|This document is available in two formats: this web page (for browsing content) and PDF (comparable to original document formatting). To view the PDF you will need Acrobat Reader, which may be downloaded from the Adobe site. For an official signed copy, please contact the Antitrust Documents Group.|
IN THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
PLAINTIFFS' JOINT REPLY
TABLE OF CONTENTS
TABLE OF AUTHORITIES
American Professional Testing Service, Inc. v. Harcourt Brace Jovanovich Legal and Professional Publications, Inc., 108 F.3d 1147 (9th Cir. 1997)
Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585 (1985)
Ball Mem. Hosp., Inc. v. Mutual Hosp. Ins., Inc., 784 F.2d 1325 (7th Cir. 1986)
Barry Wright Corp. v. ITT Grinnell Corp., 724 F.2d 227 (1st Cir. 1983)
Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263 (2d Cir. 1979)
Broadcast Music Inc. v. CBS, 441 U.S. 1 (1979)
Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209 (1993)
Corsearch, Inc. v. Thomson & Thomson, 792 F. Supp. 305 (S.D.N.Y. 1992)
Data General Corp. v. Grumman Sys. Support Corp., 36 F.3d 1147 (1st Cir. 1994)
Dial A Car, Inc. v. Transportation, Inc., 82 F.3d 484 (D.C. Cir. 1996)
Directory Sales Management Corp. v. Ohio Bell Tel. Co., 833 F.2d 606 (6th Cir. 1987)
Eastman Kodak Co. v. Image Technical Services, Inc., 504 U.S. 451 (1992)
Forsyth v. Humana, Inc., 114 F.3d 1467 (9th Cir. 1997), aff'd, 525 U.S. 299 (1999)
GAF Corp. v. Eastman Kodak Co., 519 F. Supp. 1203 (S.D.N.Y. 1981)
Gamma Audio & Video, Inc. v. Ean-Chea, 11 F.3d 1106 (1st Cir. 1993)
Gilliam v. ABC, 538 F.2d 14 (2d Cir. 1976)
In re IBM Peripherals EDP Devices Antiturst Litigation, 481 F. Supp. 965 (N.D. Cal. 1979), aff'd, 671 F.2d 1377 (9th Cir. 1983)
ILC Peripherals Leasing Corp. v. IBM, 458 F. Supp. 423 (N.D. Cal. 1978), aff'd sub nom. Memorex Corp. v. IBM, 636 F.2d 1188 (9th Cir. 1980)
Image Technical Servs., Inc. v. Eastman Kodak Co., 125 F.3d 1195 (9th Cir. 1997)
Indiana Grocery, Inc. v. Super Valu Stores, Inc., 864 F.2d 1409 (7th Cir. 1989)
International Boxing Club v. United States, 358 U.S. 242 (1959)
Jack Walters & Sons Corp. v. Morton Bldg, Inc., 737 F.2d 698 (7th Cir. 1984)
Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2 (1984)
Lee v. A.R.T., 125 F.3d 580 (7th Cir. 1997)
Lotus Development Corp v. Borland Intern., Inc., 49 F.3d 807 (1st Cir. 1995), aff'd by equally divided court, 516 U.S. 233 (1996)
LucasArts Entertainment Co. v. Humongous Entertainment Co., 870 F. Supp. 285 (N.D. Cal. 1993)
Matthew Bender & Co., Inc. v. West Pub. Co., 158 F.3d 674 (2d Cir. 1998), cert. denied, 119 S. Ct. 2039 (1999)
Multistate Legal Studies, Inc. v. Harcourt Brace Jovanovich Legal and Professional Publications, Inc., 63 F.3d 1540 (10th Cir. 1995)
National Bank of Commerce v. Shaklee Corp., 503 F. Supp. 533 (W.D. Tex. 1980)
Neumann v. Reinforced Earth Co., 786 F.2d 424 (D.C. Cir. 1986)
NLFC, Inc. v. Devcom Mid-America, Inc., 45 F.3d 231 (7th Cir. 1995)
Omega Environmental, Inc. v. Gilbarco, Inc., 127 F.3d 1157 (9th Cir. 1997), cert. denied, 119 S. Ct. 46 (1998)
Paramount Pictures v. Video Broadcasting Systems, 724 F. Supp. 808 (D. Kan. 1989)
Premier Elec. Constr. Co. v. National Elec. Contr. Ass'n, Inc., 814 F.2d 358 (7th Cir. 1987)
Reazin v. Blue Cross & Blue Shield of Kansas, Inc., 899 F.2d 951 (10th Cir. 1990)
Response of Carolina, Inc. v. Leaseco Response, Inc., 537 F.2d 1307 (5th Cir. 1976)
Richter Concrete Corp. v. Hilltop Concrete Corp., 691 F.2d 818 (6th Cir. 1982)
S.O.S., Inc. v. Payday, Inc., 886 F.2d 1081 (9th Cir. 1989)
Sony Corporation of America v. Universal City Studios, Inc., 464 U.S. 417 (1984)
Southern Pac. Communications Co. v. AT&T, 740 F.2d 980 (D.C. Cir. 1984)
Spectrum Sports, Inc. v. McQuillan, 506 U.S. 447 (1993)
Times-Picayune Publishing Co. v. United States, 345 U.S. 594 (1953)
Twentieth Century Music Corp. v. Aiken, 422 U.S. 151 (1973)
U.S. Healthcare, Inc. v. Healthsource, Inc., 986 F.2d 589 (1st Cir. 1993)
United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377 (1956)
United States v. General Dynamics Corp., 415 U.S. 486 (1974)
United States v. Loew's, 371 U.S. 38 (1962)
United States v. Microsoft Corp., 147 F.3d 935 (D.C. Cir. 1998)
United States v. Microsoft Corp., 1998 WL 614485 (D.D.C.1998)
United States v. Paramount Pictures, 334 U.S. 131 (1948)
United States v. Syufy Enters., 903 F.2d 659 (9th Cir. 1990)
Walt Disney Co. v. Powell, 897 F.2d 565 (D.C. Cir. 1990)
WGN Continental Broadcasting Co. v. United Video, Inc., 693 F.2d 622 (7th Cir. 1982)
Wilson v. Volkswagen of Am., Inc., 561 F.2d 494 (4th Cir. 1977)
III P. Areeda & H. Hovenkamp, Antitrust Law (1996)
P. Areeda & H. Hovenkamp, Antitrust Law (Supp. 1999)
2 Nimmer on Copyright
17 U.S.C. § 101
17 U.S.C. § 06
17 U.S.C. § 111(a)(3)
17 U.S.C. § 501(a)
37 C.F.R. § 202.3(b)(3)(A)
Microsoft treats as nearly an afterthought both the Court's core finding that Microsoft has monopoly power and the unifying theme of the Court's numerous findings on Microsoft's conduct: Microsoft fought a multi-front campaign, using a broad array of anticompetitive tactics that reduced rather than enhanced consumer choice, to sustain the critical barrier to entry protecting its monopoly power. Microsoft cannot avoid the comprehensive Section 2 claim for maintenance of monopoly, however, by relegating its discussion to the back of its brief and discussing the Section 1 issues (and the attempted monopoly issue) first and without regard to its monopoly power. Microsoft's approach not only disregards and fragments this Court's findings but improperly evades the substantive importance of the finding of monopoly power, protected by a powerful barrier to entry, to showing the anticompetitive character of its conduct. See Eastman Kodak Co. v. Image Technical Services, Inc., 504 U.S. 451, 488 (1992) (Scalia, J., dissenting) ("Behavior that might otherwise not be of concern to the antitrust laws -- or that might even be viewed as procompetitive -- can take on exclusionary connotations when practiced by a monopolist.").
When it finally addresses the maintenance-of-monopoly claim, Microsoft accepts the legal standards laid down by the Supreme Court. It has nothing of substance to say to undermine the showing made in plaintiffs' proposed conclusions that this Court's findings compel a holding of liability under those well-established standards. Instead, Microsoft addresses strawmen and seeks to apply to this case out-of-context passages from decisions involving patently different market circumstances.
Microsoft takes a similar tack in discussing the other issues in the case and repeatedly misstates the applicable legal standards in order to avoid the legal implications of the Court's findings. On tying, for example, Microsoft purports to rely on the D.C. Circuit's consent-decree decision but ignores the "bolting" principle of that decision, which is dispositive under this Court's findings. On the multiple exclusionary agreements with Internet access providers, online service providers, content providers, PC manufacturers, and others, Microsoft not only disregards those aspects of the agreements which this Court found went beyond legitimate interests and unjustifiably constricted competition, but relies on a legal standard that would allow all hobbling of competition that did not completely destroy it. And on the attempt claim, Microsoft ignores the effects it already has achieved in the browser market, including its dominant position and the intended stifling of innovation, creating a dangerous probability of monopoly power. In short, Microsoft should be held liable not only on the comprehensive Section 2 maintenance-of-monopoly claim, but on the two Section 1 claims and the Section 2 attempt claim as well.
Plaintiffs' proposed conclusions show that this Court's findings establish a violation of Section 2 under black-letter principles of monopolization law. In particular, this Court made legally correct findings of a relevant market and of Microsoft's monopoly power in that market -- both of which are issues of fact (see Eastman Kodak, 504 U.S. at 481; International Boxing Club v. United States, 358 U.S. 242, 245, 251 (1959); United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 381 (1956)) -- based on meticulous consideration of all of the evidence and arguments put forward by Microsoft to deny those facts. Pl. Conc. at 3-14. The Court also examined Microsoft's conduct and carefully identified a wide range of actions by Microsoft that harmed consumer choice on the merits (by stifling the middleware threat to the applications barrier to entry into the operating system market) and, therefore, were anticompetitive under the established standards for unlawful monopolizing conduct. Id. at 14-53. Microsoft fails to identify any basis for denying the plain legal consequence of this Court's findings: that Microsoft unlawfully maintained a monopoly in violation of Section 2.
A. Market Definition. Microsoft briefly challenges the market definition -- operating systems for Intel-compatible personal computers -- as too narrow as a matter of law. MS Conc. at 46-49. But Microsoft does no more than identify the very "competitive threats" (id. at 46), such as Apple products or network computers and middleware, that this Court carefully examined and found not to be in the same market under well-established standards for market definition. The Court found that none of these "threats" was a meaningful substitute for Intel-based PC operating systems, or would readily enough generate meaningful substitutes by redeployment of assets used for other purposes, sufficient to deprive a hypothetical monopolist of Intel-based PC operating systems of power over price. ¶¶ 18-32.
Unable to dispute these findings as a factual matter, Microsoft must resort to hinting at the legal proposition that any firm selling anything that might take some of the defendant's customers away, or that offers a tool for lowering barriers to competing with the defendant, must be in the same market. Microsoft cannot cite any law supporting such a view. The extent of substitutability is undisputedly critical to defining the market. See Pl. Conc. at 3-4.(1) And a firm offering a complement that relaxes entry barriers, while thus facilitating competition, is not thereby offering a competitive substitute. Id. at 6. Middleware itself is thus not properly included in the market because it must be used with an operating system and is not a substitute for an operating system. Indeed, Microsoft's assertion that a "successful vendor of middleware . . . [might] extend its platform 'downward' by adding device drivers and fully displacing Windows" (MS Conc. at 49) acknowledges that middleware is not itself an operating system or a substitute for Windows; it is this hypothetical "downward" extension software, which actually operates a PC, that would be in the operating-system market, not the middleware (which no consumer could purchase to operate a PC) from which it descended.(2)
B. Monopoly Power. Microsoft takes a series of potshots at the express finding of fact that it "enjoys monopoly power in a relevant market" (¶ 33), but nothing it says shows any legal error underlying that finding. MS Conc. at 49-54. Microsoft begins by saying that monopoly power "must be durable," quoting a decision that actually upheld a finding of monopoly power (Reazin v. Blue Cross & Blue Shield of Kansas, Inc., 899 F.2d 951, 968 (10th Cir. 1990)), but any requirement of durability is undoubtedly satisfied here. This Court found a monopoly persisting over at least a decade in which there was a "sustained absence of realistic commercial alternatives to Microsoft's PC operating-system products." ¶ 53; see ¶ 35 ("persistent").
Microsoft contends that market shares are not legally conclusive (MS Conc. at 49-50), but this Court devoted only one paragraph to the extraordinarily high market-share figures (¶ 35) and did not accord those indisputably material facts any legally conclusive status; instead, the Court analyzed all the evidence of monopoly power. Microsoft seeks to deny its monopoly power by suggesting that it could not control price and output. MS Conc. at 50, 51. But as Microsoft itself acknowledges (id. at 50, citing ¶¶ 33, 62-65), this Court expressly found that Microsoft had precisely the power to set the price of Windows "substantially above" a competitive level "for a significant period of time without losing an unacceptable amount of business to competitors." ¶ 33. The Court's findings amply demonstrate that Microsoft has repeatedly exercised the power to restrict consumer choice, and thus output, without effective discipline from competitors. ¶ 411.
Ignoring the particular features of the relevant market in this case, Microsoft suggests that no power could properly be found here because this market may lack certain structural features that have been important in confirming the absence of monopoly power in other markets. MS Conc. at 50, 51. Thus, Microsoft relies on a decision that pointed to the lack of "control of the productive assets in the business" to reject a claim of monopoly power (Indiana Grocery, Inc. v. Super Valu Stores, Inc., 864 F.2d 1409, 1414 (7th Cir. 1989)). But Indiana Grocery involved grocery stores, for which physical assets are critical. It did not concern software, or markets like those at issue here, in which network effects, sunk costs, and economies of scale (which create the applications barrier to entry here), as well as the need for huge front-end investments, constrain actual or potential competitors. Microsoft's reliance on sentence fragments from United States v. Syufy Enters., 903 F.2d 659, 66-67 (9th Cir. 1990), and Ball Mem. Hosp., Inc. v. Mutual Hosp. Ins., Inc., 784 F.2d 1325, 1335 (7th Cir. 1986), which note the absence of particular kinds of entry barriers in those cases, is similarly misplaced. Those decisions do not deny, and could not have denied, the importance of the factors that create the particular entry barrier here.
Microsoft suggests that no entry barrier exists because there is no "disadvantage that new entrants face but incumbents [here, Microsoft] do not." MS Conc. at 52. That is not the governing legal standard, which sensibly requires only that there be factors significant enough to enable the defendant profitably to charge supracompetitive prices while deterring outsiders from entering. See Pl. Conc. at 8.(3) Moreover, the standard urged by Microsoft is in any event met by this Court's findings. The Court found that Microsoft did not originally face the "highly penetrated market dominated by a single competitor" (¶ 43); that today it "does not face any obstacles nearly as imposing as the barrier to entry that vendors and would-be vendors of other PC operating systems must overcome" (¶ 44); and that Microsoft has an unmistakable advantage over new operating-system vendors based on, among other things, its massive installed base of Windows operating systems for which vast numbers of applications have been written and the prohibitive costs of adequate cloning of Windows APIs. See ¶¶ 36-52.
Finally, Microsoft notes that "[t]here is no finding that Microsoft ever did charge a supra-competitive price for Windows." MS Conc. at 53. But that observation would not undermine the finding of monopoly power even if it were less misleading in its incompleteness. The Court found that Microsoft's power over price was confirmed by its evident lack of concern over rivals' prices and its internally recognized vast range of discretion over price. ¶¶ 62, 63. As to the price Microsoft actually charged, the Court said that it could not "determine with any level of confidence" that the price was below a monopolist's short-term profit-maximizing price. ¶ 65. And it explained that there were evident reasons, wholly consistent with Microsoft's power over price, why Microsoft may have charged "what seems like a low short-term price." ¶ 65; see ¶ 66.
C. Monopolizing Conduct. Microsoft expressly accepts (MS Conc. at 54) the formulation for finding monopolizing conduct that originated with Professor Areeda. See Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585, 605 n.32 (1985); Southern Pac. Communications Co. v. AT&T, 740 F.2d 980, 999 n.19 (D.C. Cir. 1984); Barry Wright Corp. v. ITT Grinnell Corp., 724 F.2d 227, 230 (1st Cir. 1983) (Breyer, J.). This Court carefully sifted through Microsoft's conduct, assessed its effects on consumer choices, and made detailed findings as to the anticompetitive character of numerous acts and agreements of Microsoft. Those findings, as plaintiffs' conclusions explain, establish a violation under the Section 2 monopolization standard, whether under the Areeda formulation or the variant stressing the lack of any business sense to the conduct except as a means of securing monopoly power.(4) Microsoft's attacks on these findings provide no escape from the conclusion that this is a case "in which a defendant's possession of substantial market power, combined with his exclusionary or anticompetitive behavior, threaten[ed] to defeat or forestall the corrective forces of competition and thereby sustain or extend the defendant's agglomeration of power." Eastman Kodak, 504 U.S. at 488 (Scalia, J., dissenting).
1. Microsoft's initial argument is limited to defending "the integrated design of Windows 98" (excluding all the other forms of binding Internet Explorer to Windows). MS Conc. at 57-59. Microsoft first asserts that any rejection of the Section 1 tying claim would defeat Section 2 liability (id. at 57), but that assertion is manifestly incorrect, as this Court already recognized at the summary judgment stage. United States v. Microsoft Corp., 1998 WL 614485, at *23 (1998). The standards of the per se Section 1 tying rule are quite different from the governing standards for monopolizing (anticompetitive, exclusionary, predatory) conduct (see Eastman Kodak, 504 U.S. at 480-82 (treating monopolization and tying independently)), and basic antitrust policy supports the distinction. Judicial caution in construing and applying the Section 1 tying rule has reflected two basic facts about that rule: It applies to firms with only market power, not requiring monopoly power; and it is a per se rule that condemns a tying arrangement without a full assessment of its competitive effects and justifications. The reasons for concern about a Section 1 standard that may ensnare too much conduct do not extend to Section 2, which applies only to firms with monopoly power (whose actions are more likely to harm competition) and which, even then, requires proof of anticompetitive conduct (rather than relying on a per se rule). Indeed, the availability of ordinary Section 2 standards has been a critical basis for urging caution in the interpretation and application of Section 1 per se tying standards. P. Areeda & H. Hovenkamp, Antitrust Law ¶ 1746.1b, at 494 (Supp. 1999).
On the merits, Microsoft asserts that the integration of Internet Explorer into Windows 98 resulted in "improvements to the operating system." MS Conc. at 57. This assertion is either irrelevant or wrong. If Microsoft merely means that some benefits arise from Internet Explorer and the operating system working together in the ways they do in Windows 98, the point is irrelevant, as are the decisions cited by Microsoft for the proposition that product improvement does not itself violate Section 2. MS Conc. at 57.(5) The pertinent question is the benefit of denying consumers a desired choice by offering only the commingled Windows 98 -- whether, in other words, the benefits of system integration can be achieved only by hard-wiring Internet Explorer into Windows 98. This Court answered that question with its findings that forcing consumers to take Internet Explorer in Windows 98 had significant costs for some consumers (¶¶ 170-174, 184-185, 410) and "[n]o consumer benefit" (¶ 186). See ¶¶ 177 ("no technical justification for . . . refusal to meet consumer demand for a browserless version of Windows 98"), 191 ("Microsoft could offer consumers all the benefits of the current Windows 98 package by distributing the products separately and allowing OEMs or consumers themselves to combine the products if they wished.").
Microsoft cites two decisions from other courts urging deferential review of product designs.(6) Strikingly, however, Microsoft does not even refer to the relevant findings, quoted and cited in the preceding paragraph, much less show how these findings depart from deferential review. The Court's findings are sufficient to condemn Microsoft's conduct, not only under general Section 2 standards, but even under the urged deferential review.
2. Microsoft next seeks quickly to defend its panoply of restrictive agreements -- those with OEMs (covering a number of actions that bound Internet Explorer to Windows, as well as control of the desktop and initial boot sequence) and those with OLSs, ISPs, ICPs, and ISVs. MS Conc. at 59-60. Given this Court's findings, however, Microsoft has no alternative but to assert that it was permitted under Section 2 to take a host of steps that substantially impeded Netscape's commercial ability actually to secure customers, as long as Microsoft did not absolutely "prevent Netscape from getting Navigator into the hands of consumers." Id. at 60 (footnote omitted). Microsoft thus asserts that Section 2 does not proscribe actions that serve no legitimate consumer-benefitting interests (Microsoft does not challenge this Court's extensive findings to that effect) and that, in the commercial realities of the marketplace, effectively foreclosed the most important of the channels through which consumers obtained the browsers they actually used (as this Court found, see ¶¶ 357-360, 375-385, 411).(7)
Microsoft's argument relies on some unexplained and seemingly non-economic notion of "prevent," is unsupported by any cited authority, and would license a monopolist to achieve the central evil targeted by Section 2 (keeping rivals' output low enough to permit monopoly pricing) by creating just enough economic difficulties for rivals to discourage enough consumers from dealing with them. Not surprisingly, Microsoft's argument is incorrect. For example, in a decision touted by Microsoft, the Tenth Circuit, reinstating a Section 2 claim and finding anticompetitive conduct even for attempt, pointedly reversed the district court's holding that "to be actionable, the [defendant's conduct] had to make it impossible, rather than merely inconvenient, for [consumers] to take [the rival's bar-review service]." Multistate Legal Studies, Inc. v. Harcourt Brace Jovanovich Legal and Professional Publications, Inc., 63 F.3d 1540, 1553 (10th Cir. 1995). Rather, "[w]hat matters is . . . whether the [conduct] was reasonably capable of contributing significantly to a monopolization attempt by discouraging [defendant's] customers from taking [its rival's service] and, if so, whether it was nevertheless justified." Id.; see id. (plaintiff "presented enough evidence to create a triable issue as to whether this [conduct] could significantly discourage" choice of rival's service). (Emphases added).(8) This Court's findings amply establish that Microsoft's agreements "significantly discourage[d]" consumers from actually adopting Navigator. Multistate Legal Studies, 63 F.3d at 1553 n.12. That is more than enough -- as even Microsoft implicitly recognizes in endorsing (MS Conc. at 59, 54) formulations of the Section 2 standard condemning conduct that "threatens to have a significant exclusionary impact" (U.S. Healthcare, Inc. v. Healthsource, Inc., 986 F.2d 589, 597 (1st Cir. 1993)) or "reasonably appears capable of making a significant contribution to the creation, maintenance, or expansion of monopoly power" (III P. Areeda & H. Hovenkamp, Antitrust Law ¶ 650a, at 67 (1996)). (Emphases added.)(9)
3. In what is little more than a diversion, Microsoft argues at greater length that it had no duty to pre-disclose technical information about Windows 95. MS Conc. at 60-62. Plaintiffs, however, have alleged no such general duty. The pertinent anticompetitive conduct was the sequence of (a) Microsoft's proposal that Netscape conspire to maintain the applications barrier to entry by withdrawing its plan to produce and distribute a browser that exposed APIs, followed by (b) Microsoft's punishing of Netscape, when Netscape declined to collude, by withholding information otherwise routinely made available to developers of software complements. The punishment is significant for two reasons: first, because it corroborates the proof that Microsoft's proposal to Netscape was an anticompetitive scheme to exclude Netscape as a middleware threat and, second, because Microsoft's use of its power to dispense or withhold necessary operating system technical information, in order to induce other firms to collude in maintaining the applications barrier to entry, is an anticompetitive means of maintaining its operating system monopoly. Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263 (2d Cir. 1979), and Microsoft's other authorities do not address the issue of a monopolist's using its unique position to induce collusion and are therefore inapposite.
4. Microsoft argues that the findings do not establish that it engaged in "predatory pricing," as defined in Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209 (1993), that it "priced any product below an appropriate measure of cost," or that it "would be able to charge a supracompetitive price for Web browsing software on a sustained basis." MS Conc. at 62-63. This argument ignores both what the Court found and what plaintiffs allege. The Court specifically found that "price" (revenues received) was below "cost": Given Microsoft's zero price for the stand-alone Internet Explorer and its foregoing of large-scale revenue opportunities, "[e]ven if Microsoft maximized ancillary revenue, the amount of revenue realized would not come close to recouping the cost of its campaign to maximize Internet Explorer's usage share at Navigator's expense." ¶ 142. That finding both satisfies the below-cost requirement of Brooke Group and condemns the campaign under the basic Section 2 standard: It made no business sense except for its role in sustaining the operating-system monopoly.(10)
Microsoft's argument is nothing but a play on different uses of the term "predatory." The point here is not that Microsoft unlawfully lowered prices, as was alleged in Brooke Group, but rather that, given Microsoft's zero price, its costly campaign to distribute its stand-alone browser and exclude rival browsers was unprofitable and made no business sense except as part of an anticompetitive scheme. See Pl. Conc. at 41. In addition, recoupment through a supracompetitive dollar price for the browser after obtaining a new browser monopoly (the sort of recoupment at issue in Brooke Group) was unnecessary (see id. at 43); the rewards to Microsoft from its scheme were found, instead, in the exclusion of a threat to its existing operating system monopoly. The means by which Microsoft used spending and pricing to monopolize are thus critically different from those involved in cases like Brooke Group and do not merit the judicial skepticism about the plausibility of below-cost pricing followed by recoupment through a not-yet-obtained monopoly that informed Brooke Group.
5. Microsoft says that various of its actions, chiefly those not aimed at Netscape, can escape condemnation because they did not achieve the anticompetitive effects that Microsoft was trying to achieve. MS Conc. at 64-67. The standard expressly accepted even by Microsoft, however, requires no more than capability to make a significant contribution to maintaining monopoly. See MS Conc. at 54 (quoting "reasonably appears capable" standard). Under that standard, anticompetitive actions do not lose their unlawful character simply because they fail to bear immediate fruit. Taken together, moreover, the totality of Microsoft's actions did have competition-impairing effects, including the sending of a deterrent message to other potential threats. ¶ 412.
In any event, many of the non-Netscape actions defended by Microsoft clearly impaired competition. For example, Microsoft does not deny that it succeeded in stopping Intel's promotion of its own NSP software and Intel's undertaking of other software initiatives. All Microsoft can say is that its success was so great that it not only stopped Intel from acting as an independent competitive force in the market but actually strengthened its monopoly by acquiring some of the very technology at issue. MS Conc. at 65. That acquisition aggravates, rather than alleviates, the competitive harm.(11)
The actions recounted in the Court's findings, moreover, readily meet the standards for a relevant pattern and practice. The record established in this Court's findings is hardly one of "three or four episodes over a 20-year period" (quoted in MS Conc. at 67 n.40). Rather, it is one of a broad pattern of consistent actions taken during a several-year period, in pursuit of a single objective of preventing development of cross-platform middleware, whenever Microsoft saw a serious threat of that occurring. This Court's findings establish that Microsoft's actions were, indeed, its "regular response to a repeated specific situation." Wilson v. Volkswagen of Am., Inc., 561 F.2d 494, 511 (4th Cir. 1977) (quoted in MS Conc. at 67).
6. Microsoft makes two final arguments against Section 2 liability. First, it argues that the actions at issue "should be separately considered in the context of the evidence as a whole." MS Conc. at 68. Microsoft's point is anything but clear. Microsoft cannot and does not deny that the competition-harming effects of multiple actions are mutually reinforcing or that the Court can properly consider the aggregate effect of its conduct in assessing its lawfulness. See Pl. Conc. at 50-51. Microsoft's objection to adding together several legally deficient theories simply does not apply to any argument plaintiffs have made.
Second, Microsoft appears to contend that the findings do not establish a causal connection between the totality of anticompetitive acts and the maintenance of the Windows monopoly. MS Conc. at 69-70. But Microsoft cannot deny the extensive findings of the debilitating effect on middleware resulting from its numerous actions to extinguish the middleware threat to its monopoly. See Pl. Conc. at 51-53. Microsoft, accordingly, must rely on the unavoidable inability of the Court to know precisely what would have happened if Microsoft had not nipped the middleware threat in the bud and, thus, the Court's inability to find that Navigator or Java would already, by 1999, have "ignited genuine competition" in Microsoft's market. ¶ 411. Microsoft's approach, however, is both inconsistent with the legal standard that Microsoft accepts -- "reasonably appears capable" of maintaining monopoly power -- and would have the perverse effect of permitting executions of potential threats in the crib. That is not the law.(12)
Plaintiffs' proposed conclusions show that Microsoft violated Section 1's prohibition on tying arrangements by its succession of actions -- from pure contractual arrangements soon after Windows 95 was released through partial commingling of files in Windows 95 OEM Service Release 2.0 through the more extensive commingling in Windows 98 -- that compelled consumers purchasing the Windows operating system to take Internet Explorer along with it. These arrangements violate both the tying law standards established in the Supreme Court decisions and, for the later "technological" intermingling arrangements, the D.C. Circuit's "bolting" standard in the consent-decree case (see United States v. Microsoft Corp., 147 F.3d 935 (D.C. Cir. 1998)). In particular, this Court's detailed findings establish the existence of separate products under either legal standard, as well as the other requirements of an unlawful tying arrangement. See Pl. Conc. at 53-62. Microsoft challenges the conclusion of a Section 1 tying violation only with respect to Windows 98. MS Conc. at 2-15.
First, Microsoft argues that, as a matter of law, Windows 98 is one product. Id. at 3-12. Microsoft does not dispute that Windows 98 combines two products under the demand test of Eastman Kodak and Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2 (1984).(13) Instead, it relies exclusively on the D.C. Circuit's suggested deferential two-product standard for technological tying.(14) Remarkably, however, while expounding at length on the reasoning in the D.C. Circuit's opinion, Microsoft ignores the terms of that standard: Technological integration does not create a single product if it is mere "bolting," that is, "commingling for an anticompetitive purpose (or for no purpose at all)"; to create a single product, it must provide "benefits when compared to a purchaser's combination of corresponding stand-alone functionalities." Microsoft, 147 F.3d at 949 & n.12, 950. In ignoring the standard, Microsoft also ignores the clear findings that Windows 98 is not a single product under even that standard; this Court's findings establish that Microsoft commingled files in Windows 98 as it did for anticompetitive reasons and that the commingling provides no consumer benefits compared to stand-alone programs that function on their own but could be combined to achieve the integration benefits if the user wanted them. See Pl. Conc. at 59.(15)
Microsoft points to this Court's language in the summary judgment opinion referring to a comparison of Windows 98 with the functionality of a combination of Windows and "another browser." Microsoft, 1998 WL 614485, at *12; see MS Conc. at 1, 11. But that language is not and cannot be limited to non-Microsoft browsers. It must be understood to refer generically to "a stand-alone browser," or "a stand-alone browser application," to use the language of the D.C. Circuit. Microsoft, 147 F.3d at 950, 952. Indeed, the D.C. Circuit's "bolting" standard specifically refers to combinations of two Microsoft products, not a combination of Windows and a non-Microsoft product. See also 147 F.3d at 948 n.11 (applying integration standard to combination of Microsoft's operating system and "its mouse" (emphasis added)). No other reading makes sense. The lawfulness of a tie, or the determination whether there are two products, cannot sensibly depend on whether any rival has yet designed an equivalent for the particular features the defendant has designed -- especially when the tying arrangements themselves have handicapped rivals in any development efforts they may otherwise have undertaken.
Microsoft quotes several findings to support the assertion that there are certain benefits to the commingling of Internet Explorer and Windows. MS Conc. at 6-8. But this Court's findings plainly say the opposite: There are no benefits to combining a browser with Windows 98 that cannot be achieved through separate distribution of two products. Tellingly, the few benefits Microsoft claims -- spreading a tool for Web access and a set of APIs for software developers to use (id. at 6-7) -- either relate only to Microsoft's offering a bundled version to consumers and have nothing to do with forcing them to take that version (see ¶¶ 186 ("[n]o consumer benefit" from "Microsoft's refusal to offer" an unbundled version), 408) or were rejected by the Court's findings (see ¶ 193). Moreover, the claimed benefits have nothing to do with the argument for the D.C. Circuit's suggested deferential standard in the first place. They are simply benefits of common distribution, which are universal to all ties and would exist even if the operating system and the browser were concededly two products packaged -- or bolted -- together for distribution. These purported benefits have nothing to do with the kinds of technological, engineering considerations that the D.C. Circuit concluded could justify a departure from the Supreme Court's two-product standard for a subclass of "technological" ties.
Second, although it places primary focus on the two-product issue, Microsoft also argues that no "tie" is present because there is no separate charge for Internet Explorer as part of Windows 98. MS Conc. at 12-13. Microsoft's argument would immunize any bundle offered at a single price as long as the tying product in the bundle were not offered separately, thus encouraging the very result the D.C. Circuit suggested could be "affirmatively harmful" to the aims of tying law. See Microsoft, 147 F.3d at 948. Not surprisingly, the argument contradicts basic tying law. As already explained (Pl. Conc. at 60-61), this is not a case like MultiState Legal Services or Directory Sales Management Corp. v. Ohio Bell Tel. Co., 833 F.2d 606 (6th Cir. 1987) -- Microsoft's two cases -- in which consumers were simply given an option whether to take the tied product and incurred no costs in doing so. Here, not only are consumers who do not want Internet Explorer forced to take it nonetheless, incurring costs such as performance degradation, incompatibility with other software, and increased security and instability risks (¶ 173), but rival browser firms face as a result of Microsoft's binding arrangements precisely the sorts of higher costs of gaining customers that tying law condemns: OEMs are affirmatively discouraged from adding Navigator to a PC by the higher costs they must incur once Internet Explorer is already on the machine. ¶¶ 216-217, 239-241. In these circumstances, forcing OEMs and consumers to take the browser creates the very danger of harming competition in the tied product (and the tying product as well) that tying law targets. See Jefferson Parish, 466 U.S. at 14-16. Indeed, it is black-letter law that a "tie" exists, even without a forced taking of the defendant's tied product, as long as there is sufficient economic compulsion not to take rivals' versions of the tied product. See Eastman Kodak, 504 U.S. at 461. That rule flatly bars Microsoft's position.
Third, Microsoft urges that the inclusion of Internet Explorer with Windows 98, even if a tie of separate products, "does not foreclose a substantial amount" of Netscape usage because Microsoft did not totally block OEMs from installing Netscape in addition to Internet Explorer and, as this Court found, "did not actually prevent users from obtaining and using Navigator" (¶ 357). MS Conc. at 14-15. Notably, this argument does not dispute the extent of the commerce lost by Netscape as a result of the tie or the sufficiency of that loss under the minimal standards that govern this part of the tying-law test. See Pl. Conc. at 59-60. Instead, the argument is simply Microsoft's Section 2 "absolute prohibition" contention in Section 1 guise. It has no more merit or precedential support in the present context. Even Microsoft's cited authorities find "foreclosure" when the defendant's tied product has been taken by a customer, simply making it more costly (not impossible) for a rival to secure business with that customer. That is what this Court's findings establish occurred through Microsoft's tying of a browser to its operating system, effectively foreclosing the OEM channel to Netscape as a way to secure actual usage.
Microsoft pursues similar defenses to Section 1 liability for its various restrictive agreements with ISPs, OLSs, and ICPs, which the Court found served no legitimate purpose (¶¶ 291, 308, 401). MS Conc. at 15-25. Although, as noted above, Microsoft in the Section 2 context criticizes formalistic approaches such as market-share rules, its defense against Section 1 liability for these agreements adopts just such an approach. The law, however, eschews such formalism in favor of a rule-of-reason approach which directs inquiry to the particular facts of the individual case. See Pl. Conc. at 63. The Court's findings on the record in this case make clear that Microsoft's agreements harmed competition and are therefore unlawful.
Microsoft first invokes this Court's articulation of a 40% foreclosure figure in discussing Section 1 at the summary judgment stage. See MS Conc. at 17. But the Court's findings establish that, as a result of Microsoft's actions, Netscape was indeed commercially foreclosed from the two channels of browser distribution that account for "a very large majority" of browser adoption. ¶ 144; see ¶¶ 145 ("no other distribution channel for browsing software even approaches the efficiency" of the OEM and IAP channels), 239 ("Microsoft has largely succeeded in exiling Navigator from the crucial OEM distribution channel"), 241, 242, 304, 357. Those findings satisfy even the articulated standard -- which, in any event, should not be understood as any more conclusive of the governing harm-to-competition issue than are market share figures in monopolization or attempted-monopolization cases. Certainly Microsoft cannot disregard market realities, artificially divide its conduct into pieces, and then argue that each piece is legal because, by itself, it does not meet some numerical threshold. The Court's findings unmistakably establish that competition was substantially harmed by Netscape's foreclosure from meaningful access to the important distribution channels. E.g., ¶ 247 (the restrictions "did the four things they were designed to accomplish").
Those findings, made on full consideration of evidence about the market circumstances, answer Microsoft's assertions that restrictions on distributors "are generally less cause for antitrust concern" or that short duration of agreements may reduce their competition-harming effects. Omega Environmental, Inc. v. Gilbarco, Inc., 127 F.3d 1157 (9th Cir. 1997), cert. denied, 119 S. Ct. 46 (1998) (emphasis added); see MS Conc. at 18-22. As the use of the term "generally" itself reveals, neither Omega nor other decisions establishes immunity rules that override a context-specific finding, amply supported by the evidence, that particular distribution channels do not have commercially meaningful substitutes, so that choking off those channels does substantially impair competition. Indeed, Omega -- which involved agreements covering only 120 of 500 distributors and no finding that these were particularly important ones -- carefully tied its conclusion to the particular facts in the case. 127 F.3d at 1164.
Similarly, there is no basis for invoking "short duration" (including pre-suit termination) as a legal bar to an otherwise-documented competitive harm. What is "short" can be judged only in the relevant commercial context; as the Court found (¶ 271), an agreement may come at a particularly vital time; each of several "short" agreements may have effects that cumulate with other competition-harming acts; and the Court's findings (e.g., ¶¶ 377, 383, 411) make clear that the contracts were long enough to accomplish Microsoft's anticompetitive goals. See United States v. Loew's, 371 U.S. 38, 50 (1962) (discussing change in illegal conduct on eve of suit).
Finally, Microsoft cannot defend its agreements by saying that they are not true "exclusive dealing" agreements. MS Conc. at 23-24. As then-Judge Breyer made clear in Barry Wright, a case cited by Microsoft, the extent of the restriction in a contract affects the extent of any foreclosure, but rule-of-reason analysis hardly stops -- as it did not stop in Barry Wright -- after concluding that complete exclusivity is not present. 724 F.2d at 237. As a simple logical matter, what an agreement does not restrict can hardly save it from condemnation when what it does restrict causes harm to competition.
Microsoft contends that its conduct, even if it amounted to maintenance of the operating system monopoly, did not amount to an attempt to monopolize the market for browsers. MS Conc. at 35-44. Plaintiffs' conclusions demonstrate why the Court's findings, although less direct on these issues than on the monopolization issues, nevertheless establish the elements of the attempt offense. Pl. Conc. at 66-70. Microsoft's arguments that two elements are missing -- a specific intent to monopolize and a dangerous probability of success -- should be rejected.
Microsoft contends that specific intent was not found because there is no finding "that Microsoft acted with a specific intent to drive Netscape from the marketplace" (MS Conc. at 37) and that, instead, it was seeking to prevent Navigator from becoming the standard (id. at 35). This argument is doubly mistaken. First, as Microsoft itself notes, the issue is not what ultimate goal may have "motivat[ed]" it to seek monopoly power, but simply whether it had "a purpose to acquire monopoly power." Id. at 36. Second, as a matter of well-settled antitrust law (and basic economics), the intended monopoly power does not require obliteration of rivals; it is enough that the rivals be weakened enough so that the monopolist can exercise monopoly power without effective competitive constraint. Dial A Car, Inc. v. Transportation, Inc., 82 F.3d 484, 486 (D.C. Cir. 1996) (power "to control" competition is enough).
The findings establish that Microsoft's specific intent was to "maximize Internet Explorer's share of browser usage at Netscape's expense" (¶ 133) and "to win" (¶ 166 (quoting Microsoft)). (Emphases added.) That intent goes far beyond an intent to "compete vigorously" (Spectrum Sports, Inc. v. McQuillan, 506 U.S. 447, 459 (1993)) or "to increase its market share" (Richter Concrete Corp. v. Hilltop Concrete Corp., 691 F.2d 818, 826 (6th Cir. 1982)). It is no less an unlawful intent because Microsoft may have undertaken this intentional acquisition of "control" over the ability of any browser rival to innovate as a means to prevent Navigator from becoming the standard.
Contrary to Microsoft's contention (MS Conc. at 39-44), the findings also establish a dangerous probability that Microsoft would achieve power over price (or innovation, which can effectively decrease price by increasing the value of a product even while its dollar price remains at zero). That power, as noted, may exist without "driv[ing] non-Microsoft PC Web browsing software from the marketplace altogether." ¶ 385. The market share figures, which are important but not conclusive (Reazin, 899 F.2d at 968), demonstrate at least the closeness to monopoly power required for the attempt offense: Microsoft's share of new browser usage was already at 60% in 1998 -- it is such "sales," not installed base, that measure "market share" for forward-looking purposes (see United States v. General Dynamics Corp., 415 U.S. 486, 501 (1974)) -- and the unmistakable trajectory was toward an overall usage share of more than 60% by 2001. ¶¶ 372-373. The other pertinent features of the market (recited by Microsoft, see MS Conc. at 43) reinforce that inference: The only meaningful competitor (Netscape) has been severely weakened, deterred from innovation, and swallowed by AOL; there has been no significant new entry in recent years; and Microsoft's own conduct (including its tying of Internet Explorer to its operating system on virtually every PC) has created barriers to new entry. In these circumstances, Microsoft has achieved, at the least, a dangerous probability of monopoly power in the browser market.
Microsoft raises a copyright defense to some of the contract restrictions it imposed on OEMs -- prohibiting certain deletions of icons or program entries from the desktop or 'start' menu, and prohibiting addition of certain materials by an OEM during the initial boot sequence or elsewhere. See ¶ 213. Microsoft argues that the restrictions "simply restate" its rights under the Copyright Act as the owner of copyrights for the software constituting Windows 95 and Windows 98 and are therefore immune from the antitrust laws. MS Conc. at 26-33. This argument is untenable.
A. Microsoft has not established that the Copyright Act itself imposes the restrictions it placed in its OEM contracts, let alone that those restrictions were uniquely intended by Congress to override otherwise applicable Sherman Act constraints. Section 106 of the Copyright Act enumerates the "exclusive rights in copyrighted works" (17 U.S.C. § 106)(16); as the Supreme Court has made clear, Congress "has never accorded the copyright owner complete control over all possible uses of his work" but has instead limited the holder to those rights enumerated in the statute. Sony Corporation of America v. Universal City Studios, Inc., 464 U.S. 417, 432 (1984). Thus, "[e]ven unauthorized uses of a copyrighted work are not necessarily infringing," and "an unlicensed use of the copyright is not an infringement unless it conflicts with one of the specific exclusive rights conferred by the copyright statute." Id. at 446. Cf. 17 U.S.C. § 501(a) (infringement means violating specifically enumerated rights); Twentieth Century Music Corp. v. Aiken, 422 U.S. 151, 155 (1973) (not all uses of a work are within copyright holder's control; rights limited to specifically granted "exclusive rights"); NLFC, Inc. v. Devcom Mid-America, Inc., 45 F.3d 231, 234 (7th Cir. 1995).
Microsoft never identifies which statutory right would be infringed by the OEM actions at issue. Nor does Microsoft offer any reference to facts adduced at trial or found by this Court(17) or any explanation of precisely how, if at all, the conduct forbidden to OEMs by their licenses would affect any copyrighted material or, if so, how it would infringe any statutory right.(18) Instead, Microsoft asserts a limitless "right to preserve the integrity of its copyrighted works" (MS Conc. at 28), one far beyond any right protected by the statute.(19)
In attempted support of its purported right, Microsoft relies on three decisions in which it finds "express recogni[tion] that the copyright laws prohibit unauthorized modification of copyrighted works." MS Conc. at 29-31 (citing Gilliam v. ABC, 538 F.2d 14 (2d Cir. 1976); WGN Continental Broadcasting Co. v. United Video, Inc., 693 F.2d 622 (7th Cir. 1982); National Bank of Commerce v. Shaklee Corp., 503 F. Supp. 533 (W.D. Tex. 1980)).(20) But none of the cited decisions establishes what Microsoft urges because, among other things, none of them involved provisions of the Copyright Act applicable here.(21)
Moreover, even if Microsoft had identified a general Copyright Act right against alteration, that would not be enough. Microsoft's argument specifically requires that Congress have intended that the exercise of this particular asserted right, unlike the exercise of other Copyright Act rights, be immune from otherwise-applicable antitrust limitations. Microsoft, 1998 WL 614485, at *15. Microsoft advances no basis, in the Act or in its cited authorities, for attributing that singular intent to Congress.
B. Even if Microsoft had established the required Copyright Act right, its argument here would fail. Microsoft's attempted copyright defense to its contractual provisions concerning deletions rests on a silent and erroneous premise: the single-product contention that this Court has firmly rejected as a factual matter (and now should reject as a legal matter).(22) Microsoft proceeds as if there were a single work, one that includes both Windows and Internet Explorer, and contends that the Copyright Act gives it an absolute right, overriding any antitrust limitations, against "unauthorized modifications" of that work. If Microsoft's theory (applied even to separate products) were correct, however, a film studio could package several films in a single collective work (protected by copyright) and preclude all antitrust scrutiny of a requirement that licensees take the whole package. Such a result would undo settled antitrust law applied to copyrighted products. See Loew's, 371 U.S. at 45; United States v. Paramount Pictures, 334 U.S. 131 (1948).
C. In any event, Microsoft does not make any argument, beyond its simple assertion, that Sherman Act condemnation of its OEM restrictions should be rejected as somehow impairing its copyright "interests." That failure is hardly surprising. Even in the context of refusals to license, let alone in the less difficult context of anticompetitive license restrictions (like those at issue here), intellectual property defenses to conduct that violates the antitrust laws have been rejected.(23) See, e.g., Image Technical Servs., Inc. v. Eastman Kodak, 125 F.3d 1195, 1218 (9th Cir. 1997) (no absolute immunity even for refusal to license); Data General Corp. v. Grumman Sys. Support Corp., 36 F.3d 1147, 1187 (1st Cir. 1994) (same); cf. Eastman Kodak, 504 U.S. at 479 n.29 (citing several of numerous Supreme Court decisions applying antitrust laws to strike down restrictions imposed by holders of patents and copyrights). Indeed, the Supreme Court has said that "the copyright laws confer no rights on copyright owners to fix prices among themselves or otherwise to violate the antitrust laws." Broadcast Music Inc. v. CBS, 441 U.S. 1, 19 (1979) (emphasis added).(24) Where, as here, the defendant does not simply refuse to license its intellectual property, but affirmatively licenses it with restrictions on its use that harm competition and serve no legitimate business purpose, copyright protections cannot be invoked to thwart the operation of the antitrust laws. See, e.g., Interstate Circuit, Inc. v. United States, 306 U.S. 208, 230 (1939); Paramount Pictures, 334 U.S. at 159.
Microsoft does not even attempt to show how the particular restrictions at issue here could escape Sherman Act condemnation under that body of law. Any such attempt would fail in light of this Court's plain finding (¶ 228):
No copyright interest is impaired by enforcing the Sherman Act's prohibition to condemn agreements like the OEM restrictions at issue here -- restrictions that, far from being reasonably necessary to protect any legitimate copyright rights, simply bar OEMs from assisting consumers in making choices they are fully entitled to make even under Microsoft's view of copyright law.
This Court should conclude that Microsoft has violated Sections 1 and 2 of the Sherman Act and proceed to consider the appropriate remedy.
1. E.g., Times-Picayune Publishing Co. v. United States, 345 U.S. 594, 612 n.31 (1953) ("For every product substitutes exist. But a relevant market cannot meaningfully encompass that infinite range. The circle must be drawn narrowly to exclude any other product to which, within reasonable variations in price, only a limited number of buyers will turn . . . .").
2. Relatedly, Microsoft suggests in a footnote (MS Conc. at 49 n.28) that "the legally relevant product market in this case is software development platforms, not operating systems." That contention is wrong, confusing the market with the mechanism of the competitive threat. Consumers wanting software to operate their Intel-compatible PCs do not meaningfully choose between Windows and, say, Unix-based server or mainframe operating systems that software developers may use as platforms, any more than they choose between Windows and a browser or Java. The platform features of an Intel-compatible PC operating system (the exposing of APIs) are central to the applications barrier to entry into the operating system market, and it is the middleware threat to that barrier which Microsoft perceived and attacked so as to protect its operating system dominance.
3. See also American Professional Testing Service, Inc. v. Harcourt Brace Jovanovich Legal and Professional Publications, Inc., 108 F.3d 1147, 1154 (9th Cir. 1997) (barriers to entry include any factors in the market that deter entry while permitting incumbent firms to earn monopoly returns). Microsoft's own expert, Dean Schmalensee, concurred in this definition of an entry barrier. Transcr. Jan. 14, 1999 a.m. at 6:17-7:19.
4. See Neumann v. Reinforced Earth Co., 786 F.2d 424, 427 (D.C. Cir. 1986); Pl. Conc. at 17-18 n.3 (citing decisions using both formulations, each reflected in Aspen).
5. Those decisions, in any event, are simply applications of, and do not reject, the more general principle that conduct is anticompetitive if it unnecessarily restricts competition or makes no business sense except as a means of excluding rivals and thereby facilitating the exercise of monopoly power. See, e.g., In re IBM Peripherals EDP Devices Antitrust Litigation, 481 F. Supp. 965, 1003 (N.D. Cal. 1979) ("A more generalized standard, one applicable to all types of otherwise legal conduct by a monopolist . . . must be applied to the technological design activity at issue here."), aff'd, 671 F.2d 1377 (9th Cir. 1983); GAF Corp. v. Eastman Kodak Co., 519 F. Supp. 1203, 1228-29 & n.19 (S.D.N.Y. 1981) ("the 'reasonableness' of the design of a monopolist's new products . . . may be scrutinized under §2 in cases in which . . . a single firm controls the entire market or in which a monopolist engages in coercive conduct.").
6. MS Conc. at 58. In both of those cases, unlike this case, the courts expressly found that there was no proof that the complained-of design had either forced purchasers to take unwanted products or injured the defendant's rivals. See Response of Carolina, Inc. v. Leasco Response, Inc., 537 F.2d 1307, 1329-30 (5th Cir. 1976); ILC Peripherals Leasing Corp. v. IBM, 458 F. Supp. 423, 439 (N.D. Cal. 1978), aff'd sub nom. Memorex Corp. v. IBM, 636 F.2d 1188 (9th Cir. 1980).
7. To the extent that Microsoft incorporates here its assertion of a copyright defense to Section 1 condemnation of its OEM agreements, that defense is meritless for reasons discussed in Part V below.
8. The same sensible focus on anticompetitive conduct as embracing commercially significant and unjustified raising of rivals' costs, as opposed to some notion of absolute foreclosure, is reflected in other precedents. See Forsyth v. Humana, Inc., 114 F.3d 1467, 1478 (9th Cir. 1997), aff'd, 525 U.S. 299 (1999) (Section 2); Premier Elec. Constr. Co. v. National Elec. Contr. Ass'n, Inc., 814 F.2d 358, 368 (7th Cir. 1987) (Easterbrook, J.) (Section 1 violation based on discouragement through mere raising of rivals' costs by unjustified conduct); cf. Ball Mem. Hosp., 784 F.2d at 1339 (Easterbrook, J.) (recognizing, but rejecting on facts, claim of "raising rivals' costs" as monopolizing conduct).
9. For that reason, even the ICP agreements, which this Court found had not been proved, by themselves, to have "actually had a substantial, deleterious impact on Navigator's usage share" (¶ 332), were anticompetitive.
10. Microsoft relies on the Court's finding that it could defray the costs of its campaign from the "vast profits" it obtains from its Windows operating system. MS Conc. at 63. That finding, however, says nothing more than that Microsoft can use its monopoly profits to subsidize the losses caused by its browser campaign. Moreover, even if the findings were construed to mean that Microsoft's offering a browser bundled with its operating system resulted in additional Windows revenues sufficient to defray the costs of the bundle, it would still not answer plaintiffs' contention -- that Microsoft's campaign to distribute stand-alone versions of its browser was predatory. See ¶¶ 136, 141 ("profitable only to the extent that it protected the applications programming barrier to entry").
11. Focusing solely on an incident involving initial removal of a Navigator icon, Microsoft suggests that it obtained no success with Compaq (MS Conc. at 65 n.39), but it simply ignores the Court's express findings of Compaq's various Internet Explorer commitments, and decision not to re-install a Navigator icon, in 1996 and 1997. ¶¶ 233-234.
12. See III P. Areeda & H. Hovenkamp, Antitrust Law ¶ 651c, at 77 (1996) ("[B]ecause monopoly will almost certainly be grounded in part in factors other than a particular exclusionary act, no government seriously concerned about the evil of monopoly would condition its intervention solely on a clear and genuine chain of causation from exclusionary act to the presence of monopoly.")
13. Microsoft argues that Justice O'Connor's concurring opinion in Jefferson Parish "observed that a different rule applies to integrated products." MS Conc. at 4. The concurrence, however, makes no such proposal for two different standards for determining when there are two products. The court in Jack Walters & Sons Corp. v. Morton Bldg, Inc., 737 F.2d 698 (7th Cir. 1984), also cited by Microsoft (MS Conc. at 5 n.2), did not depart from Jefferson Parish.
14. As noted in plaintiffs' proposed conclusions, if it were necessary to decide the issue, the Court should rule that the Supreme Court's test establishes the standard for determining whether there are one or more products. See Pl. Conc. at 57. Microsoft complains that, under the Supreme Court's test, it would be required "to offer OEMs and end users, a Chinese menu of options." MS Conc. at 5-6. But the Supreme Court test has no such implication, for it requires that separate options be offered only when it is "efficient" to do so. See Pl. Conc. at 55-56. In any event, Microsoft's argument, and its concern about testing costs and the like, ring hollow in light of the fact that Microsoft already offers end users a "Chinese menu of options," allowing optional installation or removal through the Add/Remove utility of more than 70 software products or components shipped with Windows 98 (including most that were Internet-related, with the striking exception of Internet Explorer). See ¶ 170.
15. Microsoft quotes (MS Conc. at 10) the D.C. Circuit's statement that "Windows 95 without IE's code will not boot, J.A. 1623" (Microsoft, 147 F.3d at 948 n.11), but there is no question under the full record developed at trial that the retail version of Windows 95 initially offered without Internet Explorer boots and functions on its own and that all the relevant benefits of Windows 98 could be achieved by a user combining that version of Windows 95 with a stand-alone browser like Internet Explorer 5.0. See Pl. Conc. at 59.
16. The rights specified in Section 106(1)-(6) are the rights to reproduce the work (§ 106(1)), to prepare derivative works (§ 106(2)), to distribute copies of the work (§ 106(3)), to perform the work publicly (§ 106(4) & (6)), and to display the work publicly (§ 106(5)).
17. This court's findings contain just one paragraph on Microsoft's copyrights (¶ 228), reflecting Microsoft's failure to introduce any evidence on this point.
18. Microsoft has failed to address, for example, whether the OEM actions at issue would affect any copyrighted software code, let alone any aspect of Windows that could be deemed an "expression" by Microsoft protected by copyright. See generally Lotus Development Corp. v. Borland Intern., Inc., 49 F.3d 807, 816-17 (1st Cir. 1995) (distinguishing "method of operation" from "expression" in computer programs).
19. Presumably, were Microsoft to try grounding its asserted right in the Copyright Act, it would have to rely on the derivative works provision of 17 U.S.C. § 106(2), which is the only provision in the statute that addresses alterations to a copyrighted work. See 17 U.S.C. § 101 ("A work consisting of editorial revisions, annotations, elaborations, or other modifications which, as a whole, represent an original work of authorship, is a 'derivative work.'"). Microsoft has not cited this section, apparently because it recognizes that the provision furnishes no support for its copyright defense. Among other things, it is settled, including in the Second and Seventh Circuits (whose decisions Microsoft invokes), that not all alterations create a derivative work. See Matthew Bender & Co., Inc. v. West Pub. Co., 158 F.3d 674 (2d Cir. 1998), cert. denied, 119 S. Ct. 2039 (1999); Lee v. A.R.T., 125 F.3d 580 (7th Cir. 1997); see also Paramount Pictures v. Video Broadcasting Systems, 724 F. Supp. 808 (D. Kan. 1989) (advertisements inserted in blank leader of videotape did not create unauthorized derivative work based on video's movie).
20. Microsoft also cites three other decisions, none of which supports its contention of an unfettered right to control its copyrighted works in the hands of a licensee. S.O.S., Inc. v. Payday, Inc., 886 F.2d 1081, 1089 (9th Cir. 1989), explicitly recognizes as distinct issues the questions whether a licensee's conduct exceeds the scope of its license and whether that conduct infringes the copyright. LucasArts Entertainment Co. v. Humongous Entertainment Co., 870 F. Supp. 285 (N.D. Cal. 1993), undertakes a lengthy and painstaking evaluation of whether the copyright holder's termination of a license violated the antitrust laws, recognizing that such a ruling was entirely possible. And Corsearch, Inc. v. Thomson & Thomson, 792 F. Supp. 305 (S.D.N.Y. 1992), simply holds a pricing provision in a license agreement not to be a per se illegal price fixing agreement.
21. Neither Gilliam nor Shaklee involved the current Copyright Act at all. Moreover, Gilliam, which concerned deletion of approximately 27 percent of a derivative work, expressly rejected the copyright holder's assertion -- echoed by Microsoft in its conclusions here -- "that any editing whatsoever would constitute infringement." 538 F.2d at 23. The court there held, instead, simply that ABC, which had no license to create a derivative work, exceeded the scope of the license granted to ABC by BBC, which had obtained the copyright holder's permission to create only a different derivative work, i.e., the full taped production. Id. at 17, 21. WGN turned on the statutory issue, not raised here, whether the defendant, which had no license to retransmit WGN's news broadcast and was thus otherwise plainly an infringer, could take advantage of the "passive carrier" exemption in 17 U.S.C. § 111(a)(3).
There is one case involving conduct post-dating enactment of the current Copyright Act in which a party, like Microsoft here, claimed a copyright violation, but nonetheless did "not ground this claim on the infringement of any right found in 17 U.S.C. § 106, and instead relie[d] on" WGN, Gilliam, and Shaklee. Paramount Pictures v. Video Broadcasting Systems, 724 F. Supp. 808, 819 (D. Kan. 1989). After noting that Nimmer describes this trio as involving "moral rights" of a copyright holder, which the copyright statutes have not created and the courts have not generally entertained, and that the Supreme Court has insisted that copyright infringement claims have a statutory basis, the court was "reluctant to extend their holdings beyond their controlling facts," id. (citing 2 Nimmer on Copyright § 8.21 [C] and Twentieth Century Corp., 422 U.S. at 155), and thus rejected the copyright claim.
22. The fact that a single copyright registration was employed for both Windows and Internet Explorer cannot support a claim that those two products are, for antitrust purposes or otherwise, one product. As the Copyright Office's regulations make clear, multiple copyrightable works may be registered on a single form, see 37 C.F.R. § 202.3(b)(3)(A), without thereby rendering them a single work. The number of distinct "works" for purposes of the copyright law is determined not by the number of registration forms, but by the number of distinct copyrightable entities that are separately economically viable. See Gamma Audio & Video, Inc. v. Ean-Chea, 11 F.3d 1106, 1116-17 (1st Cir. 1993) (four episodes of television series were four separate "works" although all were included on same registration form); Walt Disney Co. v. Powell, 897 F.2d 565, 570 (D.C. Cir. 1990) (six t-shirts depicting Mickey and Minnie Mouse were two separate "works" although each of them was included on a separate registration form).
23. Microsoft's citation of several ordinary unilateral "refusal to deal" cases (see MS Conc. at 32-33) are thus of no significance.
24. The Court then stated that it "would not expect that market arrangements reasonably necessary to effectuate the rights that are granted [by the Copyright Act] would be deemed a per se violation of the Sherman Act." Id. That reason for rejecting per se condemnation, itself applicable only to those arrangements "reasonably necessary" to effectuate specific Copyright Act rights, remits copyright holders to the ordinary, context-specific Section 2 and Section 1 (rule of reason) standards.