|
Tech Law Journal |
 |
| News, records, and analysis of legislation, litigation, and regulation affecting the computer, internet, communications and information technology sectors | |
|
|
|
| |||||
|
Part II. C.
There are four elements to a Section 1 per se tying claim: (1) two separate products or services exist; (2) the sale of one product (the tying product) is conditioned on the purchase of the other (the tied product); (3) the seller has "appreciable economic power" in the tying market; and (4) the tying arrangement affects a not insubstantial volume of interstate commerce in the tied product. See Jefferson Parish Hospital Dist. No. 2 v. Hyde, 466 U.S. 1, 9-18 (1984); Eastman Kodak, 504 U.S. at 462. For the purposes of summary judgment, on which all disputed issues must be resolved against the moving party, Microsoft does not contest the third and fourth elements. Its arguments on the first and second are both factually and legally inadequate to warrant summary judgment. Indeed, its coercive bundling of Internet Explorer with Windows is per se illegal.22
Microsoft devotes a single page to the "conditioning" element of the plaintiffs’ tying claims, arguing that its refusal to allow OEMs to license Windows without Internet Explorer does not amount to "conditioning" because it has charged OEMs only a single royalty for both Windows and Internet Explorer and "has never charged OEMs a separate royalty" for Internet Explorer alone. MS Memo at 49. This formalism fails to address the evidence that, despite the [begin page 49] fact that Internet Explorer and Windows are not separately priced, Microsoft’s coercive bundling and refusals to permit unbundling have constituted, and continue to constitute, a powerful anticompetitive mechanism. Microsoft correctly states that "conditioning the availability of one product on the purchase of another is a necessary element of a tying claim under Section 1 of the Sherman Act." MS Memo at 49. However, Microsoft misinterprets the "purchase" (or "conditioning") requirement to mean that a separate charge must be assessed for Internet Explorer. To the contrary, as Professors Areeda and Hovenkamp explain:
3A P.E. Areeda & H. Hovenkamp, Antitrust Law, ¶ 760b6 (1996). Whether the market price of the tied product is zero or something higher is thus immaterial to the potential for anticompetitive harm, and accordingly should be immaterial to analysis of whether there is a tying arrangement.23 [begin page 50] Even though no separate price is charged for Internet Explorer, and even though it is licensed under the same licensing agreement as Windows, the evidence plainly indicates that Microsoft’s insistence that OEMs accept and preinstall it constitutes anticompetitive conditioning because it has materially affected OEMs’ judgment as to the browser software they preinstall on their PCs. See supra, Section II.B.1.a(2). The competitive harm which the per se rule against tying is designed to prevent includes the denial of "free access to the market for the tied product" and the forcing of consumers to "forego their free choice between competing products." Northern Pac. Ry., 356 U.S. 1, 6 (1958). Microsoft’s tying of Internet Explorer to Windows produces these precise evils, and thus imposes real costs on OEMs. Microsoft’s argument -- the mere formalism that there is no separately priced purchase of Internet Explorer -- cannot entitle it to summary judgment.
Microsoft’s real attack on the plaintiffs’ per se tying claim challenges the first element of the tying offense, whether Internet Explorer and Windows operating systems are separate products. Instead of coming to grips in any serious way with Jefferson Parish and Eastman Kodak, the controlling decisions on the separate product issue, Microsoft relies on prior and/or distinguishable lower court decisions, and on the D.C. Circuit’s decision in the consent decree action brought last year by the United States, United States v. Microsoft, 1998 WL 327855 (D.C. Cir. June 23, 1998) (hereinafter, "the consent decree case").
The Supreme Court held in Jefferson Parish, and reiterated in Eastman Kodak, that whether something that is sold by the defendant as a single package or bundle consists of one or [begin page 51] more "products" for tying purposes depends on whether there is demand for each of the bundled products separately, apart from the package, and whether, in light of this separate demand, it is efficient for the defendant to sell the different components or products separately.24 The antitrust issue raised by tying doctrine is thus whether the defendant is required also to offer an unbundled alternative (i.e., offer the tying product without the particular tied product required as a condition of that offer), based on whether there is demand for the products separately. Jefferson Parish makes clear that the test is based in economics, not technology. Under Jefferson Parish, "the answer to the question whether one or two products are involved turns not on the functional relation between them, but rather on the character of the demand for the two items." 466 U.S. at 19; see also Multistate Legal Studies, 63 F.3d at 1547; Klamath-Lake Pharmaceutical Assn. v. Klamath Medical Service Bureau, 701 F.2d 1276, 1289 (9th Cir.), cert. denied, 464 U.S. 822 (1983) ("Products that function together and are sold in combination may still be ‘separate’ if consumers would prefer to buy them individually at the price necessary to market them separately . . . . It is the relationship of the producer’s selling decision to market demand, not the physical characteristics of the products alone, that determines the existence of legally separable products.") [begin page 52]
Microsoft concedes that there is demand for Internet Explorer without or separate from Windows operating systems. For example, the evidence shows that Microsoft has for years offered Internet Explorer separate from Windows, both for Windows users and for users of operating systems other than Windows, and even Microsoft’s brief argues that many users obtain their browsers separately. See MS Memo at 9-17. Customers who acquire either Microsoft’s browser or another browser in these various ways, whether from ISPs (in most cases), through retail purchase, or otherwise, obviously do so independently of the acquisition of their operating system. Microsoft offers Internet Explorer separately, and users obtain it separately, in order to satisfy consumer demand for web browsers. Microsoft argues that there is no separate demand for the alleged tying products (Windows 95 and Windows 98) without the tied product (Internet Explorer) on the ground that the plaintiffs "cannot show that there is separate demand for operating system software products that do not provide web browsing functionality." MS Memo at 45. This argument is both wrong on the facts and rests on a misunderstanding of the applicable legal test. The issue is not whether there is demand for operating system (tying) products without any browser (tied) products whatsoever, but rather whether there is demand for the particular tying product (Windows) without the particular tied product (Internet Explorer) required by the defendant. The facts presented in Jefferson Parish itself are a clear example of this point. The Supreme Court fully recognized that there was, and could be, no suggestion that anyone wanted surgery (the tying product) without any anesthesia (the tied product); but it nonetheless asked whether there was demand for surgery without the particular anesthesiologists required by [begin page 53] the defendant, and cited numerous earlier cases in which tying arrangements had been found despite the uselessness of the tying product without some product in the tied product market. See Jefferson Parish, 466 U.S. at 19, n.30, 22-23. In any event, there is abundant, uncontroverted evidence both of demand (and a separate market) for Windows without any browser product at all, and of demand for Windows without Internet Explorer in particular. This evidence includes the following. (1) OEMs have repeatedly sought to effectively remove Internet Explorer by removing icons and other means of access to Internet Explorer, and they have sought both to sell computers both without Internet browsing capability at all and with browsers other than Internet Explorer. [redacted material] Microsoft argues that "[r]emoving `icons and other means of access’ is not the same as removing Internet Explorer, and thus does not establish separate consumer demand for Windows without those technologies." MS Memo at 47. But as discussed below, it is immaterial whether OEMs have sought to remove all "technologies" that Microsoft chooses to associate with Internet Explorer. The issue for tying purposes is the economic question whether OEMs have sought, whether by removing icons, other means of access, or otherwise, to serve demand for Windows without Internet Explorer from the perspective of the end user. The evidence shows exactly that.
(2) Some users prefer Windows operating systems with a browser other than Internet Explorer; other users, particularly corporate customers, prefer Windows operating systems with no browser at all. Microsoft ignores the evidence both that some corporate users do not want to license any browser along with Windows, and that other users wish to select from among competing browsers on the merits and across operating system platforms, and therefore do not [begin page 54] want to have Internet Explorer forced upon them. [redacted material] This evidence too shows separate demand for Windows without Internet Explorer.25 (3) Microsoft and others in the industry recognize that browsers and operating systems are separate products. As detailed in the plaintiffs’ PI Briefs, Microsoft has recognized it is in a "browser war" with Netscape, has made obtaining ever higher "browser share" a top corporate goal in order to win that war, and has meticulously tracked that browser share wholly apart from its share of Windows or any other product. See U.S. PI Memo at 19-22, 60-64, and cites therein. Moreover, Microsoft’s strategy of making its Internet Explorer browser "cross-platform" (i.e., available to run on multiple operating systems) itself demonstrates that the browser is a separate product. The cross-platform versions of Internet Explorer are designed to satisfy consumer demand for browsers as products in their own right, not as components of particular operating systems. Indeed, Microsoft recognizes that numerous users do not wish to have their choice of browser linked to their choice (or lack thereof) of operating system.26 [begin page 55] Microsoft does not dispute that it has developed versions of Internet Explorer 3.0 and 4.0 for the Macintosh, Windows 3.1, and Solaris, or that it will continue to develop future versions of Internet Explorer to be a cross-platform product. [redacted material] (4) The practices of vendors of other operating systems demonstrate that there is separate demand for operating systems. Microsoft argues that Windows and Internet Explorer should not be considered separate products on the ground that "every modern operating system for personal computers includes a variety of technologies that facilitate access to information on the Internet, including web browsing functionality." MS Memo at 45. The evidence shows just the opposite.27 Other operating system vendors approach the bundling, if any, of browser products with their operating systems in a way fundamentally different from Microsoft. Microsoft has not identified a single operating system vendor which requires licensees to install and not to remove a particular browser as a condition of licensing the operating system. Rather, other vendors either do not bundle a browser at all, see, e.g., [redacted material] Wack Dec., ¶¶ 23-26, or permit OEM licensees not to install a browser offered with the operating system (or to remove it after installation) if they wish. See, e.g., [redacted material] [begin page 56] [redacted material] Microsoft does not really contest the evidence that both it and other industry participants treat browsers as separate products or that some OEMs and computer users would like to obtain Windows operating systems without the Internet browsing functionality provided by Internet Explorer. It argues, instead, that this evidence is insufficient because there must be evidence of "widespread sales of the tying item in unbundled form." MS Memo at 45, quoting 10 P.E. Areeda, H. Hovenkamp, E. Elhauge, Antitrust Law ¶ 1745d2 at 211. But Microsoft is creating a legal "Catch-22." There obviously cannot be widespread sales of Windows without Internet Explorer because Microsoft, with its dominance of the desktop PC operating system market, has consistently required OEMs to take Internet Explorer as a condition of obtaining Windows. For this very reason, the available (and substantial) evidence of separate demand for Windows and Internet Explorer may well understate the actual extent of such demand. OEMs, having long since recognized that they have no choice, have simply acquiesced with Microsoft’s required bundling.
The efficiency question in tying cases is whether, in light of the demand for an unbundled option, it is efficient for the defendant to provide such an option. Eastman Kodak, 504 U.S. at 462. Thus, in this case the question is whether it is efficient for Microsoft to satisfy the demand for separate or standalone browsing functionality provided by Internet Explorer, and the demand for Windows operating system functionality without the web browsing functionality provided by Internet Explorer. There is no dispute that Microsoft can efficiently satisfy end users’ demand for separate or standalone web browsing functionality. Microsoft does so by offering the standalone versions of its Internet Explorer web browser product, both to users of Windows and to users of various non-Windows operating systems.
Similarly, if Microsoft wanted to satisfy the demand of OEMs and computer users for operating systems without the web browsing functionality provided by Internet Explorer, rather than wanting just to exclude its browser rivals, it could efficiently offer those OEMs or users the alternative of Windows without that particular web browsing functionality.28 In the case of Windows 95, Microsoft concededly provided the means, through the Add/Remove utility, for users to remove Internet Explorer. It could have offered OEMs a similarly unbundled version of Windows 95,29 but chose not to do so and, instead, prohibited the OEMs from themselves utilizing the Add/Remove utility or otherwise removing IE. With regard to Windows 98, where Microsoft has chosen not to offer anyone the ready means of removing Internet Explorer using the Add/Remove utility or otherwise, it would nonetheless be efficient for Microsoft to offer an unbundled alternative. Felten Dec., ¶¶ 7-10.30 [begin page 58] Of course, under Jefferson Parish, Microsoft need not necessarily offer a version of Windows 98 without utilization of any browsing functionality, but rather without that functionality necessarily belonging to a particular browser -- Internet Explorer. This is precisely what other operating system vendors universally, and efficiently, do in providing their operating system products. Plaintiffs’ evidence will establish that it is readily possible and efficient for Microsoft to offer Windows 98 in a way that satisfies the demand of OEMs and users who desire either an operating system without web browsing functionality or an operating system on which it is easier or more economical to install a different web browser product.
Microsoft argues that the removal of what it calls "Internet Explorer technologies" from Windows 98 will "severely degrad[e] the operating systems." MS Memo at 42. This argument -- the linchpin of Microsoft’s entire defense to the tying claims -- rests on a semantic sleight of hand and is insufficient for summary judgment both as a matter of law and as a matter of fact. This sleight of hand is Microsoft’s equation of "Internet Explorer technologies" with every bit of software code used to browse the Internet using Internet Explorer. This gambit, identical to the one Microsoft attempted and the Court rejected last winter, is inconsistent with the economic inquiry made in Jefferson Parish.31 The efficiency of providing an unbundled alternative is fundamentally an economic question, not a technical one. As OEMs have recognized, removing access to Web browser functionality from Windows or another software product effectively removes the web browser, [begin page 59] eliminating the ability of the remaining Windows product to satisfy the separate demand for such functionality (even if shared files are left behind). Moreover, removing such functionality or access to such functionality from Windows leaves a fully functioning operating system that satisfies the demand of OEMs and users who desire either an operating system without web browsing functionality or on which it is easier or more economical to install a different browser. Microsoft can efficiently remove or permit OEMs to remove the specific Internet Explorer Web browser functionality from its bundled Windows products and, if desired, to substitute the Web browsing functionality provided by competing browsers.
Microsoft argues that the "integration" of "Internet Explorer technologies" with Windows 98 is a "technological" tie-in and is, therefore, subject to "a specific body of case law" that prevents any inquiry into the nature of the tie, the facts bearing on its adoption and implementation, or its competitive effects. MS Memo at 20. Indeed, Microsoft argues that "technically interconnected products" are essentially "immune" from tying claims -- that the court's inquiry is at an end once a defendant has made a plausible showing that there is "some" technological benefit from the challenged combination and that the tie was not carried out "solely for the purpose of tying two separate products." MS Memo 20, 24 (emphasis added). In substance, Microsoft argues that, even though it may be a monopolist, its bundling of Internet Explorer with Windows is not subject to antitrust scrutiny so long as it can show some plausible benefit from the conduct, regardless of whether the conduct is anticompetitive regardless of whether the conduct is on balance beneficial or harmful to consumers. [begin page 60] As discussed in Section II.C.2, infra, Microsoft’s arguments do not and cannot justify its conduct under Section 2 of the Sherman Act if (as is clear here) the conduct serves to maintain monopoly power by raising barriers to entry, increasing rivals’ costs, or foreclosing competition on the merits. Microsoft rests its argument on a few lower court cases that, for two reasons, are inapposite here, even as to plaintiffs’ Section 1 claims. First, Microsoft’s interpretation of these cases conflicts with the Supreme Court’s later, seminal pronouncements on tying law in Jefferson Parish and Eastman Kodak. Second, Microsoft has not properly understood even the cases on which it relies. Both the Supreme Court and the lower courts have relied on the demand-based analysis mandated by Jefferson Parish to evaluate separate product claims in "technological" tie-ins under Section l of the Sherman Act. See Eastman Kodak, 504 U.S. at 461-63 (in context of what the Court characterized as a "high-technology" service industry, relied on Jefferson Parish and other established tying cases, without any suggestion that any different, more relaxed "body of law" should apply in technology-related cases); Data General v. Grumman Systems Support, 36 F.3d 1147, 1178-81 (lst Cir. 1994) (alleged tie of ADEX software and services); Service & Training, 963 F.2d at 683-85 (reversing grant of summary judgment to defendant on claim of tie between ADEX and repair services); Allen-Myland v. IBM Corp., 33 F.3d 194, 200-16 (3d Cir. 1994) (reversing judgment for defendant on alleged tie of large-scale mainframe computers and the labor to install upgrades to mainframes); Digidyne Corp. v. Data General Corp., 734 F.2d 1336, 1339 (9th Cir. 1984) (holding tie of NOVA computer system to NOVA operating system unlawful). [begin page 61] Similarly, the cases on which Microsoft relies in fact look to standard tie-in criteria to evaluate the lawfulness of "technological" ties and rest on facts materially different from those alleged (and not disputed by Microsoft) here. In Telex Corp. v. IBM Corp., 367 F. Supp. 258 (N.D. Okla. 1973), rev'd other grounds, 510 F.2d 894 (10th Cir. 1975), for example, which appears to be the origin of the "technological" tie-in language, the court held that, unlike the circumstances here, there was no forced tie of memory and control functions in a central controller. Telex, 367 F. Supp. at 347 ("the integrated control in the System 370 is wholly optional. IBM continues to offer central processing units without integrated controllers.")
Similarly, in Response of Carolina, Inc. v. Leasco Response, Inc., 537 F.2d 1307 (5th Cir. 1976), the court applied traditional tie-in law to hold that plaintiff's failure to prove coercion in the alleged tie (of computer hardware to a computer time-sharing software franchise known as "Response I") was fatal to its claim. Id. at 1327-30. After discussing the absence of coercion in great detail and at great length, the court suggested in a sentence of dicta that a tie of two different products accomplished by product design would be unlawful only if "the technological factor tying the hardware to the software has been designed for the purpose of tying the products, rather than to achieve some technologically beneficial result." Id. at 1330 (citing Telex).32 Here, of course, there is substantial evidence of anticompetitive purpose, see U.S. Memorandum in Support of Motion [begin page 62] for Preliminary Injunction at 60-64; and Microsoft enforced its tie, not just by product design, but also by contractual prohibitions on OEM efforts to delete the browser. In Foremost Pro Color v. Eastman Kodak Co., 703 F.2d 534, 539-42 (9th Cir. 1983), the plaintiff alleged an "implicit" tie of Kodak's new 110 camera to the film and processing supplies needed to use it. Foremost is readily distinguishable from the present case because it involved not the bundling of products, but rather the development of new technological formats that rendered competitors’ complements incompatible. The court explained there that the "so-called technological tie" of a new product that could not be used with old complements, and thus required the purchase of new ones, did not "standing alone" (without any contractual requirement that users take the two together) establish a per se unlawful tying arrangement. Id. at 542. Here, of course, the tying claim is premised not on the creation of any incompatibility, but rather on the clearly demonstrated contractual coercion lacking in Foremost.
Microsoft describes Innovation Data Processing, Inc. v. IBM Corp., 585 F. Supp. 1470 (D.N.J. 1984), as a case "squarely on point" for the superficial reason that it "involved the integration of new features into an operating system." MS Memo at 28. But the court there held that there was no unlawful tie of MVS operating system software and DFDSS software because, unlike Microsoft in this case, defendant licensed them separately as well as together, at the user's option. Id. at 1474-75.33 [begin page 63] Finally, in ILC I, the court used traditional tie-in criteria to conclude that a disk drive that integrated a drive unit and head/disk assembly was not an unlawful tie-in. The court found that the drive unit and the head/drive assembly were designed to be and would be used as a unit; that the aggregation offered dramatically larger online storage capacity previously unavailable; that the aggregation satisfied a recognized customer need; that the aggregation resulted in cost savings that were passed on, at least in part, to end users; that the drive unit and head/drive assembly were normally used by customers in fixed proportions; and that the practice of other industry participants, including the plaintiff, was to sell integrated disks and the drive on which they operated for a single price. 448 F. Supp. at 232-34. The court’s analysis, based on these factors, is flatly inconsistent with Microsoft’s argument that, once the defendant makes a plausible showing of some technological benefit, the court should look no further. Moreover, the facts that led the ILC I court to find a single product are conspicuously different than those in this case. Operating systems and browsers are not used in fixed proportions;34 not all sellers bundle the products together; no consumer savings have been shown to result from the bundling of Windows and Internet Explorer; and not even Microsoft claims that the benefits from bundling Internet Explorer (as opposed to any other browser) are "dramatic." See 448 F.Supp. at 233. Thus, even if the cases on which Microsoft relies were good law -- and, to the extent they precede and conflict with Jefferson Parish and Eastman Kodak and their progeny, they are not -- they would not provide a basis for summary judgment. Those cases ultimately turned on [begin page 64] traditional tie-in law, not some special laissez faire rules for product design cases, and that law requires an inquiry into the economic facts relevant to each element of the tie-in standard. For example, in In re IBM Peripheral EDP Devices, 481 F. Supp. 965 (N.D. Cal. 1979), aff'd, 698 F.2d 1377 (9th Cir. 1983), the court refused to defer to product design choices whenever they are "justified" or "reasonable" because that would "ignore[] the possibility that a superior product might be used as a vehicle for tying sales of other products, and would pronounce products superior even where the predominant evidence indicated they were not." 481 F. Supp. at 1003. Instead, the existing 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:
There is no basis in the cases or sound antitrust policy for courts to grant to monopolists the kind of deference Microsoft seeks. Product design questions can be complex, but they are not beyond the competence of courts; courts deal with those and similar questions in product liability, environmental, medical or engineering malpractice, and similar cases.35 [begin page 65] Microsoft would have the court believe that showing any plausible benefit from product design is the same as showing that consumers benefit from the design. That is plainly wrong. A new product design, particularly one that bundles what would be deemed under ordinary tying standards to be two separate products, can never be said unambiguously to benefit consumers unless consumers are given the choice whether to take the bundle. Such a bundled design does not just reduce cost or improve the functioning of one of the products, but rather changes the products’ various attributes. Invariably, a bundled product design will have some pluses (e.g., the kind of one-stop shopping benefits present with any tie-in) and some minuses (e.g., in the case of Windows and Internet Explorer, increased size and impairment of user access to other browsers); and, as with Microsoft’s Windows and Internet Explorer products, some purchasers will prefer the bundle and others will prefer to buy the products separately. As both the Supreme Court and lower courts have repeatedly recognized, it is for the market, not the self-serving assertions of the defendant, to determine whether products are good or bad.36 [begin page 66] It is for these reasons that the Supreme Court made clear in Jefferson Parish, and reiterated in Eastman Kodak, that tie-ins are to be assessed on the basis of consumer demand. Indeed, the defendants in Jefferson Parish, like Microsoft here, argued that the "package" of facilities and services including anesthesiology "d[id] not involve a tying arrangement at all -- that they [were] merely providing a functionally integrated package of services." 466 U.S. at 18-19 (emphasis added). The Supreme Court rejected the argument, recognizing that companies, whether in low or high technology industries, would always be able to show some plausible synergies. See id. at 25, n.41 ("[W]e reject the view of the District Court that the legality of an arrangement of this kind turns on whether it was adopted for the purpose of improving patient care."); see also Multistate Legal Studies, 63 F.3d 1540, 1547 (10th Cir. 1995) (bundled version of two products should not be considered single product simply because the combination represents an "effort to improve the [tying product] by adding elements to it"). The same is true here. Microsoft’s case, resting on the immaterial argument that Windows and Internet Explorer are "integrated," does not warrant summary judgment.
Microsoft places principal reliance for its argument that "technically interconnected products" are essentially "immune" from tying claims on the Court of Appeals’ recent decision in the consent decree case. MS Memo at 20. Microsoft argues that the consent decree decision sets the legal framework for this case: The "court's evaluation of a claim of integration must be narrow and deferential," and "an integrated product should pass muster if there are 'facially plausible benefits to its integrated design.'" MS Memo at 26, quoting slip op. at 14-15. But the Court [begin page 67] of Appeals' holding that facially plausible benefits "pass muster" at the preliminary injunction phase of a lawsuit involving the interpretation of a consent decree neither disposes of nor forecloses full analysis of the legal and factual issues raised here. The principal issue in the consent decree case was whether the challenged tie was an "integrated" product within the specific meaning of language in Section IV(E)(I) of the decree. As addressed by the Court of Appeals, that issue was not coextensive with whether Microsoft bundled one product or two under established antitrust law. The precise issue before the court was whether the government was likely to establish its claim that Microsoft had violated a provision of the consent decree, which exempted the "develop[ment]" of "integrated product[s]" from its scope -- not whether it had violated the antitrust laws. Slip op. at *10 ("the decree does not embody either the entirety of the Sherman Act or even all 'tying' law under that Act," and it cannot be read "as though its animating spirit were solely the antitrust laws," citing United States v. Armour & Co., 402 U.S. 673, 681-82 (1971)). The Court understood its task as construing the decree as a contract, looking to the parties’ intent in determining the appropriate meaning of the word "integrated," which in any event is immaterial here. Slip op. at *19. As the Court elsewhere noted: "The antitrust question is of course distinct. The parties agree that the consent decree does not bar a challenge under the Sherman Act." Id. at *15, n.14 (emphasis added). Moreover, the Court of Appeals expressly left to a more fully developed record whether, in fact, Microsoft's arguments about "technological benefit" could be substantiated, slip op. at 18, and even whether Windows 95 and Internet Explorer are an "integrated" product within the meaning of the consent decree. Id. at 16 & n.15 (Microsoft had "ascrib[ed] facially plausible benefits to its integrated design," which "the [government] may not have contested . . . as vigorously as it might . [begin page 68] . . . The ultimate sorting out of any factual disputes . . . we of course cannot resolve on the limited record before us"). The court's decision was "tentative," id. at 19, and "subject to reexamination on a more complete record." Id. at 18. It accordingly cannot serve as a basis for summary judgment in this case. To be sure, the Court of Appeals did express in dicta concern about the competence of courts to scrutinize the packaging of products in high technology industries such as the computer software business. The court noted that "[a]ntitrust scholars have long recognized the undesirability of having courts oversee product design, and any dampening of technological innovation would be at crosspurposes with the antitrust law." Microsoft, 1998 WL 327855, at *12. But those concerns do not conflict with, and cannot effect an overruling of, Jefferson Parish and its progeny. The Jefferson Parish market test does not require courts to engage in product design oversight or even to look at the relative technological benefits of various products. Rather, the market test enables courts to let the market evaluate the technological benefits and detriments of bundled and unbundled products. Cf. Digital Equipment Corp. v. System Industries, Inc., 1990 WL 5588 (D. Mass. 1990) (fact that two products are technologically interrelated is not determinative under Jefferson Parish separate products test). If there is sufficient demand, then Jefferson Parish requires the defendant to offer the unbundled alternative and to let the market, not the defendant, decide whether the bundle is desirable.
In any event, even if Microsoft were correct that the legal standard to be applied in this case is that technologically tied products are immune from Section 1 scrutiny so long as the "integration" of products produces "plausible" synergistic benefits and the integration is not purely [begin page 69] pretextual, see 1998 WL 327855, at *13, summary judgment in favor of Microsoft would still be inappropriate because there are numerous genuinely disputed material facts.37 As to Windows 98, Microsoft claims that there can be no genuine dispute that the inclusion of Internet Explorer in Windows 98 achieves technologically beneficial results. MS Memo at 33. However, the facts cast serious doubt on the assertion that there are any necessary synergistic benefits of the bundled version, and that Microsoft’s requirement that Internet Explorer be used is not pretextual. First, as discussed above, there is no compelling evidence that requiring the use of Internet Explorer in particular, as opposed to any other browser inserted by an OEM or end user, is necessary to provide the benefits Microsoft describes. For example, the other operating system vendors which have incorporated browsing functionality into their operating systems universally respect "browser neutrality" in doing so, either by bundling multiple browsers or contractually and [begin page 70] technically permitting OEMs and end users to replicate the synergies of bundling by substituting browsers different than the one selected as default by the operating system manufacturer. Second, Microsoft cannot claim without serious dispute that all (or even most) users consider the integration beneficial; there is evidence that certain users (e.g., classes of corporate customers) do not prefer the integration -- that is, they want an unbundled version. See also SJ Ex. 6. Finally, there is a genuine issue of material fact whether, under the standard applied by the D.C. Circuit to the consent decree, Windows 98 and Internet Explorer were essentially "bolted" together. As the United States has previously discussed, see PI Memo at 60-64, numerous Microsoft documents describe the bundling of Internet Explorer with Windows 98 as precisely motivated by a desire to thwart competition among browsers (as opposed to maintaining a competitive advantage among operating systems) -- and even to specifically "weld" the products together in response to this desire. In light of these factors, even under the unduly narrow standard articulated by the Court of Appeals under the consent decree, disputed issues of material fact preclude summary judgment.
The Complaints allege that Microsoft’s forced licensing of Internet Explorer as a condition of OEMs licensing its Windows operating systems significantly injures other browsers and, both alone and in concert with Microsoft’s other anticompetitive conduct, violates Section 2 of the Sherman Act. It constitutes both unlawful monopolization of the PC operating system market -- because it injures other browsers that threaten to provide an alternative platform that would [begin page 71] diminish the monopoly power of Microsoft’s operating systems -- and an unlawful attempt to monopolize the market for Internet browsers. "‘The offense of monopoli[zation] under § 2 of the Sherman Act has two elements: (1) the possession of monopoly power in the relevant market and (2) the 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.’" Eastman Kodak Co. v. Image Tech. Servs., Inc., 504 U.S. 451, 481 (1992) (quoting United States v. Grinnell Corp., 384 U.S. 570-71 (1966)). The offense of attempted monopolization under Section 2 has three elements: "that the defendant (1) has engaged in predatory or anticompetitive conduct with (2) a specific intent to monopolize, and (3) a dangerous probability of achieving monopoly power. Spectrum Sports v. McQuillan, 506 U.S. 447, 456 (1993). In substance, Section 2 prohibits the maintenance or creation of monopoly power by anticompetitive means. See, e.g., Aspen Skiing, 472 U.S. at 602-603 (Section 2 condemns maintenance of power when accomplished through improper or anticompetitive means); Barry Wright Corp. v. ITT Grinnell Corp., 724 F.2d 227, 239 (1st Cir. 1983) (Breyer, J.) (explaining that "[a] monopolist’s conduct that from a competitive point of view is unreasonable" violates Sherman Act section 2). As discussed above, Microsoft argues largely that its licensing of Windows on condition that OEMs license Internet Explorer is not an unlawful tying arrangement because Internet Explorer and Windows comprise a single product. MS Memo at 38-48. But, whatever the merits of that argument under Section 1, it does not follow that Microsoft’s forced licensing of Internet Explorer cannot be challenged under Section 2. As the leading antitrust treatise explains:
10 P. E. Areeda, E. Elhauge, and H. Hovenkamp, Antitrust Law ¶ 1752g, at 289 n.46 (1996) (emphasis added). The very "technological tying" cases Microsoft invokes hold that Section 2 may be violated when a monopolist "uses" its monopoly "power to tighten its hold on the market." Berkey Photo, 603 F.2d at 274-75; Northeastern Tel. Co. v. American Tel. & Tel. Co., 651 F.2d 76, 84-85 (2d Cir. 1981), cert. denied, 455 U.S. 943 (1982); Telex, 510 F.2d at 926-27. Those cases also make clear that, in determining whether a product design decision amounts to an unlawful use of monopoly power (or, in the language of other cases, is "predatory" or "exclusionary"), the usual Section 2 standard applies and not some special, deferential rule. See Northwest Telephone, 651 F.2d at 84-85 & 95 n.29; In re IBM Peripherals EDP Devices Antitrust Litigation, 481 F. Supp. 965, 1003 (N.D. Cal. 1979) (rejecting the argument that illegality should be found only when "the intent was solely an illegal one" and concluding that "[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.), cert. denied, 464 U.S. 955 (1983); see also 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 to affect consumer choice," and "an absolute monopolist’s design conduct is subject to antitrust [begin page 73] scrutiny by a jury"); California Computer Products, Inc. v. IBM Corp., 613 F.2d 727, 735-36 (9th Cir. 1979). Microsoft argues that its refusal to license Windows except with Internet Explorer is simply an "ordinary business practice[] typical of those used in a competitive market" and, for this reason, cannot "constitute anti-competitive conduct violative of Section 2." MS Memo at 76 (quoting Trace X Chem., Inc. v. Canadian Indus., Ltd., 738 F.2d 261, 266 (8th Cir. 1984), cert. denied, 469 U.S. 1160 (1985)). This contention, however, is both factually and legally flawed. As noted above, the evidence shows that other operating system vendors permit OEMs to license their operating system without a particular browser, and to remove browsers shipped with the operating system. Moreover, other operating system vendors, in contrast to Microsoft, lack monopoly power; and it has long been clear that, "[w]here a defendant maintains substantial market power, his activities are examined through a special lens: Behavior that might otherwise not be the concern to the antitrust laws -- or that might even be viewed as procompetitive -- can take on exclusionary connotations when practiced by a monopolist." Eastman Kodak, 504 U.S. at 488 (Scalia, J., dissenting on other grounds) (quoting 3 P.E. Areeda and D.F. Turner, Antitrust Law ¶ 813, at 300-02 (1978)).
Thus, the Section 2 issues cannot be disposed of by looking merely to the practices of firms other than Microsoft. Instead, they must be resolved by asking the question that Section 2 requires -- whether Microsoft’s insistence that OEMs license Internet Explorer as a condition of licensing Windows is anticompetitive.38 In making that assessment, the Supreme Court explained [begin page 74] in Aspen, the court should inquire whether the monopolist’s conduct "‘(1) tends to impair the opportunity of rivals’" and (2) "’does so in an unnecessarily restrictive way.’" 472 U.S. at 605 n.32 (quoting 3 P.E. Areeda & D.F. Turner, Antitrust Law 78 (1978)). In other words, Section 2 is violated when a monopolist engages in conduct that excludes rivals and the exclusion is not reasonably necessary to achieve asserted legitimate business objectives. See In re IBM Peripherals EDP Devices Antitrust Litig., 481 F. Supp. at 1003 ("If the design choice is unreasonably restrictive of competition, the monopolist’s conduct violates the Sherman Act. The standard will allow the factfinder to consider the effects of the design on competitors; the effects of the design on consumers; the degree to which the design was the product of desirable technological creativity; and the monopolist’s intent, since a contemporaneous evaluation by the actor should be helpful to the factfinder in determining the effects of a technological change."). The evidence here establishes without question material disputes on these issues. First, the evidence leaves no doubt that, because of Microsoft’s monopoly power, OEMs have no alternative to the Windows operating system and, thus, to licensing Internet Explorer; that OEMs that license Internet Explorer are less likely as a result also to distribute other browsers; and that the forced licensing of Internet Explorer thus forecloses the opportunities of rival browsers. See supra, Section II.B. The exclusion of non-Microsoft browsers in this way is not a "competitive advantage" that "‘accrues to any integrated firm,’" MS Memo at 80 (quoting Berkey Photo, 603 F.2d at 276), but rather is a consequence of Microsoft’s desktop operating system monopoly. [redacted material] [begin page 76] Second, the evidence supports the conclusion that this exclusion of non-Microsoft browsers was not reasonably necessary in order for Microsoft to achieve any legitimate business objective. Microsoft could offer OEMs the option of a Windows product without Internet Explorer being required to supply browsing functionality and/or easily permit OEMs to remove the means of access to Internet browsing functionality. See supra, Section II.C.1. It chose to tie Internet Explorer, instead, precisely in order to deny its browser rivals access to the OEM distribution channel. [redacted material] This belief in the effectiveness of the tie has remained unwavering. See, e.g., [redacted material] This evidence, and the factual disputes it demonstrates, preclude summary judgment on
the Section 2 claim.
|
|
