX1 at section 18. New York law governs the contract.

216. New York follows the “American Rule” on the award of attorneys' fees, meaning that “attorneys' fees and disbursements are incidents of litigation, and the prevailing party may not collect them from the loser unless an award is authorized by agreement between the parties or by statute or court rule.” A.G. Ship Maint. Corp. v. Lezak, 69 N.Y.2d 1, 511 N.Y.S.2d 216 (1986); see also Versatile Housewares & Gardening Sys., Inc. v. Thill Logistics, Inc., 819 F.Supp.2d 230, 241 (S.D.N.Y. 2011). Parties may override the presumption by contractually agreeing to permit recovery of attorneys' fees, in which case “a federal court will enforce contractual rights to attorneys' fees if the contract is valid under applicable state law.” U.S. Fid. & Guar. Co. v. Braspetro Oil Serv. Co., 369 F.3d 34, 74 (2d Cir. 2004) (quoting McGuire v. Russell Miller, Inc., 1 F.3d 1306, 1313 (2d Cir. 1993)). Thus, where “a contract . . . provides for an award of reasonable attorneys' fees to the prevailing party in an action to enforce the contract, ” the language “is enforceable” so long as it “is sufficiently clear.” NetJets Aviation, Inc. v. LHC Commc'ns, LLC, 537 F.3d 168, 175 (2d Cir. 2008). And, although awards of attorneys' fees typically fall within the court's discretion, “where a contract authorizes an award of attorneys' fees, such an award becomes the rule rather than the exception.” McGuire, 1 F.3d at 1313.

217. The MLA's language is clear and enforceable. Because New York law recognizes private agreements to fees like the one at issue here, the “prevailing party” is entitled to an award. Thus, the court's threshold determination must be which, if any, party is the “prevailing party.” The parties have not briefed which of them, if either, is the prevailing party. Previously, the parties agreed to resolve the issue of attorneys' fees by motion and affidavit after entry of judgment. See Dkt. 612 at 20; Hearing Tr. 16:01-19, Dec. 17, 2021; see also Dkt. 594 at 16. Consistent with Federal Rule of Civil Procedure 54 and the parties' existing agreement, the court will order briefing to determine which party prevailed under New York law.

Conclusion

For the reasons discussed above:

1. BMC's Objections and Renewed Motion to Exclude the Expert Opinions of Bruce Hartley (Dkt. 617) is DENIED;

2. BMC's Objections and Renewed Motion to Exclude the Expert Opinions of Barry Graham (Dkt. 619) is DENIED;

3. BMC's Motion for Judgment on Partial Findings (Dkt. 665) is GRANTED in PART and DENIED in PART, as stated above.

4. IBM's objections to BMC's designated testimony of Alan Ratliff (Dkt. 621) are DENIED;

5. IBM's supplemental objections to testimony of Alan Ratliff (Dkt. 670) are DENIED;

6. IBM's objections to BMC's designated testimony of Kendyl Román (Dkt. 622) is DENIED AS MOOT;

7. IBM's Motion for Judgment on Partial Findings (Mar. 17, 2022 Trial Tr. at 20:8- 47:23; see Mar. 24, 2022 Trial Tr. at 4:17-35:14) (Dkts. 751, 752) is GRANTED in PART and DENIED in PART, as stated above;

8. BMC's breach of MLA section 8 claim is DISMISSED WITH PREJUDICE;

9. BMC's DTSA, TUTSA, and common law unfair competition through misappropriation claims are DISMISSED WITH PREJUDICE;

10. BMC's breach of 2015 OA section 5.1 is DISMISSED WITH PREJUDICE;

11. BMC lost profits claim is DISMISSED WITH PREJUDICE;

12. BMC shall recover direct and punitive damages from IBM on its fraudulent inducement claim, plus prejudgment interest in the amount of 5% and post-judgment interest:

a. BMC is entitled to recover from IBM $717,739,615 in actual contractual damages.

b. BMC is entitled to recover from IBM $168,226,367.29 in prejudgment interest on the actual damages described above.

c. BMC is entitled to recover from IBM $717,739,615 in punitive damages based on fraud found by clear and convincing evidence.

d. BMC is entitled to recover from IBM post-judgment interest from the date of the entry of judgment at the federally mandated rate. 28 U.S.C. § 1961.

SIGNED in Houston, Texas