Commercial Data Servers, Inc. v. INTERNATIONAL BUSINESS MACHINES CORPORATION

262 F.Supp.2d 50 (2003)

COMMERCIAL DATA SERVERS, INC., d/b/a Xbridge Systems, Inc. Plaintiff,
v.
INTERNATIONAL BUSINESS MACHINES CORPORATION Defendant.

No. 00CIV.5008CMLMS.

United States District Court, S.D. New York.

April 4, 2003.

Bernard Persky, Barbara Jane Hart, Goodkind, Labaton, Rudoff & Sucharow, L.L.P., New York City, Bruce L. Simon, Steven N. Williams, Cotchett, Pitre & Simon, Robert T. Scott, Cotchett, Pitre & Simon, Bruce L. Simon, Cotchett, Pitre & Simon, Burlingame, CA, for Commercial Data Servers, Inc. dba Xbridge Systems, Inc., plaintiff.

Evan R. Chesler, Cravath, Swaine & Moore, L.L.P., Evan R. Chesler, Cravath, Swaine & Moore, L.L.P., New York City, for International Business Machines Corporation, defendant.

MEMORANDUM DECISION AND ORDER DENYING DEFENDANT'S MOTION TO STRIKE THE DECLARATIONS OF RONALD S. ALEPIN AND BRETT L. REED, OVERRULING PLAINTIFF'S OBJECTIONS TO DEFENDANT'S EVIDENCE, AND GRANTING IN PART DEFENDANT'S MOTION FOR PARTIAL SUMMARY JUDGEMENT

MCMAHON, District Judge.

Plaintiff, Commercial Data Servers, Inc. ("CDS") alleges that Defendant International Business Machines Corporation ("IBM") committed federal and state antitrust violations under the Sherman Act §§ 1, 2 (15 U.S.C.A. §§ 1, 2), the Clayton Act § 3 (15 U.S.C.A. § 14), and the Donnelly Act (N.Y. Gen. Bus. Law § 340). Plaintiff also brings claims for tortious interference with its relations with two vendors. Defendant denies these allegations, and counterclaims for breach of two contracts and copyright infringement.

Defendant has moved for summary judgment under Fed.R.Civ.P. 56, on all of the plaintiffs claims except one claim of tortious interference. Defendant also moves for summary judgment on one of its breach of contract counterclaims.

Both the plaintiff and the defendant have also made separate motions regarding the admissibility of evidence submitted by the other. Plaintiff objects to the exhibits submitted by defendant with its Motion for Summary Judgment, on the grounds that the exhibits were not properly authenticated. Defendant moves to strike two expert affidavits submitted by plaintiff with its Motion in Opposition to Summary Judgment, on the grounds that the affidavits are unsubstantiated and speculative, and untimely under Fed. R.Civ.P. 26.

The motions are disposed of as follows: (1) I overrule plaintiffs objections to defendant's exhibits; (2) I deny defendant's motion to strike plaintiffs expert affidavits; (3) I grant defendant's motion for summary judgment on Counts II through VI of CDS's Second Amended Complaint, and deny summary judgement as to Count VII; and (4) I grant defendant's request for summary judgment on its breach of contract counterclaim.

FACTUAL BACKGROUND

IBM is a New York corporation that maintains its principal executive offices in Armonk, New York. (Plaintiffs Second Amended Complaint, ("Plaintiffs Sec. Amd. Cmplt."), ¶ 8.) IBM describes itself as a company that "uses advanced information technology to provide customer solutions through, inter alia, the sale of a variety of computer hardware and software products." (Defendant's Amended Answer and Counterclaims, p. 11.) CDS (which is currently doing business under the name "Xbridge Systems, Inc.") is a California corporation that maintains its principal executive offices in California. (Plaintiffs Sec. Amd. Cmplt, ¶¶7, 10.) CDS was formed in 1996 by three former IBM executives. Id. at ¶ 22. According to CDS's pleadings, CDS provided its founders with an "opportunity to compete with IBM."^[1]Id. at ¶ 24.

Over a number of years, IBM developed the S/390 computing platform, "a component-based computer system consisting of hardware and software products that are compatible and interoperable with the platform's architecture." (Plaintiffs Second Amd. Cmplt., ¶ 13.) The primary hardware component of the S/390 system is a central processing unit ("CPU") board, referred to as the "P/390 Card." Id. at ¶¶ 13, 26. The P/390 card enables the computer to run OS/390 operating system software, and other predecessor IBM operating system software.^[2] Other operating system software, such as software based on the UNIX or Windows NT operating systems, generally cannot run on S/390 platform computers. Likewise, S/390 operating system software and applications generally cannot be used on alternative platform computers.

IBM builds and markets its own S/390 computer systems and products. However, it also has agreements with other companies under which they incorporate IBM technology into their products. Plaintiff CDS was one of these companies.

In 1995, CDS entered into an Original Equipment Manufacturer ("OEM") Agreement with IBM, under which CDS was able to purchase the P/390 card and S/390 operating software. Id. at ¶ 26. CDS purchased the software at the entry system level ("ESL"), which was IBM's most favorable software pricing level. Id. ¶¶ 4, 20. The OEM Agreement allowed CDS to incorporate the P/390 Card into its own product, and licensed CDS to "preload" S/390 operating system software onto these CDS systems. Pursuant to this agreement, in 1997 CDS introduced the CDS 104 system (later renamed CDS 2000). Id. at ¶ 26.

In 1997, CDS entered into another agreement with IBM, an Original Provider Agreement ("OPA"), pursuant to which CDS invested in IBM's efforts to create an improved P/390 card—the "P/390e" card. Id. at ¶ 29. In 1998, the OEM Agreement was amended to allow CDS to purchase the P/390e card. ("OEM" Agreement and amendments, Ex. 23 to Burke Dec.) CDS then developed a product incorporating the P/390e card and the OS/390 operating system, which it marketed as the "CDS 2000e." (Plaintiffs Sec. Amd. Cmplt. at ¶ 29.)

In May 1998, IBM "pre-announced" a new product, the S/390 Integrated Server (also referred to as the "Planter"), scheduled for release in November 1998. Id. at ¶¶ 16, 69. At the same time, IBM also preannounced an additional S/390 product, with higher processing power (estimated to be two times as fast). Id. Both products was based on the P/390E, and were designed to compete with CDS's computers. Id. at ¶ 51, 52. IBM released the integrated server in November 1998, as scheduled. In September 1999, IBM released the "Multiprise 3000" servers, which provided processing power approximately six times greater than CDS's product. IBM stopped offering the S/390 Integrated Server in February 2000. Id. at ¶ 71. The Multiprise products are still on the market.

CDS alleges that IBM viewed CDS as a competitive threat and sought to "eliminate" them. Id. at ¶ 6. CDS alleges that the introduction of the S/390 Integrated Server, and subsequent decision to "drop[ ]" the server from the market, was intended to destroy the CDS market opportunity. Id. CDS also alleges that IBM refused to provide CDS with ESL pricing qualification except under anti-competitive terms and conditions which limited the marketing claims that CDS could make regarding its product. Id. at ¶¶ 44-45. IBM also allegedly "threatened" and/or "coerced" two value added resellers ("VARs"), Information Technology Co. ("ITC") and Intelliware, in an effort to "prevent CDS from distributing the CDS 2000" through certain distribution channels. Id. at ¶¶ 756-66.

PROCEDURAL HISTORY

Plaintiffs original complaint in this action was filed on July 7, 2000. Counts I through VI of the original complaint alleged various antitrust violations, Count VII alleged tortious interference with prospective business relations, Counts VIII and IX alleged misappropriation of plaintiffs technology, and Count X alleged breech of the OPA Agreement. Defendant moved to dismiss Counts VIII through X and the Hon. Louis Stanton granted defendant's motion, albeit with leave to replead. Commercial Data Servers, Inc. v. International Business Machines Corp., 2001 WL 277303 (S.D.N.Y. March 21, 2001)("CZ)S I"). Pursuant to the Court's Rules for the Division of Business, the case was sent to White Plains and reassigned to me.

On May 8, 2001, plaintiff filed its first amended complaint, which repeated the original Counts I through VII and repleaded Counts VIII and IX, the misappropriation claims dismissed by Judge Stanton. Defendant moved to dismiss the complaint in its entirety. This Court granted defendant's motion on October 9, 2001. Commercial Data Servers, Inc. v. International Business Machines Corp., 166 F.Supp.2d 891 (S.D.N.Y.2001)("CDS IF). The misappropriation claims previously dismissed by Judge Stanton, Counts VIII and IX, were dismissed with prejudice. The antitrust claims, Counts I through VI, were dismissed without prejudice, due to plaintiffs failure to properly allege a relevant market. Id. at 896. The tortious interference claim, Count VII, was dismissed without prejudice, due to plaintiff failure to identify any particular "VARs" that plaintiff had a relationship with who were intimidated by defendant. Id. at 898.

On November 13, 2001, plaintiff filed a second amended complaint, realleging with greater specificity the claims that were dismissed without prejudice. Defendant moved a third time under Fed.R.Civ.P. 12(b)(6), this time seeking to dismiss the antitrust claims, Counts I through VI. On March 15, 2002, this Court dismissed Count I for failure to state a claim for per se violation of Section 1 of the Sherman Act. Commercial Data Servers, Inc. v. International Business Machines Corp., 2002 WL 1205740 (S.D.N.Y. March 15, 2002)("CDS III"). I denied the motion as to the other claims. Thereafter, defendant answered the second amended complaint and asserted three counterclaims: (1) breach of the OPA; (2) breach of the OEM Agreement; and (3) copyright infringement.

Discovery in this matter began in the summer of 2001, while the motion to dismiss the first amended complaint was pending. After accommodating the parties by twice extending the close of discovery, I finally ordered that all discovery must be completed on or before February 28, 2002. At the conclusion of discovery, the parties had deposed fourteen individuals and exchanged numerous documentary evidence, including contracts, copies of internal emails, marketing and sales materials, and internal strategy documents. CDS also provided IBM with responses to two sets of Requests for Admission and two sets of Interrogatories.

Two days after I ruled on the final motion to dismiss, defendant filed a motion for summary judgment pursuant to Fed. R.Civ.P. 56, seeking judgment in its favor as to the plaintiffs surviving claims against it (Counts II through VII) and as to its counterclaim for breech of the OPA.

In addition to IBM's summary judgment motion, each side has submitted motions challenging the admissibility of supporting documents submitted by its opponent with its summary judgment motion papers. I deal with the evidentiary motions first, and then turn to the merits.

DISCUSSION

I. Discovery Motions

When ruling on a motion for summary judgment, courts need only consider evidence that would be admissible at trial. Raskin v. Wyatt Co., 125 F.3d 55, 66 (2d Cir.1997). Each side asserts that materials submitted by the other are not admissible under the Federal Rules of Evidence and/or the Federal Rules of Civil Procedure, and thus should not be considered by this Court in deciding the summary judgment motion. In considering the admissibility of documents, the court appropriately draws all reasonable inferences in favor of the opposing party, as it is the burden of the moving party to prove that there is no genuine issue of material fact. See Wright & A. Miller, Federal Practice and Procedure § 2738, p. 342-45; see also U.S. v. Bell, 27 F.Supp.2d 1191, 1194 (E.D.Cal. 1998); Western Land Corp. v. Crawford-Merz Co., 62 F.R.D. 550 (D.Minn.1973).

A. CDS's Objection to Exhibits Submitted by IBM

Local Civil Rule 56.1 requires a party seeking summary judgment under Fed. R. Civ. Pro. 56 to include a statement of the material facts as to which the moving party contends there is no genuine issue to be tried. Each statement of material fact must be followed by a "citation to evidence which would be admissible, set forth as required by the Federal Rule of Civil Procedure 56(e)." Accordingly, IBM's Motion for Summary Judgment is supported by a Statement of Material Facts Pursuant to Local Rule 56.1 ("Statement of Material Facts") and four volumes of exhibits attached to the Declaration of Kevin A. Burke, Esq., an attorney for IBM ("Burke Declaration"). The Burke Declaration states that true and complete copies of the following documents, all of which are referenced in the Statement of Material Facts, are attached: (1) excerpts of CDS's Responses to Interrogatories and Requests for Admission; (2) excerpts of deposition transcripts; (3) the expert report of Carl Shapiro; and (4) other documents produced during discovery and referenced in the Statement of Material Facts, some of which were produced by IBM and some of which were produced by CDS. In total, there are eighty exhibits contained in the four volumes attached to the Burke Declaration.

Plaintiff objects that the exhibits attached to the Burke Declaration have not been properly authenticated as required under Fed.R.Evid. 901(a). (Plaintiffs Objection to Defendant's Evidence in Support of its Motion for Summary Judgment, submitted April 12, 2002, p. 1.) Plaintiff argues that Mr. Burke is not someone through whom the documents could be admitted into evidence at trial. Id.

Defendant responds that plaintiffs general objection to the exhibits as a whole is improper under Fed.R.Civ.P. 7(b)(1), which requires that motions "shall state with particularity the grounds therefor." (Defendant's Response to Plaintiffs Objection to Defendant's Evidence, submitted April 29, 2002, p. 1.) Defendant further responds that the documents are sufficiently authenticated because their form, their appearance, the circumstances through which they were acquired, and the record evidence indicate reliability. Id. In support of these assertions, defendant submits the Declaration of Keith R. Hummel ("Hummel Declaration"), an attorney for IBM, which attaches portions of deposition transcripts in which CDS witnesses discuss various CDS documents which were attached to the Burke Declaration. Defendant also submits the Declaration of Joanne Richardson ("Richardson Declaration"). Ms. Richardson, an IBM Legal Analyst employed by the company for twenty two years, attests that twenty five of the exhibits are IBM documents maintained in the ordinary course of business that were produced to CDS pursuant to document requests in this litigation. Finally, defendant submits a table, attached as an exhibit to its response to plaintiffs motion, which describes each exhibit in the Burke Declaration and provides the basis for the authentication of the exhibit. ("Authentication Table").

Plaintiff replies that it is unfair for IBM to attempt to authenticate documents in its opposition papers that it did not initially authenticate, and maintains that the exhibits are still are not properly authenticated. (Plaintiffs Reply to Defendant's Response to Plaintiffs Objection to Evidence, submitted May 16, 2002, p. 2.) As a preliminary matter, I reject Plaintiffs contention that it is "unfair" for IBM to submit, and for the court to consider, the Richardson Declaration. I consider both the Richardson and Hummel Declarations to be supplemental affidavits to the Burke Declaration, and under F.R. Civ. P. 56(e), the Court may permit affidavits to be supplemented or opposed by depositions, answers to interrogatories, or further affidavits.^[3]

While I agree with defendant that plaintiffs motion is very broad, the grounds are stated with sufficient particularity—plaintiff is objecting to my considering the documents attached to the Burke affidavit on the grounds that they have not been properly authenticated. However, since plaintiff has not objected on a document-by-document basis, I shall respond by discussing the admissibility of categories of documents. See Kendrick v. Sullivan, 766 F.Supp. 1180, 1192 n. 1 (D.D.C.1991) (noting that defendant's arguments against plaintiffs documentary evidence in opposition to summary judgment were overly broad and concluding that the court was compelled to respond to the motion in general terms),

1) CDS's Responses to Interrogatories and Requests for Admission

The Burke Declaration states that Exhibits 1 through 5 are true and complete copies of CDS's Responses to Interrogatories and Requests for Admission. Mr. Burke, as an attorney for the firm representing IBM, has personal knowledge that the documents attached were obtained during discovery by the firm.^[4]

Generally, under Fed.R.Civ.P. 56(c), pleadings, depositions, answers to interrogatories, and admissions on file can be utilized by the court in considering a summary judgment motion. Under F.R. Evid. 901(a), the requirement that documents be authenticated "is satisfied by evidence sufficient to support a finding that the matter in question is what the proponent claims." Under F.R. Evid. 901(b), testimony of a witness with knowledge provides appropriate authentication. Mr. Burke's declaration that these documents are true and correct copies of CDS's responses constitutes testimony by a witness with knowledge of what the firm received from its opponent in response to legitimate discovery requests. In addition, under F.R. Evid. 901(b)(4), authentication is provided by "appearance, contents, substance, internal patterns, or other distinctive characteristics, taken in conjunction with the circumstances." Each of the challenged documents contains the caption of the instant case, is signed by the plaintiffs attorney, is available to both parties in the action, and is being submitted to the Court by the defendant's attorney with a declaration that it is a true copy. Plaintiff, who is clearly in the position to know if the exhibits are not authentic copies of its Responses to Interrogatories and Requests for Admission, has not objected that the exhibits are not authentic, just that they are not authenticated. The characteristics of Exhibits 1-5, and the surrounding circumstances, indicate that these are authentic documents.

2.) Excerpts of Deposition Transcripts

The Burke Declaration states that Exhibits 6 through 22 are true and complete copies of excerpts of transcripts from depositions taken in this action. Depositions are expressly recognized by Fed. R.Civ.P. 56(c) as relevant to consideration of a summary judgment motion.

Mr. Burke has personal knowledge that the documents are deposition transcripts taken during the discovery period of this action. In fact, either Mr. Burke himself or one of his colleagues from Cravath, Swain & Moore personally attended each of these depositions. Therefore, his declaration that the exhibits are true and correct copies of discovery materials is testimony of a witness with knowledge, which provides authentication under F.R. Evid. 901(b). In addition, I find that the appearance of the documents, each of which includes a cover page and the title page, and the circumstances through which they are presented to the Court in this action, authenticates them under F.R. Evid. 901(b)(4). Finally, I note that according to the cover page of each transcript, plaintiffs counsel appeared at each of these depositions. Plaintiff never claims that the cover page is a fabrication (that plaintiffs counsel was not there) or asserts that its counsel was there and that the transcripts are not accurate depictions of what took place. Again, plaintiff is in a position to know if the documents are not authentic, but makes no such claim.

3.) The Expert Report of Carl Shapiro

The Burke Declaration states that Exhibit 80 is a true and complete copy of the Expert Report of Carl Shapiro, referenced in IBM's Statement of Material Facts. Plaintiff does not challenge the admissibility of the defendant's expert report based on its substance, only for lack of any authentication. Mr. Burke is competent to testify that Exhibit 80 is a true and complete copy of IBM's expert's report. At trail, the expert could be expected to testify as to the contents of the report.

Further, I note that CDS has itself submitted Mr. Shapiro's expert report into evidence, as Exhibit 15 to the Declaration of Steven N. Williams. CDS thus challenges a document for lack of authentication, while at the same time submitting an identical copy of the document. Such tactics unnecessarily occupy the Court's time on matters not touching the merits of this case.

4.) Additional Documents Produced by IBM and CDS

Over two thirds of the exhibits attached to the Burke Declaration are documents produced during discovery by IBM or CDS, including letters, contractual agreements between the parties, and marketing materials. The Authentication Table, attached as an exhibit to defendant's response to plaintiffs objection motion, provides the basis for authentication of each of these documents.

As to the CDS documents, the Authentication Table notes that all of these documents were produced by CDS in the course of discovery, and bear CDS production numbers. In addition, the Table notes that many of the CDS documents are also authenticated by CDS's Response to IBM's Second Set of RFAs, or by CDS employees during their depositions. The Hummel Declaration, cross-referenced by the Authentication Table, includes as exhibits the specific deposition excerpts in which certain documents were discussed.

It is disingenuous and wasteful for plaintiff to object that its own documents are not authenticated, and thus inadmissible at trial and on summary judgment. The appearance of these documents and the circumstances surrounding this motion— most importantly the fact that the plaintiff is in the best position to know if they are authentic and that they have never claimed that they are not—show that these are authentic documents. See F.R. Evid. 901(b)(4)

The IBM documents are authenticated by the sworn Declaration of Joanne Richardson, an IBM employee for over twenty two years, who attests that each of the documents is a true and complete copy of an IBM document maintained by IBM employees in the regular course of business. As discussed above, under F.R. Evid. 901(b), testimony of a witness with knowledge provides appropriate authentication for documents. Ms. Richardson is a witness with knowledge, and her sworn affidavit provides authentication for the IBM documents.

I thus find that all of the documents challenged by plaintiff as "unauthenticated" meet either the requirements of Fed. R. Evid 901(b) or 901(b)(4). There is sufficient evidence of their authenticity for the court to consider these documents on this motion for summary judgment. See Bieda v. J.C. Penney Communications, Inc., 1995 WL 437689, *9 n. 2 (S.D.N.Y. July 25, 1995) (denying defendant's motion to strike plaintiffs documents from consideration in a summary judgment motion when defendants did not challenge the authenticity of the documents, but rather the plaintiffs ability to authenticate them, and finding that there was sufficient evidence of authenticity based on the appearance of the documents, the defendant's interrogatory responses regarding the documents, and the fact that defendant had produced the documents). The court deems the documents authenticated and considers them.

B. IBM'S Motion to Strike CDS Declarations

On April 12, 2002, plaintiff filed its opposition to defendant's motion for summary judgment, along with supporting documents, including the expert affidavits of Ronald S. Alepin and Brett L. Reed. On April 29, 2002, defendant submitted a motion to strike the expert affidavit of Mr. Alepin and a separate, yet very similar, motion to strike the expert affidavit of Mr. Reed. Both motions argue that the affidavits should be stricken on two grounds. First, defendant claims that the opinions expressed in each expert's affidavit are so different from the opinions expressed in his "Rule 26 Report" (report pursuant to Fed.R.Civ.Pro. 26) that the affidavit is, in essence a new and untimely expert report. Second, defendant claims that each expert's opinion is unsubstantiated and speculative, and argues that the affidavits should thus be disregarded as conclusory. Plaintiff argues that both declarations are consistent with the underlying expert reports and deposition testimony, and that both experts' opinions are properly supported. (Plaintiffs Opposition to Defendant's Motions to Strike Declarations of Ronald S. Alepin and Brett L. Reed, submitted May 16, 2002.)

I conclude that Mr. Alepin's Affidavit is entirely consistent with his timely Rule 26 Report. Mr. Reed's Affidavit goes well beyond his Rule 26 report. However, I will overlook that for purposes of this motion, because the failure is harmless. Neither affidavit is impermissibly speculative.

Fed.R.Civ.P. 26(a)(2)(B) ("Rule 26(a)(2)(B)") requires that an expert's report contain "a complete statement of all opinions to be expressed and the basis and reasons therefor..." It is essential that the expert's opinions be known to the other party, so that they may properly prepare their opposition. As a result, under Fed.R.Civ.P. 37(c)(1) ("Rule 37(c)(1)"), "a party that without substantial justification fails to disclose information required by Rule 26(a)... is not, unless such failure is harmless, permitted to use as evidence at trial, at a hearing, or on a motion any witness or information not so disclosed." Defendant provides examples of several cases where courts have held that affidavits containing opinions that were not disclosed in a Rule 26 report would be inadmissible at trial, and thus should not be considered on a motion for summary judgment. See Brumley v. Pfizer, Inc., 200 F.R.D. 596, 603-604 (S.D.Tex.2001) (affidavit stricken to the extent that it went beyond opinions expressed in the Rule 26 report); Baker v. Indian Prairie Cmty. Unit, School Dist. No. 204, 1999 WL 988799, *2-*3 (N.D.Ill. Oct. 27, 1999) ("The Court will not allow [Plaintiffs] to ambush Defendants with new expert opinions after the expert disclosure deadline and after they filed or summary judgment."); Penland v. BIC Corp., 796 F.Supp. 877, 881 (W.D.N.C.1992) (striking expert affidavits filed in opposition to summary judgment because they "contained] purported expert opinion not previously disclosed ... as required by the Federal Rules of Civil Procedure").

I have reviewed the Rule 26 Reports and expert affidavits of both Mr. Alepin and Mr. Reed. In the light most favorable to plaintiff, I find that Mr. Alepin's affidavit is substantially similar to his Rule 26 Report, that the opinions therein would be admissible at trial (albeit subject to crossexamination), and that the affidavit may be considered on the motion for summary judgment.

Mr. Reed's affidavit, however, offers his opinions on an entirely different topic than the one discussed in his Rule 26 report. The Reed Affidavit opines on switching costs and "lock-in" of IBM customers. All of the opinions and analysis in the Rule 26 Report relate to damages, via lost opportunities to CDS or sales by IBM.

Plaintiff insists that "analysis of the lock-in was this [sic] explicitly and implicitly part of his entire analysis of this case," and that the subject matter is "coextensive." (Plaintiffs Memorandum in Opposition to Defendant's Motion to Strike the Expert Affidavits, p. 5.) It may be that Mr. Reed analyzed whether IBM locked-in its customers via high switching costs and simply omitted that analysis from his report. Or it may be that he assumed the fact (lock-in) and proceeded from there. Either way, Mr. Reed's Rule 26 Report does not discuss the issue of lock-in and does not offer any opinions on that subject, including opinions about how the asserted lock-in came into being. It is simply and solely a damages analysis—number crunching. In contrast, his expert affidavit provides detailed opinions regarding how IBM S/390 customers were allegedly locked into uneconomical IBM systems by high switching costs, how IBM allegedly took advantage of these customers, how IBM engaged in other allegedly "anticompetitive" conduct, and why the relevant market for anti-trust considerations should be limited to those customers "locked-in" to the S/390 system.

Plaintiff further argues that "the examination of Mr. Reed on [the topic of lock-in] covers many pages of his deposition." Id. But as defendant correctly notes, deposition testimony on a topic does not cure a failure to provide Rule 26 disclosure. See e.g., Ferriso v. Conway Org. 1995 WL 580197, at *2-3 (S.D.N.Y. Oct. 3, 1995) Moreover, Mr. Reed's deposition testimony about "lock-in" is very general, and is not couched in the form of opinions about how any lock-in arose or what caused it. He assumes that there is a lock-in effect for the purposes of calculating damages. Whether or not these opinions were "implicitly" part of Mr. Reed's analysis, he never explicitly discussed them in his Rule 26 Report or at his deposition. The topics of damages and lock-in/switching costs are certainly interrelated in some respects, but they are not "coextensive."

However, failure to comply with Rule 26(a) does not always provide a basis for sanctions under Rule 37(c)(1). As noted above, the text of Rule 37(c)(1) provides that the evidence should be precluded only if the party that failed to disclose the information is "without substantial justification" and if the failure is not "harmless." Fed.R.Civ.Pro. 37(c)(1). In addition, courts in this Circuit have considered imposition of sanctions under Rule 37 a "drastic remedy" that should only be applied "in those rare cases where a party's conduct represents flagrant bad faith and callous disregard" of the Federal Rules. See e.g., Johnson Elec. N. Am. v. Mabuchi Motor Am. Corp., 77 F.Supp.2d 446, 458 (S.D.N.Y.1999); Sterling v. Interlake Industries, Inc., 154 F.R.D. 579, 587 (E.D.N.Y.1994).

Here the failure is harmless, because Mr. Reed's Affidavit does not create a genuine issue of material fact that prevents me from granting IBM's motion. Therefore, the better part of valor is to consider it rather than strike it.

Defendant also argues that the Alepin and Reed affidavits consist of "speculative and unsubstantiated assertions" that are not properly supported by "underlying bases or citation" and thus do not satisfy Rule 26(a)(2)(B). Defendant urges the court to conclude that these expert opinions do not create any genuine issues of material fact, and thus should not be considered on summary judgment, citing Raskin, 125 F.3d at 66-68 and Virgin Atlantic Airways, Ltd. v. British Airways, PLC, 69 F.Supp.2d 571, 579-80 (S.D.N.Y.1999). Rather than doing that, I have chosen to consider them—and having done so, I conclude that they do not create any genuine issue of material fact.

II. Summary Judgment Standard

Defendant has moved for summary judgment pursuant to Fed.R.Civ.P. 56(c), which allows the court to grant judgment in a movant's favor when there is no genuine issue of material fact. In considering any summary judgment motion, the court draw all reasonable inferences in the nonmovant's favor, but in an antitrust case, "those inferences must be reasonable in light of competing inferences of acceptable conduct." Tops Markets, Inc. v. Quality Markets, Inc., 142 F.3d 90, 95 (2d Cir. 1998) citing Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 588, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986) "By avoiding wasteful trials and preventing lengthy litigation that may have a chilling effect on pro-competitive market forces, summary judgment serves a vital function in the area of antitrust law." Id. citing Matsushita Elec, 475 U.S. at 593-94, 106 S.Ct. 1348 (additional citations omitted). Where the nonmoving party has the burden of proof at trial, "it must go beyond the pleadings to demonstrate the existence of some specific facts that create a genuine issues as to those matters for which it has the burden of proof." Id. citing Celotex Corp. v. Catrett, 477 U.S. 317, 324, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986).

III. The Relevant Product Market

CDS alleges that defendant violated the Sherman Act and the Clayton Act, federal antitrust laws, and the Donnelly Act, a New York state antitrust law that mirrors the Clayton Act. The threshold issue in considering each of these antitrust claims is determining the relevant market. Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263, 268 (2d Cir.1979)("[T]he first step in a court's analysis must be a definition of the relevant market.")(citing United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 391-93, 76 S.Ct. 994, 100 L.Ed. 1264 (1956)).

CDS alleges that the relevant market for purposes of this litigation includes only "low-end IBM mainframe S/390 computers with processing power of 10 MIPS, or less." (Plaintiffs Sec. Amd. Cmplt. ¶74.) CDS's clarifies that not all users of "10 MIPS, or less" S/390 computers belong in the relevant market. (Reed Aff., ¶ 12.) Rather, CDS claims that the market includes only some sub-set of users who "are locked-in to the IBM platform" due to the high costs of switching to a non-IBM platform. Id. Plaintiff argues that these IBM customers had invested so much time and money in S/390 equipment and software applications written for the S/390 system that they faced prohibitively high switching costs, and were thus "lock[ed]-in" to the S/390 system. Id. at 74-75.

At the motion to dismiss stage, this Court found that plaintiff properly alleged that switching costs associated with changing platforms were prohibitive, which gave rise to a theoretically rational explanation for a distinct market at the "low-end." CDS III, 2002 WL 1205740 at *6-7. Thus, the Court concluded that plaintiffs market definition was "plausible on its face." But I also noted, "whether plaintiff will succeed in proving the alleged market is a question for a later date." Id citing Todd v. Exxon Corp., 275 F.3d 191, 204-205 (2d Cir.2001).

That "later date" has arrived. And while plaintiffs definition of the relevant market was sufficient as an allegation, plaintiff has failed to advance evidence that creates a genuine issue of material fact concerning the relevant market.

A. Standard for Relevant Product Market Definition

The relevant product market is "the area of effective competition" within which a firm's products compete.^[5] AD/SAT v. Assoc. Press, 181 F.3d 216, 227 (2d Cir. 1999) quoting Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 327, 81 S.Ct. 623, 5 L.Ed.2d 580 (1961). The market must include all products "that have reasonable interchangeability for the purposes for which they are produced—price, use and qualities considered". United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 404, 76 S.Ct. 994, 1012, 100 L.Ed. 1264 (1956); see AD/SAT, 181 F.3d at 227. Products are "reasonably interchangeable" where there is sufficient cross-elasticity of demand, i.e., where consumers would respond to an increase in the price of one product by purchasing another product. See AD/SAT, 181 F.3d at 227. Products need not be identical, however, to be included in the same market, complete functional overlap is not required. See United States v. Grinnell Corp., 384 U.S. 563, 572-73, 86 S.Ct. 1698, 1704-05, 16 L.Ed.2d 778 (1966); E.I. du Pont de Nemours, 351 U.S. at 393, 76 S.Ct. at 1006 ("one can theorize that we have monopolistic competition in every nonstandardized commodity with each manufacturer having power over the price and production of his own product. However, this power ... is not the power that makes an illegal monopoly. Illegal power must be obtained in terms of the competitive market for the product." (footnote omitted)).

B. A Reasonable Trier of Fact Would Have to Conclude that S/390 Platform Computers Are Reasonably Interchangeable with UNIX and Windows NT Platform Computers.

IBM argues that the record includes evidence of vigorous competition, based on reasonable interchangeability and cross-elasticity of demand, between S/390-compatible computers and alternative platform computers, primarily those compatible with the UNIX or Windows NT operating systems.

IBM documents acknowledge the competitive impact that alternate platform computers, particularly those incorporating UNIX or Windows NT operating systems, had on the sale of S/390 platform computers. (See e.g., "S/390 Strategy" Document, p. 5, September 1995, Ex. 41 to Burke Dec. ("Fierce competition from alternate platforms has eaten into the potential growth of the S/390 workload."); "Alternate Platform Market Model Survey Data, U.S. and Europe," January, 1995, Ex. 42 to Burke Dec. (finding that S/390 customers were projecting movement of "significantly more" workload off the S/390 platform over the next few years); "S/390 Strategy Document, p. 1, July, 1996, Ex. 47 to Burke Dec." (noting accelerating competitive pressures, predicting that UNIX vendors will "continue their strategy" to replace S/390 solutions and predicting that "in the future, the greatest threat will come from NT.").) CDS itself alleges that IBM's favorable entry server level ("ESL") pricing for S/390 products was offered in response to the price competition faced from UNIX and Windows, though it does so while alleging that IBM was seeking to "lock-in" new customers by enticing software developers to write S/390 software applications. (Plaintiffs Second Amd. Cmplt. ¶ 20.)

CDS's own sales and marketing documents identified UNIX and Windows NT based systems as "Substitutes" for CDS's own S/390-compatible products ("CDS Systems"). ("Commercial Data Servers Marketing Strategy and Implementation Plan," Ex. 30 to Burke Dec; "Commercial Data Servers Sales Strategy and Implementation Plan," Ex. 31 to Burke Dec.) These documents state that S/390-compatible manufacturers were "under severe competitive pressures from UNIX and in the near future this competition will also come from NT". (Id.) In addition, CDS's marketing material and private placement memoranda acknowledged cross-platform competition faced by the CDS Systems. (CDS Executive Summary, Ex. 29 to Burke Dec; CDS Private Placement Memorandum, July 1998, Ex. 32 to Burke Dec; CDS Private Placement Memorandum, September 1995, Ex. 33 to Burke Dec; CDS Private Placement Memorandum, September, 1998, Ex. 34 to Burke Dec.)

Both CDS and IBM witnesses attested to the existence of vigorous competition between S/390 platform computers and UNIX and Windows platform computers. For example, Ron Hankison, CDS's corporate designee, testified:

We would, no matter where we went, find customers that had an intention or desire to move away from 390 into some other architecture, and we'd run into UNIX and NT as an alternative.

(Hankinson Dep., 30:9-13, Ex. 10A to Burke Dec. accord Holstrom Dep., 67:8-22, Ex. 11A to Burke Dec; Smarting Dep., 35:22-36:2, Ex. 22A to Burke Dec.) Similarly, Robert Port, IBM Manager of S/390 Sales Support, testified:

[T]he real competition here was all the other alternative platforms and choices the customer had for deploying their applications. So, I view in this Sun was a competitor, HP is a competitor, DEC is a competitor. Unysis is a competitor. Intel based alternatives, some companies like Compaq were competitors. This was a very, very competitive environment....

(Port Dep., 172:17-173:3, Ex 17B to Burke Dec); accord Brogan Dep., 219:6-12, Ex. 7E to Burke Dee. (IBM asserts that all of the competitors identified by Mr. Port sold non-S/390 platform computers.). Mr. Port further stated:

The competition was the Unix vendors and the [Windows] NT vendors.

Those were the guys we were really competing against.

(Port Dep., 229:18-20, Ex. 17D to Burke Dec.)

CDS itself alleges that, under certain circumstances (e.g., with respect to new workload), S/390-compatible computers competed directly with UNIX and Windows-compatible computers. See Plaintiffs Second Amd. Cmplt. ¶¶ 18-19 ("competing systems offered by Windows and Unix, among others, were an increasingly tempting alternative"; "Windows NT and UNIX... [represented] cost-effective alternative[s].") CDS even admits that it lost customers to computers compatible with the UNIX and Windows NT operating systems. (CDS's Responses to IBM's Second Set of RFAs, Nos. 30, 31 (Ex. 3); Holstrom Dep., 55:16-24, Ex. 11A to Burke Dec.)

The record contains ample evidence that this competition is not limited to new customers. Surveys conducted by both IBM and CDS show that customers who had previously purchased S/390 compatible systems migrated to non-S/390 platform computers during the relevant time frame. (See e.g., Results of CDS telemarketing survey, Ex. 45 to Burke Dec; IBM Surveys, Exs. 49 and 50 to Burke Dec.) Between 1995 and 2000, the number of S/390 platform computers worldwide decreased 18% and the number of customers with S/390-compatible computers decreased 17%. (IBM Data Chart, Ex. 51 to Burke Dec.) The decrease in the installed base was even more pronounced in the "10 MIPS or less" arena on which CDS focuses. Between 1995 and 2000, the number of S/390 platform computers worldwide with processing power of "10 MIPS or less" decreased 60% and the number of customers with S/390 platform computers with processing power of "10 MIPS or less" decreased 59%.^[6] (IBM Data Chart, Ex. 55 to Burke Dec.) During this same period, spending on S/390-compatible computers was decreasing, while spending on non-S/390-compatible computers was increasing. (IBM Server Market Forecast and Analysis, p. 37, Ex. 53 to Burke Dec.) CDS itself found, based on its telemarketing results, that many of the VM, VSE and MVS S/390 customers on which it focused had "already migrated" or were considering migrating to "other vendors/architectures" as the year 2000 approached. (CDS Internal Presentation, p. 4, Ex. 46 to Burke Dec.) These facts indicate that CDS and IBM were competing in a market that included non-S/390-compatible computers.

C. S/390 Customers Are Not "Locked-In".

CDS responds to this evidence of competition by arguing that the relevant market should be limited to existing IBM S/390 customers. If it is appropriate to limit the relevant product market in this way, then there is no question that IBM has market power in the relevant market. IBM is the world's leading producer of IBM products—just as The Coca-Cola Company is undoubtedly the leading producer of Coca-Cola. But it is an established maxim that antitrust law protects competition, not competitors. Brown Shoe Co. v. United States, 370 U.S. 294, 320, 82 S.Ct. 1502, 8 L.Ed.2d 510 (1962). Thus, courts generally conclude that single brands do not constitute separate markets. Mathias v. Daily News, 152 F.Supp.2d 465, 482-83 (S.D.N.Y.2001) (dismissing antitrust claims based on a market definition limited to the "Daily News" newspaper and excluding other newspapers, and recognizing other cases in the Second Circuit rejecting a relevant market based on a single brand name); Re-Alco Indus., Inc. v. Nat'l Center for Health Educ, Inc., 812 F.Supp. 387, 391-92 (S.D.N.Y.1993) (dismissing antitrust claims when plaintiff failed to make a showing why the relevant market should be limited to a particular brand of health education materials) See also, Town Sound & Custom Tops, Inc. v. Chrysler Motors Corp., 959 F.2d 468, 480 (3d Cir.1992) (rejecting a market definition limited to Chrysler cars, finding that Chrysler cars competed with other companies' automobiles); Domed Stadium Hotel, Inc. v. Holiday Inns, Inc., 732 F.2d 480, 488 (5th Cir.1984) (rejecting market definition limited to Holiday Inn hotel rooms, finding that the relevant market was hotel rooms generally).

CDS's argues, however, that the relevant product market should be limited to existing IBM S/390 customers because they suffer the "lock-in" effect described by the Supreme Court in Eastman Kodak Co. v. Image Technical Services, Inc., et. al, 504 U.S. 451, 112 S.Ct. 2072, 119 L.Ed.2d 265 (1992). In Kodak, several independent service organizations brought an antitrust action against Kodak after it created a new policy that it would sell Kodak replacement parts only to customers who used Kodak maintenance services to maintain their photocopier and micrographic equipment (or maintained their own). Id. at 458, 112 S.Ct. 2072. The Supreme Court, affirming the Ninth Circuit's reversal of the district court's grant of summary judgment in favor of Kodak, found that there were genuine issues of material fact as to Kodak's market power in a properly defined market. Specifically, the Court found that there was a genuine issue as to whether Kodak machines and post-purchase Kodak maintenance services and parts were two distinct products with a "tying arrangement" between them or— as Kodak claimed—one unified market.^[7] Id. at 463, 112 S.Ct. 2072. The Court explicitly recognized that a single brand of a product could constitute a separate market for antitrust purposes, explaining that the relevant market was properly determined by the choices available to Kodak equipment owners, and that "[b]ecause service and parts for Kodak equipment are not interchangeable with other manufacturers' service and parts, the relevant market from the Kodak equipment owner's perspective is composed of only those companies that service Kodak machines." Id, at 482, 112 S.Ct. 2072. The Court noted that customers who had already purchased Kodak equipment would tolerate some level of service-price increases, even supra-competitive prices, if the switching costs remained higher than the service prices. Id. at 477, 112 S.Ct. 2072. If the seller could discriminate between these customers and new customers, the company could charge new customers below-marginal costs on equipment and regain the costs through service and warranty packages not available to the "locked-in" customers. Id. The Court found that the respondents (plaintiffs) had offered evidence of high switching costs, and that Kodak itself had confirmed that it varied the package price of equipment/parts/service for existing customers. Id. The Court also considered the difficulty consumers faced in informing themselves of the total cost of the Kodak "package" including equipment, service, and parts—at the time of purchase, due to the difficulty of calculating accurate lifecycle pricing for a large, complex machine. Id. at 473, 112 S.Ct. 2072. In the context of this potentially limited relevant market (for Kodak parts alone), the Court concluded that Kodak had failed to demonstrate that the respondents' inference of market power was unreasonable, particularly in light of the direct evidence that Kodak had raised prices to drive out competition. Id. at 477, 112 S.Ct. 2072. And it was also plausible from the evidence "to infer that Kodak chose to gain immediate profits by exerting that market power where lockedin customers, high information costs, and discriminatory pricing limited and perhaps eliminated any long-term loss." Id. As a result, the Supreme Court concluded that it was inappropriate for the district court to grant summary judgment in favor of Kodak. Id.

While there is no allegation of an illegal tying arrangement in the instant case, plaintiff argues that here, as in Kodak, summary judgment is not appropriate because there is a genuine issue of fact concerning the existence of a "lock-in" effect. (Plaintiffs Opposition to Summary Judgment ("Plaintiffs Opp."), p. 20.) CDS claims that IBM S/390 customers are locked-in by high switching costs and high information costs and asserts that, "faced with the `commercial realties' of the market, it is apparent that the relevant market at issue is low-end S/390 platforms." Id. at 20.

However, CDS fails to present evidence of any of the elements that the Kodak Court found could indicate "lock-in": (1) high switching costs faced by a substantial number of customers; (2) high information costs faced by a substantial number of customers; and (3) a substantial ability to exploit "ignorant" customers, i.e., those with high information costs. See, generally, Universal Avionics Sys. Corp. v. Rockwell Int'l Corp., 184 F.Supp.2d 947, 956 (D.Ariz.2001) (applying these factors to determine whether plaintiff had established "lock-in" under Kodak). Absent such evidence, CDS does not raise a genuine issue of fact.

1. CDS does not present evidence showing that a substantial number of customers faced "high switching costs".

"Switching costs" are the additional costs (over and above the market price of the item) incurred by customers who have previously invested in a particular product line. See generally, PSI Repair Servs., Inc. v. Honeywell, Inc., 104 F.3d 811, 818 (6th Cir.1997). In Kodak, the respondent plaintiffs submitted evidence of "very high switching costs" for existing Kodak customers. Kodak, 504 U.S. at 477, 112 S.Ct. 2072. There was also evidence that customers used Kodak service "even though they preferred ISO service, [and] Kodak service was of higher price and lower quality than the preferred ISO service" Id at 465, 112 S.Ct. 2072. In the instant case, however, CDS has not provided any evidence of customers who wanted to leave the IBM S/390 system but were unable to because of high switching costs.

In the case of a computer system, the cost of purchasing or programming new software is a potential switching cost. Arguing that locked-in customers face prohibitively high switching costs, CDS cites the Declaration of Robert Fleming ("Fleming Dec"). Mr. Fleming, the President of a library bibliographic supply company, states that his company used a IBM S/390 operating system (or its predecessors) and software applications for twenty years, and that "due to the complexity of the software... it was not possible to migrate the software to another non-IBM platform." (Fleming Dec, ¶ 3, 4.) Unfortunately for CDS, Mr. Fleming goes on to state that eventually his company "successfully migrated" to a CDS operating system, which allows it to run their S/390 applications. (Fleming Dec, ¶ 5.) And Mr. Fleming then reveals that his company is currently "going through a very difficult transition from OS/390 system to a non-IBM network systems," albeit one involving "very high costs for new hardware and very high costs for software." (Fleming Dec, ¶ 6.) While Mr. Fleming's affidavit provides evidence that switching costs may be high and are not incurred lightly, they are apparently not prohibitive, because his company made the switch, despite the costs.

CDS also cites to the Declaration of Ron Hankison ("Hankison Dec"). Mr. Hankison, the President of CDS (now Xbridge Systems, Inc.), opines that a "substantial portion of S/390 customers" are locked-in, and IBM customers are "generally unable to cost-effectively migrate their applications to less expensive computing platforms offered by IBM's competitors." (Hankinson Dec, ¶ 7, 8.) But Mr. Hankison provides no evidentiary basis for his wholly conclusory statements.

In addition to the Fleming and Hankinson Declarations, CDS seeks to support its claim of "high switching costs" through the expert testimony of Ronald Alepin. While Mr. Alepin concludes in his declaration ("Alepin Dec") that, "Customers are unwilling or unable to consider replacing their ... [S/390 systems] because the conversion cost is prohibitive," he offers no examples to support this opinion. (Alepin Dec, ¶11.) Alepin admits that he did not try to identify any S/390 customers who wanted to leave the S/390 platform but were unable to migrate due to high switching costs, or to quantify how many such customers there might be. (Alepin Dep. at 127:5-8, 174:19-175:9.) Likewise, CDS's other expert, Brett Reed, neither identifies any customers that are allegedly "lockedin" nor quantifies how many such customers there are.

In Virgin Atlantic Airways, plaintiff submitted an expert report which purported to explain market facts relevant to the antitrust litigation. The Second Circuit noted that plaintiff had failed to supply the hard data that the expert relied upon, and that the charts and graphs that were offered provided data for an irrelevant time period. Virgin, 257 F.3d at 264. The Circuit concluded that plaintiff failed to meet it burden of showing an actual adverse effect on the relevant market by submitting the speculative report, explaining that, "Expert testimony rooted in hypothetical assumptions cannot substitute for actual market data." Id. Likewise, the speculative affidavits of Mr. Alepin and Mr. Reed do not overcome for the lack of hard evidence. CDS has failed to identify any specific customers, let alone a substantial number, that allegedly faced switching costs of a magnitude sufficient to make migration impractical.

The only concrete evidence regarding the cost of migration in the record comes from IBM documents, in which the defendant estimated the costs to consumers of moving from various platforms. ("PS90E Competitive Analysis," Ex. H to Reed Dec; "Entry S/390 Concept DCP," Ex. I to Reed Dec.) However, these documents do not show that switching costs were prohibitive for any identified customers; rather, they generally suggest that the cost of migration is one factor among others that consumers use to determine the most efficient, beneficial, and economical computing solution for their needs. These documents show nothing more than that defendant was concerned about migrating customers, and sought to retain them by providing a competitive alternative. (See, e.g., Ex. I, p. 11-12, explaining that IBM is seeking to provide "a platform which is priced more competitively with alternate platforms" and noting that "The point is that in order to be successful in our established geographies at replacing the install base, we have to be affordable to the customer, be able to demonstrate cost of computing savings, be an economical platform alternative (including migration costs), provide server function in today's client/server world, and reach the customer.") Even Mr. Alepin acknowledges that switching costs should be considered in light of the benefits of migration. (Alepin Dep. at 127:5-8, 174:19-175:9). See also SMS Sys. Maint. Servs., 188 F.3d at 22 ("cost of shifting to another [computer] system must take into account the efficiency gains of buying new software—gains that often may dwarf hardware price in dollar terms").

The fact that existing IBM customers would need to spend money to migrate to another computing system does not establish "lock-in." See e.g. Geneva Pharm. Tech. Corp. v. Barr Lab., Inc., 201 F Supp.2d 236, 273 (S.D.N.Y.2002) (declining to limit a relevant market to one brand and noting that switching costs alone do not establish lock-in under Kodak) (citations omitted). CDS offers no evidence that "high switching costs" made migration to an alternative platform impractical for a substantial number of S/390-compatible customers.

2. CDS does not present evidence showing that "high information costs" prevented a substantial number of customers from making fully informed purchase, decisions.

"Information costs" prevent customers from obtaining relevant information about the costs and benefits of a product at the time customers initially invest in that product. See, e.g., PSI Repair Servs., 104 F.3d at 818; see generally, Universal Avionics, at 956-57 (to show absence of high information costs, "`[i]t suffices that most buyers know the risks of dealing initially with the defendant rather than with his rivals'") (quoting 10 Areeda et al, Antitrust Law § 1740, at 157 (1996)); Little Caesar Enters., 34 F.Supp.2d at 488-89 (central to the inquiry is "the degree of knowledge ... reasonably available to the buyer[s]" when they initially invest, quoting Areeda, § 1740, at 151). CDS could have supported its allegation of a limited relevant market by showing that a substantial number of existing S/390 customers were unable to make fully informed purchase decisions regarding the costs and benefits of investing in S/390-compatible computers. However, CDS presents no evidence of high information costs.

Again, there is no record evidence that any S/390 customer was unable to make a fully informed decision regarding the attendant costs and benefits of owning, or switching from, an S/390 platform computer, let alone that a substantial number of customers was unable to do so. Unlike defendant Kodak, IBM did not change any policies related to its product after purchase. And there are no allegations in the instant case that customers who bought S/390 systems had difficulty calculating things like lifecycle pricing involving service and warranties, thus preventing them from making an informed decision about what product to purchase.

CDS argues that IBM took action "equivalent to Kodak's change in parts policy" when it "hindered CDS." (Reed Dec, 18). This argument is based on the theory, advanced by Mr. Reed, that, "Customers who selected IBM in the 1980s or early 1990s (or before) could have reasonably expected that new low-cost IBM S/390-compatible technologies made available in 1997 would be made available without undue interference from IBM." This argument not only relies on an unsupported divination of what IBM customers expected; it also begs the ultimate question of whether IBM committed antitrust violations. Mr. Reed's statement does not provide evidence that supports his conclusory claim of undue inference.

Courts generally consider the viability of Kodak-based limited markets in cases where there is an allegation of an illegal tying arrangement, as there was in Kodak. In such cases, if there is no evidence of "information costs," federal courts have routinely granted summary judgment against plaintiffs basing claims on alleged "lock-in" markets. See SMS Sys. Maint. Servs., 188 F.3d at 23 (rejecting "lock-in" market in part because computer users know beforehand that their investment in a particular computer architecture will have certain consequences); PSI Repair Servs., 104 F.3d at 820 (holding that a plaintiff "cannot succeed on a Kodak-type theory when the defendant has been otherwise forthcoming about its pricing structure and service policies."); see also, United Farmers Ass'n, Inc. v. Farmers Ins. Exck, 89 F.3d 233, 238 (5th Cir.1996) (rejecting claim that insurance agents were "locked-in" to particular insurance company because the agents "would clearly have become aware of [the alleged anticompetitive] policy long before they faced significant switching costs"); Universal Avionics, 2001 WL 1771185, at 956-57, 959-60 (finding that manufacturers "know that it is costly to switch" avionics suppliers and that "it would be plainly contrary to the 'economic reality of the market at issue' to assume that [defendant's] customers did not know what they were doing" when they initially invested in defendant's proprietary flight control system (quoting Kodak, 504 U.S. at 467, 112 S.Ct. at 2082)); Laserworks v. Pitney Bowes, Inc., Civ. No. C2-96-1307, 1999 WL 33435671, at *8 (S.D.Ohio Dec.29, 1999) (rejecting "lock-in" market premised on customers' investment in Pitney Bowes fax machines because customers "were armed with all of the information they needed to make an informed decision" in the foremarket and Pitney Bowes policy was not changed after customers invested in the technology) affd 8 Fed.Appx. 380, 2001 WL 392045 (6th Cir. 2001); Metzler v. Bear Auto. Serv. Equip. Co., 19 F.Supp.2d 1345, 1359 (S.D.Fla. 1998) (rejecting "lock-in" market based on customers' investment in particular automotive diagnostic equipment where customers were not faced with a "sudden restrictive change in policy" after they had invested in the technology); Wilson v. Mobil Oil Corp., 984 F.Supp. 450, 461 (E.D.La.1997) (rejecting franchise "lockin" because investors "had sufficient information to evaluate the risks of entering [Defendant's] franchise before they signed their agreements"). Here, I am faced with an effort to extend Kodak beyond the illegal tying arena, with no evidence in the record about excessive information costs. What the First and Sixth Circuits did in SMS and PSI is equally appropriate here.

3. CDS does not present evidence that IBM exploited "locked-in" customers.

IBM argues that, even if the record contained evidence of high switching costs and information costs, IBM could not exploit these customers because they could not identify them, as they had no access to the information necessary to conduct a cost/benefit analysis of switching for each individual customer. CDS counters that the "locked-in"—and thus exploitableS/390 customers can be identified simply by the fact that they have remained customers. This argument rests on the plaintiffs wholly unsupported theory that "If customers were not locked-in or otherwise reliant on the S/390 platform, all IBM customers would have switched to Unix and NT." (Reed Dec, ¶ 20.)

CDS claims that IBM exploited these locked-in customers through price discrimination. In Kodak, respondents presented evidence that Kodak practiced price discrimination by selling parts to customers who serviced their own equipment (most likely high-volume customers who were more likely to have invested the time to gather information on the complicated pricing calculations), but refusing to sell parts to customers that hired third-party service companies. Id at 476, 112 S.Ct. 2072.

There is no similar evidence in the instant case of any type of illicit price discrimination. CDS, however, attributes a nefarious purpose to IBM's tiered pricing index, claiming that it demonstrates "IBM's ability to impose higher prices on its customer base." (Plaintiffs Opp., p. 5.) Yet CDS itself admits the rationale behind the group pricing structure: "... IBM is able to charge different prices for the same OS/390 software platform depending on the MIPS level of the CPU in the system hardware." Id. In plain English, IBM charged higher prices for software in computers with higher processing power. This is hardly evidence of exploitive pricing of locked-in customers.

* * * * * *

In sum, CDS provides no evidence to support its claim that IBM S/390 customers are locked-in in the same manner the Supreme Court thought Kodak photocopier and micrograph equipment customers might be locked-in. To the contrary, as discussed below, there is evidence in the record that S/390 customers migrated from S/390 platforms to other operating systems. CDS does not dispute this migration away from S/390 systems, but argues that, while "some IBM customers are leaving the IBM platform (emphasis in original), . .. these customers are leaving the proper relevant market, not defining a new market. These customers would not stay with IBM S/390 products if IBM reduced prices a small but significant amount, and thus the fact these customers have migrated away from the IBM platform does not demonstrate that the market must be defined as broadly as claimed by IBM." (Reed Aff, ¶ 12.) CDS offers no factual support for its claim that IBM customers would leave the S/390 market if the prices were reduced. Moreover, CDS's argument now becomes in essence: "All S/390 customers want to migrate to other operating systems, but they can't, and when they do, it just means that they shouldn't be counted." This rather hyperbolic theory is unsupported by evidence. It is not rational to assume that every single person who ever bought an IBM S/390 System would prefer to switch to another vendor.

I note that CDS repeatedly stresses that these computing platforms were not "interchangeable" because the S/390 products were not functionally interchangeable with products based on different operating systems. {See e.g., Plaintiffs Opp., p. 4.) But "interchangeability" in antitrust analysis does not refer only to products can be seamlessly swapped for each other. Within the same relevant market, competing products will have various benefits and detriments, including factors such as price and functionality, and customers will chose the product that best suits their overall needs. Only "reasonable interchangeability" is required. See generally E.I. olu Pont de Nemours, 351 U.S. at 404, 76 S.Ct. 994 ("[t]he market is composed of products that have reasonable interchangeability for the purposes for which they are producedprice, use and qualities considered.")

It is true that, due to the lack of functional interchangeability, customers who have invested resources in S/390 products face switching costs if they chose to migrate to another system. It may be that some of these customers are "locked-in" to their investment in the colloquial sense.^[8] But this does not mean that they are victims of a "lock-in" effect caused by anticompetitive behavior under Kodak. And the record evidence of migration off the system—especially given CDS's failure to identify even one specific customer who wants to leave the system and cannot— undermines CDS's theory that customers are "locked-in" by conversion costs in any meaningful sense.

A theory similar to CDS's was rejected by the First Circuit in SMS Sys. Maintenance Serv., Inc. v. Digital Equipment Corp., 188 F.3d 11 (1st Cir.1999). In SMS, an independent computer service (repair/maintenance) provider ("SMS") brought antitrust claims against a computer manufacturer ("DEC") based on an alleged attempt by the manufacturer to monopolize the aftermarket for service of its computers by providing a mandatory warranty. Id. The plaintiffs argued that current DEC customers were "locked-in." Id at 21. The Court found that there was a "serious shortcoming that permeat[ed] SMS's lock-in argument" because "being part of an installed base and being lockedin are not synonymous." Id.

The evidence offered by the plaintiffs in SMS—and the evidence that was conspicuously not offered—was similar to the evidence in and not in the record of the instant case. SMS's "most direct evidence of lock-in" came from employees of DEC customer companies, who testified that, once a company committed to a certain brand of hardware and acquired ancillary software, it would be costly to migrate. Id. However, the Court found that the evidence in the record showed that the decisionmaking surrounding the purchase of computers is qualitatively different from the decision to buy copier parts, and that a large range of factors enter into the computer user's decision most significantly, the availability of desirable software applications. Id at 22. Thus, "the cost of shifting to another system must take into account the efficiency gains of buying new software" which may be economical, "even if one considers the switching costs that are associated with retraining employees and discarding software designed to run exclusively on a particular platform." Id. And in an uncanny parallel to this case, the SMS Court noted that the very witnesses on whom SMS relied to establish the costs of migration undermined the lock-in argument through their actual behavior, because their firms were in the process of switching some of their systems to other platforms. Id at 23.

The First Circuit concluded that the record did not support the conclusion that a substantial number of installed base customers were locked-in, noting that SMS should have provided empirical evidence of the actual behavior of the installed base. Id. The Court noted, "Although we do not doubt that some customers may experience the lock-in that SMS envisions, the record affords no reason to believe that such customers are numerous or that they are representative of the installed base." Id. The Court found it significant that, in contrast to Kodak, there was no evidence of exploitation of customers in the installed base through price discrimination or other suspicious practices. Id at 24. The Court found the conclusions of SMS's expert "highly suspect," noting that he did not conduct a customer survey, but rather based his opinion on his interpretation of certain internal DEC documents. Id at 25.

Although SMS, like Kodak, involved an analysis of "lock-in" in the context of a derivative aftermarket, the similarities to the instant case are obvious. In SMS, after noting that "a lock-in phenomenon must be shown, not assumed," the First Circuit found that "the evidence, taken most favorably to SMS, does not demonstrate a lock-in, certainly not one that raises antitrust concerns." Id at 21, 23. Likewise, in the instant case, I find that CDS has not demonstrated that IBM S/390 customers suffer from an antitrust "lock-in" effect.

D. IBM Lacks Power in the Relevant Market Containing S/390 Platform Computers and Alternate Platform Computers.

Market power is the ability: "(1) to price substantially above the competitive level and (2) to persist in doing so for a significant period without erosion by new entry or expansion." AD/SAT, 181 F.3d at 227 (internal quotes omitted)(emphasis in original); see also CDC Techs, v. IDEXX Labs., Inc., 186 F.3d 74, 81 (2d Cir.1999) (defining market power as the "ability to raise price significantly above competitive level without losing all of one's business" (internal quotes omitted)). This power may be demonstrated by direct evidence of specific conduct indicating the defendant's ability to control price or exclude competition. See CDC Techs., 186 F.3d at 81; K.M.B. Warehouse Distribs., Inc. v. Walker Mfg. Co., 61 F.3d 123, 129 (2d Cir.1995). In the absence of such direct evidence, a firm's share in the relevant market may be used as a proxy for determining power. See CDC Techs., 186 F.3d at 81; K.M.B. Warehouse, 61 F.3d at 129. The record evidence here plainly demonstrates that IBM lacks power—however measured—in a properly defined market.

First, there is no evidence in this record that IBM had the ability to set prices at a supracompetitive level, or to exclude competition. The evidence shows just the opposite—IBM was engaged in competitive pricing. According to CDS, the price of S/390-compatible computer processing was "dropp[ing] significantly," falling 75%, from September 1995 to July 1998. (See CDS Private Placement Memorandum, September 1995, p. 11. Ex. 33 to Burke Dec. (cost approximately $20,000/MIPS); CDS Private Placement Memorandum, July 1998, p. 14, Ex. 32 to Burke Dec. (cost approximately $5,000/MIPS); see also IBM Large Systems Customer I/T Panel Survey, Fall 1998, p. 25, Ex. 50 to Burke Dec. (IBM confirmation of price decrease).) Thus, customers were getting more S/390-compatible processing power for less money. In addition (and as discussed more fully below), Sun Microsystems, Hewlett Packard, and Compaq, to name a few competitors, significantly increased their market shares during the relevant time frame at the expense of IBM. (Expert Report of Carl Shapiro, Table 4, Ex. 80 to Burke Dec.) These trends are entirely inconsistent with the exercise of market power by IBM.

Second, the evidence shows that IBM had a relatively low, and declining, market share. This is true regardless of whether the market includes all computer systems available in the world or only those computers at the "low end" of the market.^[9] IBM submitted evidence showing that, with respect to all computer systems, IBM's share of sales based on integrated hardware was 23% in 2000. (Expert Report of Carl Shapiro, p. 29, Ex. 80 to Burke Dec.) IBM's share based on operating system software was 20% in 2000. Id. Both of these shares with respect to all computer systems declined from 1996 to 2000. Id.

IBM also submitted evidence regarding the "low end" of the market, which it defined by price bracket rather than processing power. The evidence shows that, based on operating system software, IBM's market share for computer systems priced at $250,000 or less (which includes the IBM and CDS systems at issue here) declined from 20% in 1996 to 8% in 2000.^[10] Id. at Table 3. Based on integrated hardware, IBM's market share for systems priced at $250,000 or less declined from 23% in 1996 to 14% in 2000. Id. at Table 4. Moreover, as of the year 2000, IBM ranked fourth (based on integrated hardware), behind Compaq (21%), Sun Microsystems (18%) and Hewlett-Packard (16%). Id. CDS does not provide contradictory evidence or dispute the validity of IBM's data in its summary judgment papers.

These low market share numbers are dispositive. "Courts have consistently held that firms with market shares of less than 30% are presumptively incapable of exercising market power." Union Carbide Corp. v. Montell N.V., 27 F.Supp.2d 414, 417 (S.D.N.Y.1998) (Section 1 claim); see also Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 26-27, 104 S.Ct. 1551, 1566, 80 L.Ed.2d 2 (1984) (Section 1 claim; finding that 30% market share does not constitute "dominant market position"); AD/SAT, 181 F.3d at 229 (Section 2 claim; "33% market share does not approach the level required for a showing of dangerous probability of monopoly power"); Twin Labs., Inc. v. Weider Health & Fitness, 900 F.2d 566, 570 (2d Cir.1990) (same); Hayden Co. v. Siemens Med. Sys., Inc., 879 F.2d 1005, 1017-18 (2d Cir.1989) (Section 2 claim; finding that 20% market share is insufficient to establish "dangerous probability"); Valley Prods. Co. v. Landmark, 128 F.3d 398, 402 n. 3 (6th Cir.1997) (Section 1 claim; noting that federal courts "have repeatedly held that a 30% market share is insufficient to confer... market power"); State of New York v. Anheuser-Busch, Inc., 811 F.Supp. 848, 873 (E.D.N.Y.1993) (Section 1 claim; finding, in context of vertical restraint, that "39% share is below that which has been deemed sufficient to confer market power in any previous decision").

The fact that IBM's share was declining—while spending on non-S/390-compatible computers was increasing—is further evidence that IBM did not possess market power. See Nifty Foods Corp. v. Great All. & Pac. Tea Co., 614 F.2d 832, 841 (2d Cir. 1980); United States v. Syufy Enters., 903 F.2d 659, 666 (9th Cir.1990); Winter Hill Frozen Foods and Servs., Inc. v. Haagen-Dazs Co., 691 F.Supp. 539, 547-48 (D.Mass.1988).

CDS does not dispute these facts, instead relying on its narrowly defined market of "existing IBM S/390 customers" and arguing that IBM holds market power in that market. Again, I have no doubt that IBM does, in fact, possess market power in a market limited to its own products. However, the properly defined relevant market includes products with which the relevant IBM products compete; and in that market, there is no question that IBM lacked market power.

IV. IBM IS ENTITLED TO SUMMARY JUDGMENT ON CDS's CLAYTON ACT SECTION 3 CLAIM (COUNT IV).

To establish a violation of Section 3 of the Clayton Act, CDS must prove: (1) sales of goods for use, consumption or resale within the United States; (2) sales conditioned on the purchasers not dealing in the goods of a competitor; and (3) a potential effect or substantial lessening of competition. See Reisner v. Gen. Motors Corp., 511 F.Supp. 1167, 1176 (S.D.N.Y. 1981); see also Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 328, 81 S.Ct. 623, 628, 5 L.Ed.2d 580 (1961) (competition foreclosed "must be found to constitute a substantial share of the relevant market"). CDS cannot prove elements (2) and (3).

Even if IBM's Business Partner relationships were deemed "exclusive" for purposes of the Clayton Act, a substantial foreclosure of competition cannot be found where, as in this case, numerous alternative channels of distribution existed. See Omega Envtl. v. Gilbarco, Inc., 127 F.3d 1157, 1163 (9th Cir.1997). IBM points to evidence that CDS entered into a number of marketing, distribution and sales agreements with numerous business partners of its own. (Def. SMF ¶¶ 51-52.) In a private placement memorandum, CDS described its marketing approach to "use multiple channels depending on the system type and application," and explained that "a primary channel will be the computer service companies to include Amdahl, Fujitsu, and Decision One." (Private Placement Memorandum, p. 5, Ex. 32 to Burke Dec.) Indeed, CDS described Decision One, as "the largest independent service provider in the USA" Id. Ron Hanikson stated in his deposition that CDS had "several working arrangements" with independent service providers to distribute their products, including "major software firms" with whom it would "be difficult for IBM itself to partner". (Hankison Dep. 61:19-22, Ex. 10A to Burke Dec.) CDS employees also provided evidence that they advertised to VARs, distributors, and other potential customers through trade shows and other direct methods, including telemarketing, by which it completed sales of its products. (Johnson Dep. at 56:16-57:22, Ex. 12A to Burke Dec; Holstrom Dep. at 220:7-13, Ex. 11B to Burke Dec.)

The Ninth Circuit has held that where the plaintiff could have sold its product directly to end-users or used distribution channels other than those used by defendant, defendant did not, as a matter of law, violate Section 3 of the Clayton Act:

If competitors can reach the ultimate consumers of the product by employing existing or potential alternative channels of distribution, it is unclear whether such restrictions foreclose competition from any part of the relevant market.... Competitors are free to sell directly, to develop alternative distributors, or to compete for the services of existing distributors. Antitrust law requires no more.

Omega Envtl, 127 F.3d at 1163 (citations omitted)(emphasis in original); see also CDC Techs., Inc. v. IDEXX Labs., Inc., 7 F.Supp.2d 119, 120 (D.Conn.1998) affd on other grounds 186 F.3d 74 (2d Cir.1999) (granting summary judgment on Clayton Act Section 3 claim where plaintiff could have availed itself of "existing, or potential alternate channels of distribution" and reached consumers through direct sales, non-distribution sources and distributors that had not contracted with defendant). Moreover, the IBM Business Partner agreements were terminable by either party "without cause on three months' written notice." ("IBM Business Partner Agreements for Solution Providers," p. 5, Ex. 58 to Burke Dec.) Thus, CDS was free to compete with IBM for distribution/reseller arrangements. See Paddock Pubis., Inc. v. Chicago Tribune Co., 103 F.3d 42, 45 (7th Cir.1996) ("Competition-for-the-contract is a form of competition that antitrust laws protect rather than proscribe ... A termination clause works just like a stated time limit in facilitating competition for the contract."); see also Omega Envtl, 127 F.3d at 1163 (one year duration of, and 60 day terminability clause in, agreements "negate substantially their potential to foreclose competition") accord CDC Techs., 7 F.Supp.2d at 129.

CDS counters this evidence by arguing that IBM interfered with its relationship with Information Technology Company ("ITC"). (Plaintiffs Opp. at 14.) CDS claims that this alleged interference creates a genuine issue of material fact as to the existence of a Clayton Act violation, because CDS "relied upon [VARs]" and ITC was "one of the most important VARs." While CDS does not explicitly argue that IBM's alleged interference with its relationship with Intelliware also supports a Clayton Act violation, I construe the complaint in its favor to include this allegation. Even so, CDS's allegations of interference with these two VARs fails to create a genuine issue of fact as to whether IBM violated the Clayton Act.

CDS attempts to support its argument by citing to the Hanikson Declaration, in which he claims that "[t]here were not very many VARs" and characterizes ITC as "one of the most important." (Hanikson Dec, ¶ 15.) This statement directly contradicts Mr. Hanikson's deposition testimony, where he estimated that there were "in the hundreds, one hundred and fifty" VARs selling S/390 computers. (Hanikson Dep. 72:17-22, Ex. 10A to Burke Dec.) It is well settled that a party cannot create a disputed issue of fact by submitting an affidavit that is at variance with his deposition testimony. Mack v. U.S., 814 F.2d 120, 124 (2d Cir.1987). Moreover, CDS admits that it entered into marketing, distribution and sales agreements with at least eleven other companies. (CDS's Responses to Defendant's First Set of Requests for Admissions, 92-102.) And among two dozen firms that CDS strategy documents identified as prospective or desirable business partners, neither ITC nor Intelliware is mentioned. (CDS Sales Strategy and Implementation Plan, p. 11, Ex. 31 to Burke Dec.) There is absolutely no record evidence that any specific VAR was particularly "influential" within the distribution/reseller community. And CDS cannot refute the availability of its other options by arguing that certain firms did not have the same "proven finances, abilities and customer relationships" as other resellers. See Omega Envtl., 127 F.3d at 1163 ("[T]he antitrust laws were not designed to equip the plaintiffs' hypothetical competitor with [defendant's] legitimate competitive advantage.").

Thus, even though I am denying summary judgment to IBM on CDS's claim that IBM interfered with its relationship with ITC and IntelliWare, there is no evidence that any such interference could have had the potential effect or resulted in a substantial lessening of competition. At the motion to dismiss stage, I allowed CDS's antitrust claims to proceed because they alleged that ITC and Intelliware were such "influential" VARs that IBM's interference with them had the effect of substantially foreclosing CDS from the relevant market. CDS III, 2002 WL 1205740 at * 7. CDS has failed to provide any evidence to support its claim that these VARs were particularly "influential."

Because the record evidence establishes that CDS had available, and availed itself of, numerous distribution channels, no reasonable trier of fact could find a substantial foreclosure of competition. IBM is entitled to summary judgment on CDS's Section 3 Clayton Act claim (Count IV).

V. IBM IS ENTITLED TO SUMMARY JUDGMENT ON CDS's SHERMAN ACT SECTION 1 CLAIMS (COUNTS II-III).

A violation of Section 1 of the Sherman Act occurs when there is (1) a combination or some form of concerted action between at least two legally distinct economic entities, and (2) such combination or conduct constituted an unreasonable restraint of trade under the rule of reason.^[11]See Tops Markets Inc. v. Quality Markets, Inc., 142 F.3d 90, 95-96 (2d Cir.1998). Under the rule of reason, the plaintiff must show that the challenged conduct had an actual adverse effect on competition in the market, not just that it harmed competition. Id at 96.

CDS does not claim, and there is absolutely no evidence of, any horizontal agreements between any two or more distributors regarding the distribution of CDS products. Rather, CDS alleges discrete vertical agreements between IBM and its business partners. (Plaintiffs Sec. Amd. Cmplt, Counts II & III.) In its Motion in Opposition to IBM's Motion for Summary Judgment, CDS argues that summary judgment is inappropriate as to the Sherman Act claims because it has presented evidence of arrangements between IBM and two individual Business Partner VARsITC and Intelliware. (Plaintiffs Opp., p. 12.) Regarding these two companies, CDS claims that "these initial potential customers were extremely important to CDS's business plan, and were essential to building momentum in the market place." Id. By allegedly causing ITC and Intelliware to stop dealing with CDS, CDS claims that IBM "effectively shut[ ] [them] out of the market." Id.

The allegations underlying CDS's Sherman Act Section 1 claims are thus essentially the same as those underlying its Clayton Act Section 3 claim. The Supreme Court has held that if conduct does not fall within "the broader proscription of [Section] 3 of the Clayton Act", it is not forbidden by Sections 1 or 2 of the Sherman Act. Tampa Elec. Co., 365 U.S. at 335, 81 S.Ct. at 632; see also CDC Techs., 7 F.Supp.2d at 122. For substantially the same reasons that no violation under Section 3 of the Clayton Act exists on these allegations—there is no evidence of a relationship which caused substantial foreclosure in the market—no violation exists under Section 1 of the Sherman Act.

However, even if IBM had not prevailed on the Clayton Act claim, it would be clear that CDS had not established a violation under Section 1 of the Sherman Act. CDS provides no evidence of an unreasonable restraint of trade under the rule of reason, and thus fails to establish the second element of a Section 1 Sherman Act claim. Unreasonable restraint of trade under the rule of reason requires "an actual adverse effect on competition as a whole in the relevant market". Virgin Atlantic Airways, 257 F.3d at 264 (internal quotes omitted). The "overarching standard is whether defendant's actions diminish overall competition, and hence consumer welfare." K.M.B. Warehouse, 61 F.3d at 128 (internal quotes omitted).

The traditional indicators of an adverse effect—reduced output, increased price, or decreased quality—are entirely absent in the record evidence. See Virgin Atlantic Airways, 257 F.3d at 264. Indeed, as discussed above (see Part III. D, supra), prices for S/390 computing capacity decreased drastically (by 75%) during the relevant time frame—a trend that is inconsistent with anticompetitive behavior.

Lacking direct evidence of an adverse effect, CDS would have to show not only that IBM possessed market power, but also that "other grounds [existed] to believe that the defendant's behavior will harm competition market-wide such as the inherent anti-competitive nature of defendant's behavior or the structure of the interbrand market." Clorox, 117 F.3d at 59 quoting K.M.B. Warehouse, 61 F.3d at 129. IBM plainly does not possess market power in a properly defined market. (See Part III. D, supra.) And even if it did, there is no evidence in the record of "other grounds" to believe that IBM's actions could harm market wide competition. As noted above, there is no evidence that, even if IBM did interfere with CDS's relationships with ITC and Intelliware, its actions harmed market-wide competition. (See Section III, supra.) Rather, the evidence shows that CDS had numerous distribution channels available to it and used them. See Id.; Ryko Mfg. Co. v. Eden Servs., 823 F.2d 1215, 1234 (8th Cir.1987) (no Section 1 violation where plaintiff failed to show that other means of promoting and selling its product were unavailable). As IBM points out, there is nothing anticompetitive about the fact that some of IBM's Business Partners, among a much larger population of available VARs and other distribution channels, elected not to do business with CDS. See CDC Techs., 7 F.Supp.2d at 129 ("[A] distributor's reluctance to establish ties to a reasonably small company rather than a company with a strong reputation ... is not surprising or anticompetitive. In fact, it is the essence of competition." (internal quotes omitted)). And the evidence shows that there were multiple vigorous competitors in the relevant market with shares greater than IBM; these competitors' shares increased at the expense of IBM during the relevant time frame. See Part III. B; K.M.B. Warehouse, 61 F.3d at 129 (no "other grounds" to substantiate competitive harm where evidence demonstrated "healthy" interbrand competition).

CDS also argues that IBM saw CDS as a "threat" and "targeted the CDS S/390 with their own product," the "aptly named Piranha (subsequently renamed Planter)." (Plaintiffs Opp., p. 12.) Again, CDS attempts to support a claim not with evidence, but with nefarious characterization of legitimate behavior. Developing a new product in an attempt to compete with another company's product is not evidence of anticompetitive behavior. It is, rather, the essence of competition.

For all of these reasons, IBM is entitled to summary judgment on CDS's Section 1 claims (Counts II—III).

VI. IBM IS ENTITLED TO SUMMARY JUDGMENT ON CDS's SECTION 2 MONOPOLIZATION CLAIM (COUNT V).

The first issue with Count V to decipher is what is alleged. The claim is entitled "Attempt to Monopolize," but it also contains allegations that IBM's conduct constituted both an attempt to monopolize and a conspiracy to monopolize. (Plaintiffs Sec. Amd. Cmplt., 11126.) In its Prayer for Relief, CDS asks that the Court find a violation of Section 2 of the Sherman Act based on "monopolization, attempt to monopolize and conspiracy to monopolize." Id., p. 48. IBM suggests in its motion for summary judgment that the references to conspiracy and monopolization claims are drafting errors left over from previous complaints, and focuses in its motion on the "attempt to monopolize" allegation. CDS responds by arguing both that IBM attempted to monopolize the market and that its attempts were successful- a monopolization claim. CDS does not advance any arguments to support a "conspiracy to monopolize" claim, presumably abandoning the theory because there is no record evidence to support it.

To establish attempted monopolization, CDS must prove: (1) that IBM engaged in predatory or anticompetitive conduct; (2) with a specific intent to monopolize; and (3) there was a dangerous probability of achieving monopoly power in a relevant market. See Spectrum Sports, Inc. v. McQuillan, 506 U.S. 447, 456, 113 S.Ct. 884, 890-91, 122 L.Ed.2d 247 (1993); accord Virgin Atlantic Airways, 257 F.3d at 266. To establish monopolization, CDS must prove: (1) possession of monopoly power in a relevant market; and (2) willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident. See Volvo N. Am. Corp. v. Men's Int'l Prof I Tennis Council, 857 F.2d 55, 73 (2d Cir.1988). In other words, to prove monopolization under the Sherman Act, plaintiff must show both that defendant has a monopoly and that defendant willfully engaged in anticompetitive conduct in the relevant market, Invamed, Inc. v. Barr Laboratories, 22 F.Supp.2d 210, 218 (S.D.N.Y.1998).

CDS fails to establish essential elements of both attempted monopolization and monopolization.

A. CDS Has Not Raised an Issue of Material Fact Concerning Either IBM's Possession of Monopoly Power in a Relevant Market or a Dangerous Probability of Achieving Monopoly Power.

CDS has not shown that IBM possessed monopoly power in the properly defined relevant market. See Part III, supra. CDS thus fails to prove the first element of a monopolization claim. Furthermore, to demonstrate a dangerous probability of achieving monopoly power, CDS must prove that IBM possessed significant market power in the relevant market. See Walsh Trucking Co., 812 F.2d 786, 791 (2d Cir.1987). As established in Part III D, supra, IBM did not have such market power. Accordingly, the attempted monopolization claim must also be rejected. See AD/SAT, 181 F.3d at 229; Twin Labs., 900 F.2d at 570-71.

B. CDS Cannot Establish Anticompetitive Conduct or Acts of Willful Acquisition or Maintenance of Monopoly Power.

In the context of discussing its attempted monopolization claim, CDS argues that IBM engaged in anticompetitive conduct by using its power in the market to force ITC and Intelliware to refuse to deal with CDS. (Plaintiffs Opp., p. 13). In the context of discussing its monopolization claim, CDS makes a general statement that there is "ample" record evidence that precludes summary judgment, including evidence of "(1) IBM's economic coercion and threats to ITC and Intelliware, which shut CDS out of the market; ... (2) IBM's prohibition of honest marketing claims by CDS; ... (3) raising CDS's software prices, to the detriment of consumers and to protect the locked-in market... and (4) preannouncing a product to chill the market for the CDS product, and then dropping that product once CDS left the market, again to the detriment of consumers." (Id., p. 14.) CDS's allegations, if proven, could be used to support either a monopolization or an attempted monopolization claim, as both require proof of anticompetitive behavior on the part of the defendant, Invamed, 22 F.Supp.2d at 218. However, CDS fails to support any of these allegations with evidence raising a genuine issue of fact, and thus fails to sustain a valid claim under either theory.

1. The Allegations Regarding ITC and Intelliware Do Not Support a Claim Under Section 2 of the Sherman Act.

The alleged interference by IBM with the ITC and the Intelliware relationships was the unsuccessful basis of the Section 3 claim. (Section III, supra.) The Supreme Court has held that if conduct does not fall within "the broader proscription of [Section] 3 of the Clayton Act", it is not forbidden by Sections 1 or 2 of the Sherman Act. Tampa Elec. Co., 365 U.S. at 335, 81 S.Ct. at 632; see also CDC Techs., 7 F.Supp.2d at 122. CDS failed to show that IBM's alleged interference with these VARs had an anticompetitive effect on the market, and thus failed to support its allegation of a violation under Section 3 of the Clayton Act. This alleged interference thus does not establish a violation under Section 2 of the Sherman Act.

2. There is No Evidence that IBM's Actions Regarding ESL Pricing Were Anticompetitive.

CDS's second and third accusations of anticompetitive conduct relate to IBM's actions regarding ESL pricing. CDS claims that IBM took "direct action" against it by "refusing to provide ESLpriced software to CDS except under anticompetitive terms and conditions, and by limiting marketing claims that CDS could make regarding its product." (Plaintiffs Opp., p. 9 citing Hankison Dec, ¶13.) IBM counters that all system manufacturers were held to the same standard—they were permitted to offer their customers ESL pricing on IBM software if the manufacturers' performance claims were consistent with IBM's uniform criteria for ESL pricing. (Defendant's Mem. in Support of Summary Judgment, p. 20.)

The record evidence supports IBM's contention that the "ESL" pricing level was offered by IBM to CDS based on explicit performance claims about its products. Under the OPA signed by both IBM and CDS, it was clearly stated by IBM that "Based on your claims regarding your System's performance as compared to other IBM S/390 Systems IBM will assign your System the appropriate S/390 group software price. The lowest possible S/390 software price for Systems using P/390E is ESL." (See OPA Agreement, Ex. 2, § 5.2, Ex. 23 to the Burke Dec.) The OEM Agreement stated that: "Buyer agrees and understands that IBM has classified the Buyer System as an Entry Level System (ESL). Buyer further agrees that IBM, in its sole discretion, may change this classification which may result in changes to the Eligible Program Price." ("Amendment to the OEM Agreement," Ex. 1, § 1.3, Ex. 23 to the Burke Dec).

In addition to the explicit terms of the contract, there is record evidence showing that CDS was reminded of the guidelines for ESL pricing several times. On October 16, 1997, an IBM employee sent an email to Ron Hankison explaining that, "The class of system that CDS compares itself to will dictate the price paid for the software used on the CDS system. Comparison to Entry Level class systems will assure that the software price is ESL." (Email to Ron Hankison; Ex. 68 to the Burke Dec). This email indicates that CDS was permitted to decide what marketing claims it wanted to make about its product, but that it might face higher software prices for claims of higher performance. Id. ("Basically CDS can make whatever claims they can support by their own benchmarks and testing... Marketing claims that clearly are defining a system that has attributes beyond that of an Entry Level system could raise the question of whether this is an Entry Level system.") In an earlier email, another IBM employee had provided Ron Hankison with the same message—IBM provided ESL group software pricing only for Entry Level systems. (See Email to Ron Hankison and attached Facsimile, Ex. 66 & 67 to Burke Dec.) ("Systems that use the P/390EE and are defined and compared to other Entry Systems will qualify for ESL software pricing... Systems, either IBM or OEM, that use P/390EE but are defined as and compared to other ESA level IBM Systems will be assigned the software group prices for that class of System.")

When CDS introduced its new product, however, it chose to align itself with an IBM system that was not an entry level system. Instead of naming its product the "CDS-O," as originally planned, CDS named its offering the "CDS-104." Ron Hankison acknowledged that, in doing so, CDS was attempting to position its product to resemble IBM's existing "Multiprise-104" S/390 computer. (Hankison Dep., 28:23-29:10, Ex. 9 to Burke Dec.) But IBM's Multiprise-104 was a more powerful computer than the entry system level computers, and it did not qualify for ESL pricing. (Leahy Dep., 115:11-20, Ex. 13C to Burke Dec.) (explaining that the Multiprise product family had the full I/O subsystem, wasn't subject to the I/O limitations [producing greater thoroughput], and thus did not qualify for ESL pricing.) IBM also claims that CDS made performance claims for its computer that were outside the guidelines allowable for an Entry Level system. (Defendant's Motion for Summary Judgment, p. 20.)

IBM thus informed CDS that, based on its performance claims, the CDS system was no longer qualified for ESL classification, and assigned the system to a higher software pricing group. (Letter to CDS from IBM, Ex. 71 to Burke Dec.) However, IBM and CDS subsequently reached an agreement under which CDS was able to once again receive ESL software prices. (Hankison Dep., 29:14-23, Ex. 9 to Burke Dec.) Pursuant to this agreement, CDS changed the name of its product from "CDS-104" to "CDS-2000." Id. The following year, IBM qualified the CDS2000e system for ESL software pricing. (Letter from IBM to Kevin Murphy, arch 13, 1998, Ex. 72 to Burke Dec.)

CDS has provided no evidence to support its allegation that IBM's actions regarding ESL pricing were retaliatory and anticompetitive. To the contrary, the record evidence shows that IBM was acting within its rights under the OEM agreement and the OPA. The record evidence also shows that CDS customers had to pay for IBM operating system software based on the performance claims made by CDS, but there is nothing that indicates CDS was prohibited from accurately describing its products. There is certainly no evidence in this record that IBM allowed customers other than CDS to take advantage of ESL pricing while advertising products outside the allowable guidelines. Rather, it appears that IBM imposed the same software pricing guidelines on itself that it imposed on CDS. (Leahy Dep., 115:11-20, Ex. 13C to Burke Dec; Email to Ron Hankison and attached Facsimile, Ex. 66 & 67 to Burke Dec.) CDS thus fails to provide any evidence which would raise a genuine issue of material fact as to whether IBM's decisions regarding CDS's qualification for ESL pricing were anticompetitive.

3. There is No Evidence that the May 7, 1998 Announcement Was Anticompetive.

In May 1998, IBM announced plans to offer a new S/390 Integrated Server in the Fourth Quarter of 1998. ("Preview: IBM S/390 Integrated Server," Ex. 77 to Burke Dec.) At the end of the announcement, IBM, in a "Statement of Direction," stated that in 1999, "S/390 will announce and deliver a processor with 2X performance of the processor shipped in fourth quarter 1998." Id. Immediately after this statement, IBM noted that the notice was "subject to change or withdrawal without notice" and represented IBM's "goals and objectives only." Id

To base an antitrust claim on IBM's "pre-announcement" of the Integrated Server—or the statement of direction regarding the two-times performance follow-on product—CDS must prove that the announcement was "knowingly false" at the time it was made. See AD/SAT, 181 F.3d at 231; accord MCI Comm. Corp. v. Am. Tel. & Tel. Co., 708 F.2d 1081, 1130 (7th Cir.1983) (announcement not anticompetitive absent "deception or knowing falsehood"). That is because "reasonable good faith statements about research, development, and forthcoming production serve the social interest in maximizing the relevant information available to buyers". MCI Comm., 708 F.2d at 1128 (internal quotes omitted).

There is no evidence suggesting that IBM's "pre-announcement" was anything but a "reasonable good faith statement." To the contrary, the internal IBM documents regarding the product—which CDS suggests show anticompetitive behavior— actually demonstrate that IBM was working strenuously to develop and price the product, in an effort to get it out on the market to compete with the CDS product. (See e.g., Internal IBM email, 3/30/98, Ex. L to Reed Dec. ("I am charged with pushing this forward to give it the opportunity to succeed. If the DBB decides today otherwise than we'll reset the plan... but if they go for success we will be pushing very hard over the next few weeks for the SOD"); Internal IBM email, 7/30/97, Ex. 12 to Williams Dec. (discussing a plan to respond to the CDS product by "counter[ing] with our PS390/R390 offerings, which are even more cost effective in this segment. Regardless of the specifics behind this, CDS is getting good press, and we need to counterattack. We have the offerings, but not a good marketing strategy at this time" and noting that Piranaha (Planter) is the product to "go[ ] after CDS full bore with.").) Once again, CDS's "evidence" of anticompetitive conduct is actually evidence of the very competition that antitrust laws are designed to protect.

Moreover, IBM did offer an Integrated Server in November, 1998, as pre-announced. (CDS's Second Amend. Cmplt. 181.) It is hard to characterize a statement as "knowingly false" when it turns out to be true. And in September 1999, IBM released the "Multiprise 3000" servers, S/390 products that provided more than twice the performance of the Integrated Server. ("S/390 Multiprise 3000 Server" Ex. 79 to Burke Dec.) CDS expert Brett Reed claims that the product IBM ultimately introduced was not the product pre-announced in May 1998, because the smallest of the Multiprise servers was 60 MIPS, which was "five times larger than the 2X performance enhancement suggested in the May 1998 preannouncement." (Reed Dec. 1127.) The fact that IBM released a more powerful processor than originally planned, at a time when processing power was becoming cheaper, is not evidence that IBM's original announcement was made in bad faith.

Mr. Reed suggests that the pre-announcement was not a "reasonable good faith statement" because internal emails showed concern that the development team was "overworked" and that it was too early for the announcement. (Reed Dec. 127, citing Internal IBM email, 3/30/98, Ex. L to Reed Dec.) The emails to which Reed refers were written five weeks before the announcement was made. They discuss what needed to happen within IBM, and the development team, to announce the integrated server and go forward with the statement of direction successfully. While one email indicates that the "team" is being "driven hard" to "get to announce readiness," there is absolutely no evidence in any of the emails of a bad faith decision to announce a product that IBM did not intend to offer. And, as noted above, IBM released the product.

CDS does not submit any evidence to support its nefarious interpretation of IBM's decision to stop offering the S/390 Integrated Server in February 2000. None of the documents in evidence that discuss the development and introduction of the Integrated Server suggest that IBM planned to "drop" the product once CDS had left the market. But the record evidence does suggest a reason why both the IBM and the CDS products failed. Ron Hankison explains that the CDS S/390 products were withdrawn from the market due to "declining sales and dwindling cash balances." (Hankinson Dec, 1113.) While Hankinson attributes CDS's lack of success in the S/390 market to the anticompetitive conduct of IBM, the record evidence shows that sales of all S/390 products were decreasing at this time— and that customers were increasingly turning to non-S/390 platform computers and non-S/390 compatible software applications. See Section III B & D, supra. The rational inference to be drawn is that as the demand for S/390 products declined, both IBM and CDS experienced decreased sales. In CDS's case, this resulted from a withdrawal from the market, and for IBM, it lead to a reduction in their S/390 product offerings. CDS's contrary assertion that the removal of the IBM offering from the market was anticompetitive is based solely on the conclusory allegations of its own witnesses, and is not supported by any record evidence.

CDS fails to provide any evidence which would raise a genuine issue of fact as to whether the May 1998 IBM announcement was anticompetitive.

* * * * * *

No reasonable trier of fact could conclude that CDS has established the essential elements of its monopolization or attempted monopolization claims. IBM is entitled to summary judgment on Count V.

VII. IBM IS ENTITLED TO SUMMARY JUDGMENT ON CDS's NEW YORK GENERAL BUSINESS LAW SECTION 340 CLAIM (COUNT VI).

Section 340 of the New York General Business Law, known as the Donnelly Act, is modeled after the Sherman Act and should be construed in light of federal precedent. See X.L.O. Concrete Corp. v. Rivergate Corp., 83 N.Y.2d 513, 518, 611 N.Y.S.2d 786, 789, 634 N.E.2d 158 (1994). For the same reasons that CDS has not raised critical issues of fact on its federal Sherman Act claim, CDS cannot establish a violation of the Donnelly Act, and IBM is entitled to summary judgment on Count VI. See, e.g., Empire Volkswagen, Inc. v. World-Wide Volkswagen Corp., 814 F.2d 90, 98 n. 4 (2d Cir. 1987).

VIII. IBM IS NOT ENTITLED TO SUMMARY JUDGMENT ON CDS's CLAIM FOR TORTIOUS INTERFERENCE WITH PROSPECTIVE BUSINESS RELATIONS (COUNT VII).

To establish a claim for tortious interference with prospective business relations under New York law, CDS must prove: (1) business relations with a third party; (2) IBM's interference with those business relations; (3) that IBM acted with the sole purpose of harming CDS or used dishonest, unfair or improper means; and (4) injury to the business relationship. See Nadel v. Play-by-Play Toys & Novelties, Inc., 208 F.3d 368, 382 (2d Cir.2000).

Along with its Opposition to Summary Judgment, CDS produced the Declaration of Stan King. ("King Dec") Mr. King, President of ITC, states that he was threatened by an IBM employee. IBM concedes in its Reply Memorandum that the King Declaration raises a genuine issue of material fact as to whether IBM interfered with CDS's relation with ITC. While IBM makes a cogent argument that this claim should be dismissed insofar as it deals with Intelliware, I decline to do so on motion. While CDS has not offered similar direct evidence of covert threats by IBM against Intelliware, a trier of fact confronted with the evidence about ITC could view IBM's contacts with Intelliware in a different light. Therefore, the claim of tortious interference will go to trial as to both customers. Because there is diversity between the parties, the Court retains jurisdiction over this matter under 28 U.S.C. § 1332(a).

IX. IBM IS ENTITLED TO SUMMARY JUDGMENT ON ITS CLAIM FOR BREACH OF THE ORIGINAL PROVIDER AGREEMENT.

To prevail on its claims for breach of the OPA under New York Law, IBM must prove: (1) a valid contract; (2) performance of the contract by IBM; (3) breach of the contract by CDS; and (4) damages. See Terwilliger v. Terwilliger, 206 F.3d 240, 245-46 (2d Cir.2000).

IBM argues that CDS has admitted all the elements of IBM's claim: the OPA constituted a valid and binding contract (CDS's Answer to IBM's Counterclaims f 33); IBM has fully performed its obligations under the OPA (CDS's Answer to IBM's Counterclaims 1(40) ("CDS admits that IBM created and wrote IBM's copyrighted code in the Eligible Programs"); and CDS never paid IBM $155,000 in principal due and owing under the OPA (CDS's Response to IBM's First Set of RFAs, No. 39.). As a result of CDS's failure to make these payments, IBM claims that it has suffered damages in the principal amount of $155,000, plus interest.

CDS does not dispute the factual basis for IBM's claim, and instead pleads affirmative defenses based on IBM's alleged misconduct: fraud, constructive fraud, acquiescence, inequitable conduct, unclean hands, breach of covenant of good faith and fair dealing, estoppel, and laches. (CDS's Reply to the Amended Answer and Counterclaims of IBM, ¶ s 48-56.) Conflating all these different theories, it asks the court to exercise its equitable power and bar IBM from recovering against CDS because the antitrust violations alleged by CDS "demonstrate that IBM has breached the law and its contract with CDS." (Plaintiffs Opp., p. 24.)

Unfortunately for CDS, I have dismissed its antitrust claims.^[12] The tortious interference claim that remains bears no relation to the OPA between IBM and CDS. Accordingly, there are no genuine issues of material fact to be tried, and IBM is entitled to summary judgment for liability on its counter-claim.

As this case will be going to trial on the tortious interference claims, I will hear evidence regarding the appropriate interest calculations for this claim at that time.

CONCLUSION

This constitutes the decision and order of the Court.

NOTES

[1] CDS referred to their "second" opportunity to compete with IBM, as these three individuals had previously founded another computer company, Amdahl Corp.

[2] The term "S/390 operating system software" is used to include the OS/390 system as well as compatible predecessor system software, including the MVS, VM/ESA, and VSE/ ESA operating system software.

[3] No doubt IBM did not proffer the Richardson and Hummel Declarations (or equivalent information) along with the original motion papers because it did not anticipate CDS's motion. In this Court's experience, it is unprecedented to have reputable counsel (or any counsel, for that matter) challenge the authenticity of deposition transcripts, discovery responses, and documents produced from his client's files.

[4] At the relevant time, Mr. Burke was associated with Cravath Swain and Moore, the law firm representing defendant.

[5] The relevant geographic market is not in dispute. CDS pleads, and IBM agrees, that the relevant geographic market is worldwide. (Plaintiff's Sec. Amd. Cmplt., 1173; Defendant's Mem. in Support of Summary Judgment, p. 4.)

[6] The testimony of both IBM and CDS employees suggests that one factor behind this substantial migration was that computer software, not hardware, drove customers' purchase decisions. (Holstom Dep. 79: 21-25, Ex. 11A to Burke Dec; Port Dep. 204:23-205:8, Ex. 17C to Burke Dec.) Since 1997, more applications were being written for UNIX and Windows NT than for S/390. (CDS's Response to IBM's Second Set of RFAs, No. 19, Ex. 1 to Burke Dec.)

[7] "A tying arrangement is `an agreement by a party to sell one product but only on the condition that the buyer also purchases a different (or tied) product, or at least agrees that he will not purchase that product from any other supplier.'" Kodak, 504 U.S. at 461, 112 S.Ct. 2072 (quoting Northern Pac. Ry. Co. v. United States, 356 U.S. 1, 5-6, 78 S.Ct. 514, 2 L.Ed.2d 545 (1958)) A tying arrangement violates Section 1 of the Sherman Act "if the seller has `appreciable economic power' in the tying product market and if the arrangement affects a substantial volume of commerce in the tied market." Kodak, 504 U.S. at 462, 112 S.Ct. 2072 (quoting Fortner Enter. Inc. v. United States Steel Corp., 394 U.S. 495, 503, 89 S.Ct. 1252, 22 L.Ed.2d 495 (1969)).

[8] CDS argues that IBM "acknowledge^]" that IBM customers are locked-in, quoting a document where an IBM employee states. "... we do not need this upstart company [CDS] wedging its way into a marketplace we should have locked-in the low end arena." (Plaintiff's Opp., p. 2, quoting Ex. 12 to Williams Dec.) First, the statement that something "should" be done implies that it has not yet been done. More important, a statement by an employee regarding "locking-in" business or customers is not an admission by the company of the existence of a "lock-in" effect for antitrust purposes.

[9] The record evidence does not clearly establish that there is a definable "low end" market within the computer systems market. There are some IBM documents that discuss the "low end" of the market (usually defining the low end in terms of processing power). This distinction may be one that IBM makes for marketing and strategic planning purposes. Or it may indicate that there is a distinct product market at the lower cost/lower processing power end of the market. However, there is no concrete data that suggests that the relevant product market must be subdivided in such a way, and it is unnecessary to hypothesize about whether such line-drawing is appropriate. The evidence presented shows that IBM had a relatively small, and declining, market share of both the computing system market as a whole and at the "low end" specifically.

[10] IBM's expert, Carl Shapiro, defined the "low end" by price bracket, rather than by processing power, and analyzes the data within that framework. The data in Tables 3 and 4 is for all computers in the $250,000 or less price bracket (defined by Shapiro as the Entry and Midrange Systems), and includes both IBM and CDS Systems.

[11] As noted above, this Court previously dismissed Count I of CDS's complaint, which alleged a Sherman Act § 1 violation for per se unreasonable restraint of trade.

[12] CDS submits no evidence that IBM breached its contract with CDS. CDS originally asserted a breach of contract claim against IBM, but chose not to replead it after Judge Stanton dismissed it.