|
|
|
IN THE UNITED STATES DISTRICT COURT
DECLARATION OF TABLE OF CONTENTS Table of Contents
Declaration of David S. Sibley I. Qualifications and Introduction 1. My name is David S. Sibley. I am the John Michael Stuart Centennial Professor of Economics at the University of Texas at Austin. I received the degree of B A. in Economics from Stanford University in 1969 and a Ph.D. in Economics from Yale University in 1973. In addition to my current teaching responsibilities, I have taught graduate level courses in economics at the University of Pennsylvania and Princeton University. Prior to joining the University of Texas. I was Head of the Economics Research Group at Bell Communications Research. I have also served as a Member of the Technical Staff in economics at Bell Laboratories. During the last twenty-five years, I have carried out extensive research in the areas of industrial organization, microeconomic theory, and regulation. My publications have appeared in a number of leading economic journals, including the Journal of Economic Theory, Review of Economic Studies, Rand Journal of Economics, American Economic Review, Economeirica, and the International Economic Review, among others. I am also a co-author (with Steven J, Brown) of a leading textbook on monopoly pricing, THE THEORY OF PUBLIC UTILITY PRICING, which was first published by Cambridge University Press in 1986 and is now in its fourth printing. A copy of my curriculum vitae is attached to this declaration.1 2. I have been asked by the Antitrust Division of the United States Department of Justice to examine the competitive effects of Microsoft Corporation's ("Microsoft's") contractual restrictions (including the bundling of its Internet browser with the Windows 98 operating system) in its agreements with personal computer original equipment manufacturers ("PC manufacturers" or "OEMs"), Internet access providers ("IAPs"), and Internet content providers ("ICPs"), In particular, I examined Microsoft's bundling and other contractual practices to determine whether these actions represent anticompetitive behavior undertaken by Microsoft to maintain a monopoly in the market for operating system software used to run Intel-compatible desktop PCs ("PC operating system market," or "OS market"), I also studied whether Microsoft's actions represent anticompetitive behavior undertaken to establish a monopoly in the market for browsers. 3. In conducting this analysis, I examined: (1) public documents and websites containing relevant information; (2) certain confidential Microsoft and third party documents supplied to me by the Department of Justice; (3) deposition transcripts of personnel from Microsoft and other companies, including OEMs, IAPs, and ICPs; and (4) relevant books and articles from the economic literature in antitrust and industrial organization. 4. In analyzing the competitive affects of Microsoft's conduct, I proceeded in three steps. Each of these steps, and the conclusions drawn from my analysis of relevant data, are summarized as follows. First, I examined whether Microsoft possesses a dominant share of the PC operating system market, and whether it has monopoly power in that market. I found that Microsoft has monopoly power because of entry barriers in a bottleneck input in computing -- the operating system. Second, I examined whether Microsoft's contractual practices affecting browser choice would be profitable on their own, regardless of its monopoly power; or, alternatively, if the profitability of its practices depended on impeding competition in the operating system and browser markets. In particular, Microsoft appears to be concerned about the potential of the browser to serve as a software applications platform that would run on operating systems other than those produced by Microsoft. I conclude that, if Microsoft were not attempting to impede this type of competition, it would encourage the sale of all browsers (including those of its rivals), since greater browser sales would increase the demand for its operating system and increase the profit which could be obtained from users wanting to access the Internet (e.g.,, by increasing the price of the OS). In fact, I found Microsoft's contractual practices constrained consumers' choice of browsers, thus resulting in a diminution of profits that could most plausibly be recouped only because of competitive restraints in the OS market. Importantly, these contractual restrictions have minimal or non-existent offsetting efficiency justifications. Third, I examined whether Microsoft likely can recoup the reduction in profits resulting from its restrictive practices. I found that Microsoft is likely to recover its lost profits by blunting direct and indirect competitive threats to its OS monopoly and by gaining market power in Internet-related markets, Microsoft's restrictive practices add to the natural entry barriers associated with the OS market by inhibiting the use of competing internet browsers as software applications platforms that are independent of any particular underlying operating system. This deters future entry into the OS market and protects Microsoft's current monopoly power. 5. An alternative line of reasoning leads to the same conclusion as that drawn from the three-step analysis described above. Because contractual agreements with OEMs, IAPs, and ICPs impose exclusionary restrictions on the use of a competing browser, Microsoft blunts the threat to its OS monopoly posed by the browser's potential as an alternative platform for software applications. Based on my review of internal Microsoft documents, these exclusionary restrictions were imposed with this threat very much in mind. Furthermore, the analysis presented below demonstrates that these restrictions have few apparent purposes aside from discouraging the use of competing browsers. These findings, as well as the three-step analysis outlined in paragraph four, lead me to conclude that Microsoft's contractual practices constitute anticompetitive behavior. 6. The development of the high-tech, computer industry has had a profound positive effect on the nation's economy and the economic welfare of all consumers. The growth of software and computer companies has accounted for almost thirty percent of the increase in gross domestic product since 1994, and it is estimated that commerce conducted over the rapidly expanding Internet will surpass $180 billion dollars annually in the beginning of the next millennium.2 The absence of effective competition in two important segments of the software industry (PC operating systems and browsers) will not only affect the price that consumers pay for these products but, perhaps more importantly, it will also change the pattern of innovation that has characterized this dynamic industry. Innovation is a driving force in economic progress because it affects the quality of existing products in the marketplace and the development of new and better products. Actions taken to increase competition in this industry may well be among the most critical economic policy decisions facing the nation today. The artificial entry barriers erected by Microsoft (through its restrictive practices, including bundling) will have the effect of slowing or halting the natural tendency of the marketplace to provide alternative technologies, reducing the welfare of many consumers. If current trends in market share continue, the browser market is likely to tip to a Microsoft monopoly soon, eliminating the competitive threat Microsoft faces from Internet browsers that have the ability to run software applications independent of any particular operating system. If this happens, the Microsoft browser may well become the bottleneck input for Internet-related markets. II. Microsoft Has Monopoly Power
7. For the purposes of antitrust analysis, the relevant operating system market is the market for personal computer operating systems that are compatible with Intel x86/Pentium (or Intel-compatible) microprocessors. This definition is motivated by the product market definition given in the U.S. Department of Justice and Federal Trade Commission Horizontal Merger Guidelines, and is, in my view, consistent with accepted economic theory.3 A proposed product market defined under the Merger Guidelines is accepted as corret if a price increase in the proposed market would not cause demand to shift to products outside the proposed market to such an extent as to dissuade a hypothetical profit maximizing monopolist from imposing a "small but significant and nontransitory" price increase.4 Because I am evaluating here an allegation of maintaining monopoly power as opposed to evaluating a merger, I adopt this definition with the clarification that the initial price level prior to the price increase be a competitive price level. 8. In order to implement this methodology, I begin by noting that the licensing fee for the vast majority of the operating systems currently licensed by Microsoft to OEMs is approximately $65.5 If a hypothetical monopolist were to increase price by ten percent above this level6, this would represent an increase of no more than four-tenths of one percent in the price of a $1,500 computer, (For a $1,000 computer, the additional $6.50 represents an increase in price of no more than seven-tenths of one percent.) It seems implausible that a price increase of less than one percent would cause a significant number of buyers of PCs with Intel (or Intel-compatible) microprocessors to switch to Apple or UNIX PCs not using an x86 or Pentium chip, let alone forego the purchase of a PC entirely. Industry experts corroborate this conclusion.7 Therefore, a hypothetical monopolist that was the only supplier of operating systems for Intel and Intel-compatible PCs would find it profitable to increase price by ten percent over a competitive level The reasoning is that for an end user to switch from the Intel-based PC to an Apple computer, for example, usually requires the end user to incur substantial switching costs, in both time and money, to learn to use the different system. Moreover, as discussed below, even if the product market were more broadly defined to Include desktop operating systems for all PCs, such as those offered by Apple and UNIX that do not use Intel (or Intel-compatible) microprocessors, Microsoft's market share would fall only slightly (see discussion below) and its monopoly power would remain.8 9. Internet browsers are in a separate relevant product market. An Internet browser is a specialized software program that allows computer users to locate, access, display and use applications located on the Internet's World Wide Web or on internal corporate computer networks known as intranets.9 Today, a browser is the predominant conduit to the Internet and allows computer users to access and view "pages" on the World Wide Web. Browsers constitute a separate product market because a distinct consumer demand exists for browsers, apart from operating systems, as evidenced, for example, by the many PC users who obtain the Internet browsers they use as a stand-alone product apart from the OS.10 In fact, many corporate users purchase only an operating system and do not want a browser at all.11 Moreover, OEMs have a separate demand for a particular browser apart from the Microsoft operating system, as evidenced by CEM's desire to remove the version of Internet Explorer ("IE") packaged with Windows 95 (with the option of replacing IE with another browser).12 10. The view that Internet browsers constitute a separate product market is further supported by the fact that browsers are marketed and distributed independently from PC operating systems. Significantly, Microsoft independently markets, distributes, and monitors the distribution of its Internet Explorer browser software even though it is also bundled with its Windows 95 operating system. In addition, Microsoft intends to offer a version of IE (referred to as IE 5.0} after Windows 98 is released.13 Microsoft's marketing and distribution channels for IE provided apart from its operating system are numerous: IE is available on the Microsoft website for consumers to "download" directly to their PCs; Microsoft has agreements with IAPs that allow them to distribute IE to their subscribers; and Microsoft sells IE as a stand-alone product at retail software outlets. Microsoft has even developed and distributed versions of the IE browser for Apple Computer's Macintosh and the Sun Solaris operating system, both non-Microsoft operating systems. Finally, Netscape, Microsoft's principal browser competitor, also distributes its browser product apart from an operating system. 11. In many software markets, including the OS market, there are fundamental forces which may lead to one firm being dominant at a given time and which tend to create barriers to entry. These forces have been widely discussed in the economics and computing literature. Very briefly, they are as follows:
Economies of Scale. For complex software such as an OS. the initial, or first-copy, costs to writing software are often very large, whereas the incremental cost of producing additional copies is small. Hence, average cost declines as the scale of output rises. Both economies of scale and increasing returns in consumption, when combined with the lock-in effect created by consumer switching costs discussed below, create natural barriers to entry, 12. Increasing returns to consumption and economies of scale give rise to a phenomenon which lies at the heart of antitrust analysis of network industries: monopoly tipping.15 If a large set of users adopts a new network technology, then that technology becomes more attractive to everyone else due to increasing returns in consumption. As more users join, the technology becomes still more attractive until it becomes dominant; in economic terminology, the market has "tipped" to the new technology. Once tipping has taken place, there are forces that tend to keep the victorious technology dominant until a new, much improved product is developed; incremental improvements are unlikely to be successful. Users invest time and money in learning to use a given technology proficiently. If at some future date a still "newer" technology were to become available, potential users of the newer technology would have already incurred those learning costs on the older one. Therefore, in considering whether or not to adopt the newer technology, potential users would weigh the possible benefits of the newer technology against the switching costs of learning that technology, given that they already know the old technology. Thus, for the newer technology to succeed, it would have to offer a substantial improvement in performance, i.e., enough of an improvement at least to overcome the switching costs associated with the change. In the normal course of markets and competition, such improvements do, in fact, occur. An example is the displacement of slide rules with pocket calculators. 13. The economic theory of network effects describes well the performance of the OS market. As Microsoft itself has stated: "The availability of a rich variety of quality applications software that will ran on a particular operating system is fundamental to its success. This fact has been recognized by publishers of operating systems for years."16 In the computer industry, the theory of network effects is known as "positive feedback." The concept of positive feedback recognizes that as an operating system gains popularity, the incentive to develop software for that operating system grows since the number of potential customers for the application developer is larger. This, in turn, increases the value of the operating system (and likely its market share) which is determined by the quality and variety of software applications written for it. As the OS gains market share, software developers find it even more advantageous to produce additional applications for that system, and so on. As noted by Dr. Nathan Myhrvold, a Senior Vice President of Microsoft, the positive feedback effect has been responsible for the phenomenal strength of leading software products in both applications and operating systems,"17 Dr, Myhrvold also notes that the exchange of data between computer users is facilitated by computer and computer software compatibility. He observes "the laws of positive feedback govern any system where compatibility with other users is either directly or indirectly a key factor in the utility of a product or service."18 Positive feedback explains the importance of the number of complementary software applications and the installed base of these applications as a natural barrier to entry, and also why alternative operating systems already in the market at a small scale (such as Unix) are not effective competitors, 14. A predictable consequence of the installed base of applications is evidenced by Table One, which presents market share information for operating systems that rely on Intel x86/Penttum (or Intel-compatible) microprocessors, Microsoft's market share in each period from 1991 to 1997 holds consistently at about ninety percent.19 Microsoft's OS dominance is stable; has hardly fluctuated in the face of determined attempts at entry by rival operating systems; and is forecast to remain stable in the future.20 Table One
15. To illustrate the importance of the installed base as an entry barrier, it is instructive to review the experience of IBM's OS/2 operating system. This operating system, first released in 1987, was designed to compete with Microsoft Windows. In 1994 and I995} IBM installed OS/2 on its own Aptiva line of computers, but retailers and consumers routinely demanded software applications that were not available on OS/221 In response, IBM abandoned its practice of installing OS/2 on the Aptiva computer, choosing instead to license Windows from Microsoft.22 Similarly* Packard Bell negotiated with IBM in 1994 to license OS/2 for preinstallation on the Packard Bell PC because Packard Bell was interested in the possible price advantage that OS/2 might offer. But, when IBM stated "in addition to our operating system you need Windows"23 to ensure software compatibility, Packard Bell decided to license Windows from Microsoft directly and not install OS/2, 16. The importance of an installed base of applications software as an entry barrier is further demonstrated in the analysis of Mr. Joachim Kempin, Senior Vice President of OEM Sales for Microsoft. In discussing what could "derail" Microsoft's OS pricing strategy, Mr. Kempin stated:
Our high prices could get a single OEM (Compaq might pay us 750M$ next year) or a coalition to fond a competing effort (say in India). While this possibility exists I consider it doubtful even if they could get a product out that they can market it successfully, leapfrog us and would not deviate from their own standard to differentiate. Could they convince customer [sic] to change their computing platform is the real questions [sic]. The existing investments in training, infrastructure and applications in windows computing are huge and will create a lot of inertia. No kindling of OS on low end systems would be the easiest way to hurt us - but who would want to start with this and loose [sic] business.24 17. Based on the above discussion, it should be clear that in markets such as OS software, certain economic forces can lead naturally to dominance by a single firm, even apart from exclusionary restrictions. (There are, of course, other countervailing factors that tend to perpetuate competition, such as product differentiation,) ff a single firm does achieve dominance, these forces lead to entry barriers which protect that dominance. In themselves these facts do not warrant competitive concern. Preeminence in a market is the market's own way of rewarding excellence. Indeed, there is no reason to think that the market, left to itself, will not generate alternatives to Microsoft's operating system that will be sufficiently superior to overcome the entry barrier advantage that Microsoft enjoys. For example, in an insightful article, Evans and Schmalensee25 point out how tipping was reversed in the case of stereo long playing record players, a technology with a huge installed base of recordings, by the CD technology, which started with no recordings at all. 18. The bundling and other contractual browser restrictions that Microsoft insists upon in its agreements with OEMs, IAPs, and ICPs add artificial entry barriers to those that occur naturally, and are therefore a source of competitive concern. These artificial entry barriers have the effect of slowing or halting the natural tendency of the marketplace to provide alternative technologies that compete with Microsoft. Specifically, these restrictions are currently threatening the potential of competing Internet browsers that can run software applications independently of the Windows operating system and that therefore challenge the dominance of the Windows OS. Bill Gates recognized this threat when he stated: "They (Netscape] are pursuing a multi-platform strategy where they move the key API into the client [browser] to commoditize the underlying operating system." 26 This is the subject of the next section. III. Microsoft's Bundling and Contractual
Restrictions 19. In this section, I carry out the second step of the three-stage analysis outlined in paragraph four above. After describing the browser restrictions contained in the Microsoft agreements with OEMs, IAPs, and ICPs, I analyze their competitive effects and conclude that they are likely to enable Microsoft to monopolize the browser market, thereby removing the Internet browser as a software platform that can exert competitive pressure on the OS market where Microsoft has monopoly power. I also examine possible legitimate business justifications for the browser restrictions and conclude they are likely to be minimal or non-existent,, so that the main effects of the restrictions are to reduce profit to Microsoft from browsers and the Internet. Such practices therefore appear to make sense only in their entry-deterring effects in the OS market. Finally, I address the issue of Microsoft's distributing its browser without charge,
20. Table Two presents a summary of the restrictions in Microsoft agreements with OEMs, IAPs (including online service providers and Internet service providers ("ISPs")), and ICPs which have anticompetitive effects. Taken together, these constitute a powerful force against the use of non-Microsoft browsers in four main distribution channels. (Whether or not these restrictions have countervailing benefits will be discussed in Section IZI.B.) 21. Consider first the OEM agreements. To obtain licenses to preinstall Windows on the PC, OEMs have agreed to restrictions designed to deter them from preinstalling and promoting a competing browser. The requirement that the OEM not modify the OS software has been used to prevent OEMs from offering the OS without the IE browser. It also prevents OEMs from removing the IE icon or replacing it with the Table Two Original Equipment Manufacturer Agreements:
Online Service Folder Agreements:
ISP Internet Referral Server Agreements:
Internet Content Provider Agreements:
icon of a competing browser.27 This requirement is exclusionary even though the OEM has the option of installing a competing browser. While it is true that OEMs may place two browser icons on the desktop, OEMs are generally reluctant to do so for two reasons. First, OEMs believe that the two icons would cause customer confusion, thereby increasing their own "customer support" costs.28 Second, the desktop constitutes scarce real estate to OEMs. As such, OEMs are generally reluctant to preinstall more than one product in each functional category.29 22. Microsoft's Windows license agreements also require that the folders or icons added to the desktop must be substantially the same size and shape as those delivered by Microsoft. This requirement appears innocuous by itself, but biases browser choice since OEMs cannot remove the IE icon. An OEM that wanted to feature a competing browser, but still had to keep the IE icon on die desktop, might wish to have a larger icon for its preferred browser so as to minimize customer confusion and the resulting support calls; however, limitations on size and shape of icons prevent this and bias the OEM away from promoting or preinstalling a competing browser. 23. Three additional OEM browser restrictions affect the initial boot and startup screen, and should be viewed as a group, OEMs cannot alter the content or sequence of screens viewed by the end user on the initial boot of the PC; OEMs cannot insert a screen or image of their own during startup; and OEMs cannot cause a non-Microsoft program to execute on the initial boot, as they could IE, for example. One effect of this set of restrictions is to make the initial boot confusing for unsophisticated users who have a PC with a rival browser preinstalled. In Windows 95, the startup screens all refer to Windows and! IE so that even with another browser preinstalled, the end user cannot see any reference to it until startup is complete and the desktop appears, showing the icon of the preinstalled competing browser. More generally, these restrictions prevent OEMs from featuring another browser or an IAP that promotes a competing browser during the startup sequence prior to the Welcome to Windows screen where Microsoft promotes its own IAP business partners (see paragraph 25 below). Perhaps more importantly, the startup screen restrictions help to strengthen Microsoft's bargaining power with IAPs and ICPs. If OEMs could make their own deals directly with lAPs or ICPs and feature them on screens during the startup sequence, this would tend to devalue Microsoft's provision of access to the Windows desktop. (As I will describe below, the value of space on Microsoft's channel bar, the Online Service Folder, or the Internet Connection Wizard is what induces IAPs and ICPs to agree to Microsoft's restrictions on their use and promotion of competing browsers.) 24. Microsoft's Online Service Folder Agreements provide the three major online services and AT&T Worldnet with a highly desirable position on the Windows desktop - their Internet access software in the Online Service Folder - in exchange for agreeing to restrictions which deter them from distributing or promoting competing browsers.30 The agreements typically specify that an online service provider must distribute IE and not any third party browser with physical distributions of its software and promotional material unless a customer requests that third party browser. The online service provider is not permitted to ship a non-Microsoft browser with more than fifteen percent of its physical distributions of software even if more than fifteen percent of its customers were to request another browser. The online service folder agreements require that the online service provider place download links to the IE download site on its website and place limitations on a provider's ability to provide links to use and to download third party browsers. Again Microsoft's requirements make it more difficult for the end user to download a competing browser than to download IE. While the quantitative importance of this may not be large, it does require the online service provider to bias website visitors' selection of browsers. 25. Microsoft's Internet Referral Server Agreements include the ISP in a list of providers that are shown to an end user who selects his Internet access provider using a process called the Internet Connection Wizard, which is bundled with the Windows operating system. The Internet Connection Wizard can be invoked by the end user not only from the desktop but also from the Welcome to Windows screen -- that is the first screen end users see after initiating their PCs for the first time. In return for this preferential access to potential subscribers, the ISP must not express or imply to any potential subscriber (whether or not the subscriber was obtained from the Internet Connection Wizard) that an alternative browser is available unless specifically requested by the subscriber. The ISP must also accept limitations on links to its home page that allow users to download non-Microsoft browsers. The ISP must display the IE logo on its home page, along with a "hot link" to the Microsoft IE web page, and must use the IE logo in its promotional and packaging material. These requirements make it easier for an end user to download IE from an ISP's home page than to download another browser, thus requiring the ISP to bias its website visitors' browser choice. The Internet Referral Server Agreements typically also specify that the ISP must include IE as the only browser shipped seventy-five percent of the time. Thus, when more than twenty-five percent of its customers prefer another browser, an ISP is restricted from providing a third party browser to its customers even if they requested it to do so. 26. Microsoft's Internet Referral Server Agreements also provide discounts on referral fees for use of IE-specific technologies, thereby providing ISPs a financial incentive to configure their own web pages in such a way so as to reduce cross-platform threats to Microsoft's OS monopoly. For example, one of the IE-specific technologies is ActiveX controls31 My understanding is that ActiveX controls are generally operating system specific. That is, only Windows users can run an ActiveX control developed for Windows embedded in an ISP website. For a competing browser to support ActiveX controls would thus encourage the development of applications on the Internet that are Windows specific, thereby blunting one of the advantages of Internet browsers - the ability to run applications on any operating system. 27. The Internet Content Provider Agreements grant ICPs placement on the Channel Bar that appears on the Windows desktop.32 In return, the ICPs must implement certain IE-specific technologies on their websites. This requirement is similar to that imposed on ISPs and has the same anticompetitive effect of reducing the cross platform threat to Microsoft's OS, The ICP agreements also specify that the ICP must promote IE and the Active Desktop, and display the IE logo and links to IE on its home page. These requirements make it easier to download IE from an ICP's home page than it is to download another browser, thus requiring the ICP to bias its customers' browser choice. The ICPs are required to distribute IE exclusively, eliminating these ICPs as a distribution channel for a competing browser. The ICP cannot provide non-paid promotions or endorsements of non-Microsoft browsers on its Web and Content pages. Nor can the ICP pay or receive payment for any Content agreement with a non-Microsoft browser company. These requirements taken together limit the possible financial gains that an ISP and a competing browser supplier could achieve by an agreement with each other. 28. As of August 2, 1996, Microsoft had signed approximately 2,500 "IE preferred" licenses with providers of access to the Internet.33 An internal Microsoft document lists the top severity-five Internet access providers of North America by dial-up users, and the status of their Internet Explorer agreements with Microsoft.34 According to this document, Microsoft entered into agreements that require a minimum distribution of IE as a percentage of browsers distributed with fifteen of the top seventy-five providers, These fifteen providers offered access to eighty-nine percent of the dial-up users served by the top seventy-five firms.35 Of the top seventy-five providers, fifty-three had signed agreements Microsoft termed "IE preferred" in which IE is the "preferred and default browser" ISPs distribute to their customers.36 These data demonstrate that Microsoft's restrictive agreements are with those ISPs that are the first point of Internet access to the majority of PC users. 29. Table Three contains market share data for the IE and Netscape Navigator browsers for the period February 1996 through February 1998. Initial versions of Internet Explorer (versions 1.0 and 2.0) were not comparable to the Netscape browser and did not capture significant share. With the release of Internet Explorer 3.0 in May 1996, Microsoft first offered browser features and quality comparable to those of Netscape, and its market share grew rapidly during the period May 1996 through August 1996. During this period, Microsoft went from the "alpha release" to the commercial release of Internet Explorer version 3.0. It was also during this period that Microsoft began to sign the agreements with ISPs that contained the restrictions listed in Table Two. The ICP restrictions listed in Table Two were contained in agreements that became effective between June and September of 1997. With the release of Microsoft's Internet Explorer 4.0, its browser market share has continued to climb, to the point where its share and that of Netscape are roughly equal. As shown in Table Three, Microsoft's share of the browser market increased from three percent to four percent in early 1996, to twenty percent in early 1997, to more than fifty percent in early 1998.
TABLE THREE
You see browser share as job 1. The real issue deals with not losing control of the APIs on the client and not losing control of the end-user experience. For Netscape this is synonymous with winning the browser battle. That is because they don't have Windows. We have an asset which has APIs and controls the end-user experience: Windows. This passage unifies the exclusionary practices presented in Table Two. The bundling of the IE browser is precisely the "integration" that Mr. Allchin describes and, as he states, its purpose was for IE to win a competitive battle with Netscape that (apparently) could not be woo based on price and quality. The other restrictions perform the same role: by linking the exclusion of competing browsers to space on the desktop and in the Internet Referral Server, Microsoft's agreements with OEMs, IAPs, and ICPs perform the essential function of leveraging" Windows. 31. Microsoft would be correct in pointing out that the anticompetitive practices in question do not remove completely the ability of final consumers to obtain rival browsers. For example, notwithstanding Microsoft's practices, it remains theoretically possible for Netscape to continue distributing its browser via download from its website. A number of consumers do obtain it in this way, though for many the greater time and expense of downloading (not to mention the nervousness of less computer-literate users who may be wary of trying to obtain their browser in this way), make obtaining the browser directly from an IAP or an OEM the preferred choice. Moreover, in those cases where an OEM has been willing to bear the added customer confusion and support costs, competing browsers can be obtained preinstalled on a PC as well. And, of course, Microsoft would be correct in pointing out that there are at least some iAPs and ICPs with which it does not have exclusionary licenses. Finally, the screen and boot-up restrictions do not present an impenetrable barrier to a consumer's ability to access competing browser technologies. Nevertheless, Microsoft's practices, although they undeniably stop short of choking off all possible means by which competing browsers can be obtained, collectively amount to a strong "thumb on the scale" that significantly advantages Microsoft, disadvantages competitors, and restrains effective competition. Furthermore, in markets subject to network effects, such as those at issue here, even modest artificial advantages are capable of skewing the browser market Microsoft's way and making it significantly less likely that a competing browser will be able to weaken Microsoft's current dominant position in the OS market.
32. At this point, I will consider possible efficiency justifications for Microsoft's browser restrictions. I will analyze these restrictions by asking two questions. First, are there any efficiency justifications that apply in this case? Second, if one does apply, could the legitimate business purpose of that restriction be served without restricting other parties' distribution and promotion of a competing browser?
33. The most important OEM restriction is that the OEM not modify the operating system software as provided by Microsoft, This restriction appears justifiable to the extent that it performs a quality assurance function by prohibiting OEMs from modifying the OS in ways that undermine Microsoft's ability to provide ISVs with a consistent platform to which they can write applications. If each OEM were to modify the OS software it receives from Microsoft in different ways so as to create different versions of Windows, ISVs might have to rewrite their applications for each version. Microsoft refers to this scenario as "balkanization" of its operating system.39 34. However, the desire of Microsoft to provide ISVs with a consistent applications development platform does not provide economic justification for biasing the OEMs' choice of which browser to feature. In the case of Windows 95, Microsoft itself provided the ability to remove Internet Explorer 3,0 and Internet Explorer 4.0 using an uninstall utility called "Add/Remove" supplied in Windows, presumably with the intent that such a removal feature would be used.40 If removal would have undermined the consistency of the platform for ISVs in any meaningful way, it seems doubtful that Microsoft would have provided such a removal ability. In the case of Windows 98, it is my understanding that the ability to browse the web using IE can be removed and replaced with a competing browser in such a way that the consistency of the Windows platform for ISVs would not be frustrated in any appreciable manner. 35. Nor can the restriction be justified in the case of Windows 98 by Microsoft's claim that IE and Windows 98 are integrated in a manner designed to give users a superior integrated browsing experience from the Windows desktop to the Web. It is my understanding that a comparable browsing experience can be achieved by the combination of Windows 98 and a competing browser. Furthermore, even if this were not true, restrictions on the use of competing browsers would be unnecessary because end users would select the integrated Microsoft product of their own volition, if it were superior. 36. The anticompetitive effects of the screen restrictions were discussed above in paragraphs 22 and 23. One possible reason for these restrictions is that they prevent OEMs from obtaining referral fees from IAPs through an OEM-sponsored ISP referral server (similar to Microsoft's ISP referral server) presented to the end user during the initial boot. My impression, based on Microsoft documents, is that this motive is a minor one.41 In any case, given its OS monopoly, Microsoft could probably recapture such referral fees by charging an OEM for the right to insert its own screens or alter those provided by Microsoft during startup. Restrictions that tilt the scales against competing browsers are not essential to Microsoft's retention of TAP referral fees that might otherwise go to OEMs. 37. A second possible justification for the screen restrictions is that there may be marketing or quality benefits to be obtained from keeping the Window's startup sequence and desktop uniform across all users, regardless of whose PC the end user is buying. For example, Microsoft devotes resources to ensuring that there are no failures during the initial boot. If OEMs were allowed to put in their own screens and change the icons on the desktop, it is conceivable that this could, for example, result in a longer initial boot-up process or increased confusion lo consumers which could reflect poorly on Microsoft. This is a variation on the ISP balkanization argument discussed above but applied to end users. 38. As an initial matter, replacement of the IE icon with the icon of a competing browser would not appear to interfere with any marketing or quality benefits associated with uniformity of the startup sequence and the desktop screen. In fact, the screen restrictions as applied to Windows 98 support this view since Microsoft does permit the OEM to add items of any shape and size to the desktop screen if the OEM turns on the Active Desktop as the default configuration on its PCs. I therefore see no legitimate business interest in Microsoft requiring OEMs to limit the size and shape of additional folders and icons on the desktop. 39. Potentially, the balkanization argument is more significant in the case of OEMs inserting their own screens highlighting a competing browser or an ISP service that promotes a competing browser prior to the desktop screen because more significant changes in the appearance and startup sequence of Windows could result. Therefore, although these startup sequence restrictions probably have some anticompetitive effects, legitimate business justifications may be important. On balance, I consider the anticompetitive effects of these restrictions to outweigh concerns about balkanization of the startup sequence. I base this judgment on the fact that Microsoft has negotiated exceptions to these startup sequence restrictions with OEMs. 42
40. Collectively, these browser restrictions are remarkably uniform in their focus on promoting IE and discouraging the use of competing browsers. One possible justification for these restrictions could be that Microsoft and its Internet partners are engaged in joint marketing of their respective products. Microsoft offers promotional facilities on the desktop, and the IAPs and ICPs promote Microsoft on their web pages and Microsoft technologies with their services. The anticompetitive browser restrictions, it might be argued, perform the necessary function of limiting ex-post opportunism43 without which the joint marketing might not take place. For example, an Internet partner, such as an ICP, could get top billing on the Microsoft desktop but would then look for a similar deal on the channel bar of a competing browser in return for allowing that competitor to advertise on the ICP's web pages, too. From the standpoint of economic efficiency, though, anticompetitive restrictions such as these are justifiable only if there is no obvious alternative that is less anticompetitive. In this case, the restrictions are not reasonably necessary for the joint marketing to occur. The reason is that it is in Microsoft's interest to continue this joint marketing whether or not the Internet partner also deals with a competing browser vendor in the same way. 41. To see this, suppose that the Internet partner does joint marketing with a competitor in the browser market. What is Microsoft's best response? If it does not allow the Internet partner access to the Windows desktop, Microsoft leaves the competing browser as the only means by which an end user can be a customer of that IAP or ICP, If the IAP or ICP is of any importance, Microsoft would still want it on the Windows desktop in order to boost the use of IE. Similarly, because Microsoft has an OS monopoly, it is to the advantage of the IAP or ICP to allow Microsoft to advertise on its Web pages and install links to Microsoft's own Web pages, etc. Thus, I would expect that the joint marketing activities between Microsoft and its Internet partners would continue in the absence of the browser restrictions. Indeed, Microsoft has voluntarily dropped some of these restrictions, which appears inconsistent with a claim that they are necessary for joint marketing. This being so, the restrictions cannot be justified as a means of deterring ex-post opportunism by Microsoft's Internet partners. 42. A second justification sometimes given in the economics literature for certain types of exclusionary restrictions is that they can facilitate usage metering and thereby make possible efficient price discrimination. In this case, there seems to be no such metering function that is related to pricing and is facilitated by excluding competing browsers. A third type of justification in the literature is that exclusionary restrictions may be needed to maintain quality control. For this justification to apply, the use of a rival browser with Windows would somehow degrade the operation of Windows. To my knowledge, Microsoft often points out how well rival browsers such as Netscape Navigator works with Windows 95.44 (As pointed out above, if the technical integration of the browser with the OS in Windows 98 is required in order to provide seamless and efficient operation, then exclusionary restrictions would be unnecessary.) 43. The discussion in this section and in the preceding section demonstrates three main points. First, the exclusionary restrictions heavily bias end users' choices towards IE and away from competing browsers. Second, absent these restrictions, many end users would have been offered a competing browser by their OEM, TAP, or ICP. Third, these restrictions have little apparent business purpose besides being anticompetitive regarding a competing browser's cross platform role. These points by themselves would lead me to conclude that Microsoft's browser restrictions are intended to reduce competition to its OS monopoly. There is, however, an additional line of reasoning which leads to the same conclusion. This is the subject of the next section.
44. At this point, I turn to the question of whether the restrictions just described are aimed solely at expanding Microsoft's profit in Internet-related markets by increasing IE browser usage, or whether they are undertaken to impede competition in the OS market. To analyze the foil economic effects of these restrictions, I rely on the main finding of Section IT, that Microsoft has monopoly power over a bottleneck input in computing - the operating system. In this setting, browsers are complementary to the bottleneck input, and the restrictions described above tend to exclude competing browsers from being used with Microsoft's dominant OS. This paradigm, a monopoly with alleged exclusionary practices in a complementary market that it also services, has been analyzed by a number of economists and antitrust scholars.45 The general conclusion from this body of work is that if the price level in the complement's market is limited by competitive forces, then in the absence of efficiency justifications such as those discussed above, the monopolist's control over the bottleneck input does not give it any profit incentive to restrict or exclude a competitor's product in the complements market. The reason is that control over the bottleneck input allows the monopolist to extract value from consumers no matter whose version of the complementary good the consumer buys. 45. Applied to the case at hand, the bottleneck input is Microsoft's OS and the complementary product is the browser. Profits earned from browsers come not from the browser software itself, which is generally offered free to end users at the current time, but from Internet-related businesses engaged in by Microsoft, such as referral fees from IAPs and advertising revenue from the Microsoft website. Potentially, revenue could also come from the sale of desktop space to IAPs. 46. The analysis of Microsoft's restrictions involves several steps. Assume, for the sake of argument, that Microsoft markets a "better" browser than any of its rivals. For example, suppose hypothetically (and contrary to my previous analysis) that the integration of IE with the OS could provide a more "seamless" experience than would the use of another browser In that case, the bundle of Windows and Internet Explorer would create more value for the end user than would Windows with a competing browser. Given its dominant position in the OS market, Microsoft could extract higher value in the form of profits, whether through a higher royalty for the OS software, through raising customer referral fees paid by ISPs to Microsoft, or by charging IAPs and ICPs for space on the Windows desktop (other methods are doubtless possible, as well). Under these assumptions, absent anticompetitive motives, if IE is superior to competing browsers, Microsoft would have no need to impose contractual restrictions to encourage the use of IE. 47. Now suppose that a competing browser is superior to IE, In this case, it might well seem that Microsoft would gain by imposing exclusionary restrictions against the users of competing browsers. However, the economic logic described above applies in this case, as well. If the combination of Windows with a competing browser did provide more value to the user than bundling Windows with IE, Microsoft's bottleneck OS monopoly would allow it again to extract this value through a higher OS royalty (or by one of the other means described above), even though the increase in consumer value came from another company's browser and not its own. 48. The evidence establishes, however, that Microsoft has put in place contracting restrictions that indeed had force and constrained licensees of various versions of Windows from offering Netscape Navigator in circumstances where they wished to do so.46 Making the reasonable assumption that OEMs, IAPs and ICPs who wanted to offer Navigator wanted to do so because it would provide value to their customers, Microsoft's restrictions prevented end users from obtaining this additional value. Microsoft could have extracted this additional value from consumers {or OEMs, IAPs and ICPs), but chose not to in order to obtain the restrictions, In economic terms, Microsoft has incurred an opportunity cost by foregoing the additional value it could have extracted from consumers (or OEMs, IAPs, and ICPs). An example of the foregone value it could have obtained is the provision of tree space on its Channel Bar to ICPs who were willing to pay a positive price for placement on the Channel Bar as evidenced by deals Netscape struck with a number of ICPs for placement on the Netcaster Channel Bar.47 49. Why would Microsoft incur the opportunity cost of these restrictions? In my view the economic rationale is that it is attempting to preserve its OS monopoly. This behavior is easy to understand once we take account of the competitive threat to Microsoft's operating system monopoly by the competitive browser platform. The threat of Netscape's browser to Microsoft's OS dominance has been known for some time and has been described by Microsoft's Steve Ballmer:
If I wanted to compete with Microsoft, I wouldn't say 'Oh yeah, let's try to write a better memory manager, and we'll have lighter-weight threads than they do.' In this day and age it's not sexy enough, and there's no user interface. The OS/2 experience proves full well the value of that. I would do what Netscape is doing. I'd say I'll build on top of [Windows] and I'll take their future away from them.48 More recently, Microsoft's James Allchin stated: "The goal, the stated goal of Netscape was to replace the user interface of Windows where you couldn't see it and to create a new set of APIs that developers would write to."49 When asked how Netscape could threaten Windows and the sales of Windows, Mr. Allchin stated: "You get developers to write the APIs, you cover op Windows, you've just got this layer running on top, and if the size, performance was acceptable, it becomes irrelevant, Windows becomes irrelevant. 50 50. Because the browser can serve as a software applications platform independent of the underlying OS, a new entrant in the OS market would not have to create an installed base of software applications complementary to its OS and comparable to Microsoft's in its size and use in order to succeed. Developers would no longer have to incur the extra cost of rewriting software to be "ported" to the new OS. Instead, these applications written to the browser platform (perhaps using Sun Microsystems' Java programming technology51) would be accessible to a user using any OS that supported that browser. This would reduce both the installed base entry barrier and the switching costs entry barrier in the OS market. Therefore, I conclude that Microsoft's browser restrictions are likely to have the effect of weakening a potential competitive threat to its dominant position in the OS market 51. Moreover, Microsoft could obtain an additional benefit if these restrictions force major browser vendors to leave the market. The browser market could tip to monopoly, and entry barriers are likely to arise naturally, much as they do in the OS market. Thus, the browser market and associated Internet markets will become Jess competitive. With IE dominating the browser market, websites will tend to be written to the IE technology. This will induce more end users to switch to IE, increasing software developers' incentives to build websites around IE. As this occurs and as end users incur the sunken adjustment costs to IE, the natural entry barrier of installed base (of websites and other technology) and switching costs will act to give Microsoft market power for Internet services. Thus, in the long term, Microsoft will have the opportunity to benefit from a less competitive market structure in the Internet just as it has in the OS market. 52. An important part of the exposition in this section is the conclusion reached in paragraphs 45 through 47, that if these restrictions were aimed solely at expanding Microsoft's profits in Internet-related markets by increasing IE browser usage, it could do better by capturing such profits through the price of its OS, or through selling Internet products tied to the OS (such as desktop space and ISP referral fees). The implication is that these restrictions serve an anticompetitive purpose. However, as stated in paragraph 42, my conclusion that these restrictions are anticompetitive does not depend solely on the framework presented in those paragraphs. There are three additional reasons why I regard Microsoft's browser restrictions to be anticompetitive. First, by restricting the use of competing browsers, Microsoft automatically blunts the threat to its OS monopoly that is posed by the browser's potential as an alternative platform for software applications. Second, the internal Microsoft sources cited above make it clear that OEM, IAP, and ISP restrictions described earlier were imposed with this threat very much in mind. Third, the discussion in Section B above demonstrates that the restrictions listed in Table Two have little apparent purpose aside from their anticompetitive effects,
53. The discussion above demonstrates that the intent and effect of the contractual restriction is predominately anticompetitive. An additional Microsoft practice of considerable importance in light of these anticompetitive effects is that of distributing its browser without charge. This practice has forced Netscape to reduce its browser price to zero and has apparently cut into Netscape's revenue stream considerably. I will now analyze the effects of Microsoft's distributing its browser without charge, taken in the context of the anticompetitive non-price restrictions described above. 54. The starting point for discussing pricing is cost. For a software product like a browser, the vast majority of costs are first-copy costs, i.e., the cost of developing the product. Given that the product has been developed, additional costs related to volume of output are low, consisting mainly of packaging, marketing, and distribution costs. Therefore, a zero browser price certainly does not recover any of the first-copy costs, although the extra, or "marginal," cost of an additional browser may well be close to zero. Hence, Microsoft has priced the browser in such a way that it cannot recover the cost of the browser from the browser market itself 55. Furthermore, there are significant interactions between zero pricing of the browser and the non-price restrictions analyzed above. For example, by bundling IE with Windows 98, Microsoft ensures that many end users will decide just to go with IE due to convenience, since there is a nontrivial amount of effort involved to download and install another browser after the end user's initial boot of the PC, Because Microsoft does not offer a cheaper version of the OS without IE installed, the zero price of IE simply reinforces the convenience of buying a PC with IE preinstalled. Therefore, zero pricing reinforces the anticompetitive effects of the OEM license requirement that IE cannot be uninstalled. The agreements with certain on-line service providers, ISPs, and ICPs all have elements which make it onerous to use a non-Microsoft browser when attempting to access websites based on IE technology or which do not support non-Microsoft browsers. Zero pricing adds a strong additional inducement for such distributors to go with IE. 56. Microsoft's decision to integrate the IE browser with the Windows 98 operating system suggests the browser is not just free now, but "forever" free, as long as Windows 98 is used. In effect, Microsoft has made a binding, credible commitment to a free browser. That being true, for a rival to enter or remain in the browser market, that rival must be able to support a free browser with revenue from other sources, Even if Microsoft chooses not to maintain a forever free policy by bundling IE and Windows 98, consumers may form the expectation that such a policy will continue, causing the market to "tip" in its favor more rapidly. 57. It is possible as a matter of theory that (possibly temporarily) free distribution of browsers could be justified without anticompetitive concern, as a penetration policy designed to expand Internet markets. However, given the forever free commitment on the browser price, if the browser market "tips" easily, then if Microsoft drives rivals from the browser market, entry barriers will arise, some as a result of network effects, and some due to Microsoft's contractual restrictions concerning the use of non-Microsoft browsers. At that point, Microsoft can raise the price of its OS, the bottleneck input, so as to exploit this extension of its monopoly power.52 Alternatively, it could increase referral fees from IAPs, or begin to charge for space on the desktop. In the end, as others have observed, "there is no free browser,"53 IV. Microsoft Will Recoup the Opportunity
Costs of its 58. In the preceding discussion, I have established two points. First, Microsoft has monopoly power in the market for OS that is protected to a considerable degree by natural entry barriers (such as switching costs and increasing returns to scale in consumption) and by the artificial entry barriers created through its contractual policies regarding competing browsers. Second, Microsoft has adopted contractual restrictions with OEMs, ISPs, and ICPs which have a positive opportunity cost, but seem likely to remove competing web browsers from their potential role as a platform for software applications which would run on computers independently of the underlying operating system. Such a platform would "commoditize" Microsoft's Windows operating system, as applications no longer would be specific to a given OS. 59. Because the likely effect of Microsoft's contractual restrictions will be to remove the competitive threat to its operating system, Microsoft will likely recoup whatever profits it has foregone from ICPs and IAPs due to its browser restrictions. These restrictions are profitable overall because they will help to ensure Microsoft's immensely profitable OS business is without any serious competition for some time to come. Therefore, I find it likely that Microsoft would recoup the lost profits associated with its restriction through higher profits earned on its OS than if those restrictions were not imposed. 60. In reaching this conclusion, I do not need to rely solely on my own analysis. Microsoft's senior executives have made It clear in documents cited above that they consider the browser platform threat to be serious and that their contractual restrictions regarding competing browsers are intended to tie any user benefits from these technologies back to the use of Windows. The company's actions have been entirely consistent with this concern. Clearly, in Microsoft's own calculations, the benefits from protecting its OS monopoly outweigh the opportunity costs of the restrictions that I have analyzed. 61. I have concluded that the Microsoft contractual restrictions that I have analyzed are anticompetitive for four main reasons:
62. What would be the consequence to consumers if Microsoft's bundling and restrictive practices enabled it to maintain its OS monopoly and create a monopoly in the market for browsers? Consumers would be harmed in two significant ways. First, virtually all consumer and business users would pay prices for personal computers higher than those that would exist in a competitive market. Given the increased reliance on personal computers both at home and in the work place, even a small increase in price would create substantial economic welfare losses. Second, and perhaps more importantly, innovation in areas that threaten Microsoft would be inhibited. This could potentially have significant adverse consequences for the pace and direction of innovative activity in the software industry. 63. How is it that Microsoft's anticompetitive behavior, aimed at maintaining its dominance, could adversely affect the type and extent of innovation in the computer software industry? As a general proposition, firms invest in research and development only if they expect to earn, at a minimum, a competitive rate of return, that is, a return equal! to that obtained on an investment of comparable risk. To the extent that Microsoft's anticompetitive practices make it that much more difficult for an innovator to prove successful in markets where Microsoft currently dominates, independent innovative activity that might otherwise have been profitable may not be. At the extreme, if Microsoft's practices were to erect an insurmountable barrier to successful entry in its core monopoly markets, independent third party innovation in these markets would be less likely, and the prospects for technical advance in these markets would rely solely on the efforts of Microsoft itself. While it does not seem likely that Microsoft's practices will have quite so extreme an effect, it remains the case that innovative activity is likely to be skewed inefficiently, as firms seek to "invent around'' the Microsoft bottleneck, rather than compete directly against it. 64. Beyond any adverse effect on innovative effort in the PC operating system market itself, Microsoft's practices threaten also to deter efficient innovative activity aimed at developing technologies complementary to its operating system (such as Internet browsers). As discussed earlier in this declaration, and as recognized in much of the economics literature, even firms with monopoly power generally have an incentive to encourage the development and success of products complementary to their own. All else equal, this tends to enhance the value of the dominant firm's product. However, where the development of such products and technologies has the potential eventually to threaten the dominant firm's market power, for example, by creating a computing environment in which application software would work regardless of which operating system was installed on the user's computer, the deterring of efficient innovation in this area could be in the self-interest of Microsoft. 65. In sum, Microsoft's actions are likely not only to affect the prices consumers pay for products over which Microsoft has monopoly power, but also to affect adversely the path of innovation in the computer software industry. The artificial entry barriers erected by Microsoft (through bundling and its other restrictive practices) will have the effect of impeding the natural tendency of the marketplace to provide improved and alternative technologies that compete with Microsoft's OS monopoly, reducing the welfare of society as a whole. I declare under penalty of perjury that the foregoing is true and accurate. Executed on May 15 , 1998 in Austin, Texas.
IN THE UNITED STATES DISTRICT COURT
APPENDIX TO DECLARATION OF David S. Sibley Professor, Department of Economics Education:
1969 B. A- in Economics, Stanford University Teaching Fields:
Industrial Organization, Economics of Information Research Fields:
Economics of asymmetric information, telecommunications policy* public utility pricing, models of firm's internal organization. Professional Experience:
March, 1992 - Present: John Michael Stuart Centennial Professor of Economics, University of Texas at Austin. Teaching:
September 1991 - Present: Introductory Microeconomics, undergraduate and graduate Industrial Organization. Publications:
A. Journal Articles: Completed Papers:
''Incumbency and Bidding Behavior m the Dallas-Ft. Worth Milk Market," (with C. Hewitt and J. T. McClave), submitted to Southern Economic Journal. Current Research Areas:
Editorial Duties:
Associate Editor of the Journal of Regulatory Economics. Other Professional Activities:
References:
May 1998
UNITED STATES CISTRICT COURT
TRANSCRIPT OF PROCEEDINGS
How to Uninstall Internet Explorer 4.0 Last reviewed: October 17,1997
SUMMARY This article describes how to uninstall (or remove) Internet Explorer 4.0 (version 4.71.1712.6). For information about uninstalling Internet Explorer Platform Preview 2.0 (version 4.71.1008). please see the following article in the Microsoft Knowledge Base:
ARTICLE-ID: 0170757 When to Remove Interner Explorer Remove Internet Explorer 4.0 before you perform any of the following actions:
There are some configurations in which unistalling Internet Explorer 4.0 does not work. Uninstalling Internet Explorer 4.0 is not supported in the following situations:
- You have removed the Internet Explorer 4.0 backup files. If you removed the backup files, you are unable to uninstall Internet Explorer 4.0 without reinstalling Windows into a different folder. For additional information, please see the following article in the Microsoft Knowledge Base: ARTICLE-ID: 0172919 MORE INFORMATION Before you uninstall Internet Explorer, quit all running programs. For information about problems that can occur if you do not quit all running programs before uninstalling Internet Explorer, please see the following article in the Microsoft Knowledge Base:
ARTICLE-ID: 0265646 To uninstall Internet Explorer 4.0 in Windows 95 and Windows NT 4.0, follow these steps: NOTE: You must have administrative privileges to install Internet Explorer 4.0 on Windows NT. This inclueds having administrative privileges the first time you start your computer after you install or uninstall Internet Explorer 4.0. 1. In Control Panel, double click Add/Remove Programs and then click the Install/Uninstall tab. 2. In the list of installed programs, click Microsoft Internet Explorer 4.0, and then click Add/Remove Programs. 3. In the Internet Explorer 4.0 Active Setup dialog box, click "Uninstall Internet Explorer 4.0 and all its components." and then click Advanced to specify any additional components you want to remove. 4. Click OK. 5. When Setup starts in Maintenance mode, click Remove All.
If there are any files in the Recycle Bin, you recieve the following message: Known Issues with Uninstalling
General Troubleshooting If Maintenance-mode Setup does not work or you cannot run Setup, try one of the following methods
ARTICLE-ID: 0166313
Additional query words: 4,000 THE INFORMATION PROVIDIDED IN THE MICROSOFT KNOWLDGE BASE IS PROVIDED "AS IS" WITHOUT WARRANTY OF ANY KIND. MICROSOFT DISCLAIMS ALL WARRANTEES EITHER EXPRESS OR IMPLIED, INCLUDING THE WARRANTIES OF THE MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE IN NO EVENT SHALL MICROSOFT CORPORATION OR ITS SUPPLIERS BE LIABLE FOR ANY DAMAGES WHATSOEVER INCLUDING DIRECT, INDIRECT, INCIDENTAL, CONCEQUENTIAL LOSS OF BUSINESS PROFITS OR SPECIAL DAMAGES EVEN IF MICROSOFT CORPORATION OR ITS SUPPLIERS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. SOME STATES DO NOT ALLOW THE EXCLUSION OR LIMITATION OF LIABILITY FOR CONSEQUENTIAL OR INCEIDENTAL DAMAGES SO THE FOREGOING LIMITATION MAY NOT APPLY. Last reviewed: October 17, 1997
Removing Internet Explorer 4.0 for Windows 95 Ising Ieremove.exe Last reviewed: October 17, 1997l Article ID: Q166313
SUMMARY This article describes how to remove Internet Explorer 4.0 for Windows 95 using the Iremove.exe tool and a resistry that was created before you installed Internet Explorer 4.0 (for example, System 1st) Follow the steps described in this artivle in the following situations:
Do not follow the steps described in this article in the following situations:
ARTICLE-ID: 0174265 MORE INFORMATION To remove Internet Explorer 4.0 by using hte Ieremove.exe tool and a registry that was created before you installed Internet Explorer 4.0 follow these steps: 1. If you have not already done so, restore a registry that was created before you installed Internet Explorer 4.0. NOTE: Internet Explorer 4.0 replaces the original System.1st file with a copy of the registry created during the Internet Explorer 4.0 installation. The original System.1st file is saved in the Integrated Browser.dat file (Integr-1.dat) during the Internet Explorer 4.0 installation. To extract the original System.1st file from the Integr-1.dat file,
type the following commands at a compand prompt. ed<path>\uninst-1 where <path> is the path to the folder in which Internet Explorer is installed, and <dir> is the folder in which the extracted file will be saved. For additional information about using the System.1st file to restore the registry to the state when you first started Windows 95 successfully, see the following article in the Microsoft Knowledge Base:
2. Restart the computer. When you see the "Starting Windows 95" message, press the F8 key, and then choose Safe Mode Command Prompt Only from the Startup menu. 3. Type the following commands, pressing ENTER after each command
cd <path>\uninst-1
where <path> is the path to the folder in which Internet Explorer is installed. 4. Restart the computer. 5. Click Start, click Run, type the following command in the Open box (with quotation marks), and then click OK.
"<path1>\ieremove.exe" /i:"<path2>\integrated browser.dat" where <path1> is the full path to the Ieremove.exe file, <path2> is the full path to the Integrated Browser.dat file, and <path3> is the full path to the ie4bak.dat file. For example, if the Ieremove.exe file is located in the C:\Windows\System folder, the Integrated Browser.dat file is located in the C:\Program Files\Internet Explorer Uninstall Information folder, and the Ie4bak.dat file is located in the C:\Windows folder, type the following command (with quotation marks)
"c:\windows\system\ieremove.exe" /i:"c:\progeam files\internet 6. Click OK, click Yes, and then restart your computer after the Ieremove.exe tool is finished running 7. Reinstall Internet Explorer 4.0, and then unistall it. For information about how to unistall Internet Explorer 4.0 please see the following article in the Microsoft Knowledge Base:
ARTICLE-ID: 0174265 TITLE: How to uninstall Internet Explorer 4.0 This step is necessary to ensure that the additional components included with Internet Explorer 4.0 are removed. NOTE: Ieremove.exe is a command line executable designed to remove Internet Explorer and it's components. Ieremove.exe may not remove Internet Explorer without using line options. For information about the command line options available for Ieremove.exe, type "IEREMOVE/?" (without quotations).
Additional query words: 4.00 THE INFORMATION PROVIDED IN THE MICROSOFT KNOWLDGE BASE IS PROVIDED "AS IS" WITHOUT WARRANTY OF ANY KIND. MICROSOFT DISCLAIMS ALL WARRANTEES EITHER EXPRESS OR IMPLIED, INCLUDING THE WARRANTIES OF THE MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE IN NO EVENT SHALL MICROSOFT CORPORATION OR ITS SUPPLIERS BE LIABLE FOR ANY DAMAGES WHATSOEVER INCLUDING DIRECT, INDIRECT, INCIDENTAL, CONCEQUENTIAL LOSS OF BUSINESS PROFITS OR SPECIAL DAMAGES EVEN IF MICROSOFT CORPORATION OR ITS SUPPLIERS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. SOME STATES DO NOT ALLOW THE EXCLUSION OR LIMITATION OF LIABILITY FOR CONSEQUENTIAL OR INCEIDENTAL DAMAGES SO THE FOREGOING LIMITATION MAY NOT APPLY. Last reviewed: October 17, 1997
How to Manually Unistall Internet Exporer 4.0 Last reviewed: October 24, 1997
SUMMARY This article describes how to maually uninstall Internet Explorer 4.0 for Windows 95 when you are unable to uninstall Internet Explorer with the Add/Remove Programs tool in Control Panel or the Ieremove.exe tool. NOTE: Do not follow the steps in this article if you have not tried to remove Internet Explorer 4.0 by using the steps described in the following Microsoft Knowledge Base articles:
ARTICLE-ID: 0174265 MORE INFORMATION To uninstall Internet Explorer 4.0 after trying the steps in the articles listed above, use the following steps. NOTE: If the Windows Desktop Update is not installed on your computer, or if Windows 95 does not start, skip to step 4. 1. Click Start, point to Settings, click Control Panel, and then click Add/Remove Programs. 2. On the Install/Uninstall tab, click Microsoft Internet Explorer 4.0, and then click Add/Remove. 3. Click Remove the Windows Desktop Update component, but keep the Internet Explorer 4.0 Web browser, click OK, and then click OK. NOTE: Even though you are unable to remove Internet Explorer 4.0, the option to uninstall the Windows Desktop Update should be available in the Add/Remove programs tool. If you cannot uninstall the Windows Desktop Update with the Add/Remove programs tool, you must extract Explorer.exe and Shell32.dll from the original Windows 95 disks or CD-ROM. See step 5. 4. NOTE: If you installed Windows 95 from CD-ROM and you are using real-mode CD-ROM drivers (therefore you can access your CD-ROM drive in MS-DOS), install real-mode CD-ROM drivers. For more information anout installing real mode CD-ROM drivers, consult the manufacturer's documentation included with your CD-ROM, or contact the manufacturer of the CD-ROM and do not continue with the remaining steps. Restart your computer. When you see the "Starting Windows 95" message, press F8, and then choose Command Prompt Only. 5.Type the following commands, pressing ENTER after each command:
cd windows\system NOTE: This procedure assumes you do not have files with these names with the .old extension. If you do have files with these names with the .old extension, use a file name extension that is not currently in use. If you cannot remove the Windows Desktop Update from the Add/Remove Programs tool, type the following commands, pressing ENTER after each command:
cd \<windows> From the original Windows 95 disks or CD-ROM, extract the Explorer.exe file to the Windows folder and then extract the Shell32.dll file to the Windows\System folder. To do so, use the appropriate method: Windows 95 CD-ROM Insert the Windows 95 CD-ROM into the CD-ROM drive, and then type the following commands, pressing ENTER after each command:
extract /a <drive>:\win95\win95_02.cab explorer.exe /l windows where <drive> is the drive letter assigned to your CD-ROM drive. Windows 95 Floppy Disks Insert the Windows 95 floppy disk 1 into the floppy disk drive, and then type the following commands, pressing ENTER after each command
extract /a <drive>:\win95_02.cab explorer.exe /l windows where <drive> is the floppy disk drive letter.
For more information about using the Extract tool, type "extract" (without quotation marks) at a command prompt, or see the following article in the Microsoft Knowledge Base: 6. Reinstall Windows 95. To do so, insert floppy disk 1 (or the Windows 95 CD-ROM) into the appropriate drive, and then type the following commands, pressing ENTER after each command
<drive>: where <drive> is either the drive letter assigned to your CD-ROM, or the floppy disk drive letter. Click No when you receive the following prompt:
A file being copied is older than the file currently on your computer. It is recommended that you keep your existing file. 7. After Setup finishes, reinstall Internet Explorer 3.02. For more information about installing Internet Explorer 3.02, please see the following article in the Microsoft Knowledge Base:
ARTICLE: 0164475 You can now reinstall Internet Explorer 4.0. To completely remove Internet Explorer 4 from your computer, reinstall Internet Explorer 4, and then uninstall Internet Explorer 4.0 For information about reinstalling Internet Explorer 4, please see the following article in the Microsoft Knowledge Base: ARTICLE: 170993
Additional query words: 4.00 3.00 THE INFORMATION PROVIDED IN THE MICROSOFT KNOWLDGE BASE IS PROVIDED "AS IS" WITHOUT WARRANTY OF ANY KIND. MICROSOFT DISCLAIMS ALL WARRANTEES EITHER EXPRESS OR IMPLIED, INCLUDING THE WARRANTIES OF THE MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE IN NO EVENT SHALL MICROSOFT CORPORATION OR ITS SUPPLIERS BE LIABLE FOR ANY DAMAGES WHATSOEVER INCLUDING DIRECT, INDIRECT, INCIDENTAL, CONCEQUENTIAL LOSS OF BUSINESS PROFITS OR SPECIAL DAMAGES EVEN IF MICROSOFT CORPORATION OR ITS SUPPLIERS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. SOME STATES DO NOT ALLOW THE EXCLUSION OR LIMITATION OF LIABILITY FOR CONSEQUENTIAL OR INCEIDENTAL DAMAGES SO THE FOREGOING LIMITATION MAY NOT APPLY. Last reviewed: October 24, 1997
Solving problems Installing or Uninstalling Internet Explorer 4.0 Can you uninstall Internet Explorer 4.0 using the Add/Remove Programs Control Panel? Note
To uninstall Internet Explorer 4.0 in Windows 95 and Windows NT 4.0
If you have any files in the Recycle Bin, you receive the following warning:
Internet Explorer 4.0 has modified the format of the
You can now Install a later build of Internet Explorer 4.0. To restore a working installation of Internet Explorer version 3.02, you must remove extra files associated with Internet Explorer 4.0. To remove remaining files associated with Internet Explorer 4.0
These files are associated with the Internet Connection Wizard and are not removed during Internet Explorer uninstallation.
For information about the availability of Internet Explorer 3.02, see the following article in the Microsoft Knowledge Base:
This table tracks your status in the troubleshooting process. If you need to change your respond questionr you can do so below; Problem: Cannot Uninstall Supported Configuration: Yes Unknown
Press Pass Feature Story April 29,1998 OVERVIEW OF SECTION 2 OF THE SHERMAN ACT AND ITS APPLICATION TO MICROSOFT By Charles F. (Rick) Rule A great deal of the public discussion concerning Microsoft seems to assume that because Microsoft has been highly profitable and has engaged in various practices that have placed a number of rivals under intense competetive pressure, the company is fair game for whatever "remedies" the Department of Juslice might choose to impase. In fact, however, the Department's power to impose remedies on Microsoft is dependent on its abity to establish in court that Microsoft has violated section 2 of the Sherman Act. Specifically, the Department must prove not only that Microsoft has monopoly power but also that Microsoft has acquired or maintained that power through "exclusionary" or "predatory" acts. In light of those legal requirements, there simply is no sound basis for a section 2 suit against Microsoft. The various theories that have been advanced by Microsoft's detractors as grounds for a section 2 suit would require a radical departure from existing case law. In effect, the law's current focus on consumers and innovation would have to be diverted to protection of competitors at the expense of consumers. Moreover, those theories would require courts to second-guess Microsoft's product design and distribution efforts - a task that the courts are simply not equipped to perform. And, even if the Department could persuade the courts to transform the antitrust laws so radically, any remedy that the Department might seek to impose would inevitably be highly regulatory and would almost certainly reduce consumer welfare and impede innovation. I. Section 2 of the Sherman Act and "Monopolization" As the Supreme Court has stated, "Congress designed the Sherman Act as a 'consumer welfare prescription.'" Relter v. Sonotome Corp, 442 U.S. 330.343 (1979), quoting R. Bork, The Antitrust Paradox 66 (1978). In other words, the law protects the marketplace from private conduct that interferes with the competitive process. Or stated differently, the antitrust laws protect "competition, not competitors" Brown Shoe Co, v. Untied States, 370 U.S. 294, 320 (1962). Antitrust law as a general matter does not attempt to regulate the unilateral activities of a person or firm, in addition, the law does not condemn commercial success achieved through competion on the merits, even when it results in the competitive triumph of a single firm. Rather, the law - section 2 of the Sherman Act specifically - only condemns "monopolization* (or attempts to "monopolize"). 15 U.S.C, § 2. As intepreted by the courts, "monopolization" has two elements: (A) the possession of monopoly power in a relevant market and (B) 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," United States v. Grinnell Corp., 384 U.S. 563. 570-71 (1966). A. Monopoly Power (1) The Law - A necessary, though not sufficient, prerequisite for a violation of section 2 of the Sherman Act is a showing that the defendant hag "monopoly power." The Supreme Court has defined such power as "the power to control market prices or exclude competition." United States v. E.I. du Pont de Nemours & Co.. 351 U.S, 377,391 {1956). While a very large market share may raise an inference of monopoly power, "[w]hen there are better ways to estimate market power, the court should use them." Balt Memorial Hosp. v. Mutual Hosp. Ins., 784 F.2d 1325, 1336 (7th Cir. 1986). For example, a large market share is only indicative of monopoly power if it is durable and persistent. If there are low barriers to entry or to expansion by fringe firms or if the market is highly dynamic, then a firm's current market position is likely to be fragile and any hope of exercising market power will be futile. See, e.g., Los Angeles Land Co. v. Brunswick Corp., 6 F.3d 1422 (9th Cir. 1993) (holding that the defendant did not possess monopoly power despite the fact that it had a 100 percent share of the relevant market). (2) Application to Microsoft - Microsoft's critics claim that, because Microsoft's operating systems are installed on a large percentage of the PCs being shipped today, Microsoft has monopoly power. That claim, however, is extremely misleading for several reasons. First, a "market" limited to operating systems installed by computer manufacturers on Intel-compatible PCs is undule circumscribed. Not only can a computer manufacturer choose amoung various PC operating systems, including IBM's OS/2, Solaris x86 from Sun Microsystems, UnixWare from Santa Cruz Operation, and Linux (the "open" operating system based on Unix), but consumers and businesses can choose systems other than a PC, such as an Apple Macintosh, a so-called Net-Computer ("NC"), or even bigger systems such as mini-computers. In addition, any market definition that ignores the fact that Micrsoft must compete against its own installed base fails to take account of an important competitive dynamic. Computer software does not wear out. A current computer owner will switch to a new operating system (whether Microsoft's or anyone else's) only if he or she perceives the incremental value of the new operating system to be worth its price. And, one cannot ignore so-called "middleware" layers, such as Netscape's browsign software or Lotus Notes, that provide independent software vendors or developers ("ISV's") with platforms complete with thier own application programming interfaces ("API's") that are being touted as substitutes for Windoes. Indeed, powerful competitors such as IBM and Sun Microstystems are investing hundreds of millions of dollars to develop and market new technologies that they hope will replace Windows. Second, the "market share" of a software product is much less significant than the typical market share possessed by a manufacturer of a producer of physical goods. Once written, a piece of software can be copied an infinite number of times at virtually zero marginal cost. In other words, the productive "capacity" of every piece of software, once written, is virtually infinite, even if its current sales are de minimis. If the owner of the "dominant" software tries to reduce output, the fringe firms can costlessly increase the output of thier software, completely replace the output withheld by the dominant software publisher and thus thwart any effort to raise price. Therefore, any effort by Microsoft to raise price (without providing commensurate value) would be doomed to failure because consumers are able to switch to competing operating system they currently own. The price Microsoft charges and the market share it achieves are nothing more than a reflection of the inherent value (or superior efficiency) of Microsoft's operating systems as compared to the available alternatives. Third, because of the rapid change in technology in the computer market, there are always new opporturities for start-up firms or established players (such as IBM, Oracle, or Sun} to introduce new operating systems or to enhance existing ones. And, if those new or improved operating systems indeed represent the proverbial "better mousetrap" then they have a real chance to supplant the existing market leader. These opportunities we driven by charges, Inter alla, in software, in hardware, in communications technology, and in consumer preferences. Microsoft has only succeeded by constantly bringing new technologies to consumers more efficiently, more attractively, and at lower prices than the competition. Some of Microsoft's critics claim that becamse of so-called "network externalities," it is virtually impossible for anyone to challenge Microsoft's "dominance" on the PC desktop. Certainly, Windows is very popular, in part, because it provides for literally thousoands of ISVs a valuabvle platform that simplifies the task of creating compatible applications and because millions of consumers have become familiar with the "look and feel" of Windows and have assembled libraries of Windows applications. Nevertheless, the notion that these "externalities" insulate Microsoft from competition is simply wrong. The same "lock-in" story was being told just a few short years ago about Nintendo's "monopoly" position in video games. Yet, in the interveneing years, Sega and now Sony have overtaken Nintendo. The same thing will happen to Microsoft if it down not continue to innovate at a rapid pace and continue to offer its operating system at competitive prices. In short, notwithstanding the percentage of PCs currently being shipped with Windows, Microsoft has no power to suppress the changes that almost daily create opportunities for those hoping to supplant Microsoft. The company's only hope of sustaining its success is to continue to do the best, most efficient job of making each technological advance available to the consumer. That technological treadmill is hardly the quiet life and "deaden[ing] of initiative" experienced by a true monopolist. United States v. Aluminum Co. of America, 148 F.2d 416, 427 (24 Cir. 1945). B. "Willful" Acquisition or Maintenance (1) The Law - Even if Microsoft did have (or were ever found by a court to have) monopoly power, possession of such power alone would not constitute a violation of sscfion 2 of the Sherman Act. The acquisition, maintenance or extension of monopoly power few virtue of foresight, skill, industry, or even luck is not unlawfal. See, e.g., United States v. Grinnel Corp., 384 U.S. at 570-71. Section 2 of the Sherman Act does not condemn a firm with lawfully abtained monopoly power from competing vigorously on the merits; rather, the law prohibits such firm from engaging in "exclusionary" or "predatory" acts that lead to monopoly or that maintain an existing monopoly. According to the Supreme Court, "[i]f a firm has been 'attemping to exclude rivals on some basis other than efficiancy,' it is fair to characterize its behavior as predatory." Aspen Skiing Co. v. Aspen Highlands Skiing Corp.,472 U.S. 585, 605 (1985), quoting R. Bork, The Antitrust Paradox 138 (1978). For example, in United States v. Lorain Journal Co., 342 U.S. 143 (1951), the Court found that the only local newspaper in Lorain, Ohio had monopoly power over the sale of local advertising and had exgaged in illegal exclusionary conduct. That conduct was blatant - the Journal flatly refused to accept any advertisements from companies also advertising on a new radio station that broadcast into Lorain and providied the only competing medium for local advertising. According to the Supreme Court, advertising in the Journal was "essential" for local advertisers, and the evidence was clear that the Journal was engaged in the conduct to drive its only competitor out of business. The Journal moreover, offered no efficiency rationale or other legitimate business justification for its conduct. On the other hand, conduct for which there is a legitimate business justifcation (i.e., conduct designed to lower the cost of production, distibution or marketing or to increase demand for the product or service) is not exclusionary or predatory even if the conduct disadvantages cometitors. See, e.g., Multistate Legal Studios, Inc. v. Harcourt Brace Jovanovich Legal and Prof'l Publics., 63 F.3d 1540,1550 (10th Cir. 1995) ("A defendant may avoid liability by showing a legitimate business justification for the conduct"). In addition, so long as the justification is "asserted in good faith and not as a pretext, the defense does not require any kind of 'balancing' of social gains against competitive harms, except in the gross sense that trivial justifications should be disregarded." 3 P, Areeda & H. Hovenkamp, Antitrust Law ¶ 658f, at 127 (Rev. ed. 1996). Moreover, section 2 proscribes conduct not intent, and illegality ultimately turns on objective evidence of anticompetitive effect, not on subjective evidence of intent. See e.g., Ocean State Physicians Health Plan v. Blue Cross & Blue Shield, 883 F.2d 1101, 1113 (1st Cir. 1989) (a "desire to crush a competitor, standing alone, is insufficient to make out a violation of the antitrust laws"), cert. denied, 494 U.S. 1027 (1990). Statements by an alleged monopolist's officers- no matter how aggressive or colorful the rhetoric - cannot substitute for objective evidence that the alleged monopolist has actually engaged in exclusionary or predatory conduct. See e.g., 3 P. Areeda & H. Hovenkamp, Antitrust Law ¶ 651a, at 74 (Rev. ed, 1996). (2) Application to Microsoft - When evaluating all the allegations concerning exclusionary conduct by Microsoft, it is useful to recall the outcome of the first wave of investigations of Microsoft in the first half of this decade. Notwithstanding careful examination (by both the Federal Trade Commission and the Antitrust Division of the U.S. Department of Justice) of numerous allegations, the Department found only a single practice - the so-called per processor license - to be worthy of a challenge. And, even then, the Department stated on the record that the practice did not account for Microsoft's success in licensing operating systems to computer manufacturers but that the practice represented only a future threat to competition. According to the Department, Microsoft's position in the market was obtained by lawful means. Microsoft's rivals, of course, did not stop making allegations when the court entered the consent decree preventing Microsoft from continuing to offer the option of a per-preocessor license. Indeed, not long after it agreed to the consent decree, the Department declared itself to be the "Microsoft complaint department." Now, the critics are alleging that Microsoft has engaged in various exclusionary practices in order to maintain its "monopoly" or to extend that monopoly into new markets, primarily the Internet. Nevertheless, most of the recent allegations are similiar if not identical to those investigated and rejected by the Department during its original investigation. And, with respect to the current allegations that reflect actual (as opposed to confabulated) Microsoft practices, the conduct represents vigorous competition on the merits, not unlawful exclusion.
Quite simply, the principal reason Microsoft has Integrated Internet functionality into its operating systems is that ISVs have demanded it. For example, ISVs are increasingly writing their applications to display information, whether stored locally (e.g., on a PC's hard drive) or downloaded from the internet, in the common language of the internet - so called, hypertext mark-up language ("HTML"). Putting that funcationality in the operating system allows the thousands of ISVs to call on the same HTML "shared library" in the operating system and to avoid the need for each ISV to write its own version of HTML rendering in its applicatin. Having integrated HTML rendering and other internet fuctionality into the operating system in order to satisfy the demands of ISVs, it is logical (and efficient) for Microsoft to provide a means (for example, the Internet Explorer ("IE") icon on the desktop) to enable consumers to launch that functionality to "browse" the Internet. While the efficiencies of integrating internet functionality into Microsoft's operating systems seem clear, the exclusionary effect of that integration is difficult to discern - except to the extent that such integration is so technologically superior that it diverts demand from web browsing software such as Netscape's Navigator and Communicator. Indeed, some level of Internet functionality has been integrated into Windows 95 from the time it was first shipped to computer manufacturers; however, it was not until IE's functionality (embodied in versions 3.0 and 4.0) surpassed that of Netscape's products that consumers in large numbers started choosing IE instead of these products. Moreover, in as consent decree case against Microsoft, the Department has pointed to no evidence that Microsoft attempted to prevent any computer manufacturer from installing Netscape's web browsing software. Rather, Microsoft simply requires computer manufacturers to install Winders as shipped, thereby precluding them from deleting features, including Internet functionality and the IE icon. Ironically, Microsoft enforcement of this license provision prevents Netscape - whose web browsing software is still used by more than 50 percent of Internet surfers - from obtaining exclusives with computer manufacturers; yet, the critics suggest it is Microsoft's conduct that is exclusionary!
Of course, Microsoft also allows existing owners of Windows to download the latest version of IE off the Internet at no charge and gives away to the users of non-Microsoft operating systems, such as Apple's Macintosh, the web browsing software that Microsoft has written for those other operating systems. However, those "give aways" are also not predatory. The Supretme Court has indicated that an alleged monopolist's pricing is not "predatory" unless Its prices are "below an appropriate measure of [that firm's] costs."Brooke Group v, Brown & Williamson Tobacco Co, 509 U.S. 203, 222 (1993). Once a piece of software is written, the cost of copying it is esscentially zero. Thus, even giving away copies of software would not be "predatory," as the Supreme Court has defined the term. Also, it was Netscape, not Microsoft, that began the practice of giving away web browsing software. Moreover, like Microsoft, Netscape gives away its web browsing software in the expectation of earning revenues elsewhere - for example, by stimulating sales of related software and services.
As with the past allegations, there is simply no substance to the claim that Microsoft has manipulated APIs in Windows to harm Netscape or anyone else. First, based on both Microsoft's own tests of as operating systems (both Windows 95 and Windows 98) and the observations of unbiased third parties (such as journalists), Netscape's browser products and related software install and function well on Windows - there is no evidence that Netscape's products routinely crash or that their performance is degraded. Second, Microsoft treats Netscape no differently from any of the other thousands and thousands of ISVs that successfully and profitably write applications for Windows. Netscape has full access to Windows APIs, and Microsoft's large software-developer relations group has always been available to work with Netscape to ensure that Netscape's browsing software functions to its maximum capabilities on Windows. Third, as even Microsoft's severest critics note, a central reason for Microsoft's commercial success has been the support Microsoft provides to ISVs writing applications for Microsoft's operating systems. That strategy, which Microsoft has traditionally and consistently followed, is fundamentally at odds with the allegations being resurrected by Netscape. In short, there is no evidence that Microsoft has done anything to make Windows incompatible with Netscape's products. It would make no sense to do so given the popularily of those products among Windows users. Moreover, given the importance that Microsoft attaches to good relations with software developers, it is in Microsoft's interest to avoid even the appearance of treating one of those software developers unfairly by depriving it of information needed to develop applications that are compatible with Windows. Nevertheless (and somewhat ironically), it is not unlawful for even a monopolist to make improvements in its products in such a way as to give as own related products (such as applications in the case of an operating system) a competitive advantage over competing products sold by rivals. And the law imposes no obligation on a monopolist to reveal proprietary information (such as APIs) to rivals simply because they are competitively disadvantaged without that information. A fortroni, if a company such as Microsoft systematical discloses large amounts of information about its new operating systems well in advance of commercial release to facilitate the development of compatible products, the fact that a feature of an operating system works better than a substitute stand-alone application would certainly not violate the antitrust laws.
The provision renewed nothing more than a normal and legitimate understanding in a cross-marketing initiative, and it was not exclusionary. The provision appeared in Microsoft's agreements with only twelve ISPs in the United States, and those agreements did not prohibit the affected ISPs from distributing the web browsing software of Microsoft's rivals. The ISPs also were free to offer other Internet services (not participating in the referral program) that promoted competing web browsing software. In short, the provision was no more restrictive than similar provisions typically found in cross-marketing arrangements. Regardless of the industry, few if any firms (including those without the slightest hint of monopoty power) are willing to refer customers to third parties without some assurance that those customers will not be switched to the competition. While the provision was hardly remarkable and clearly justified, it proved to be not very important to Microsoft's business. As a result, earlier this year when Microsoft was reviewing its agreements with participating ISPs, it modified the contracts to require merely that an ISP promote Microsoft's web browsing software no less favorably than the ISP promotes competing web browsing software. (Microsoft subsequently made essentially the same change to similar protional limitations that appeared in its contracts with a few Internet content providers that participate in Microsoft's Channel Bar, a feature of WIndows by which Internet content is automatically downloaded to a user's PC.)
While the "first screen" provision generates significant efficiencies, it is also crucial to keep in mind that it is in no legal sense exclusionary. Under the terms of Microsoft's license, a computer manufacturer is free to install on its PCs additional non-Microsoft user interfaces to which the consumer can switch with a single "click of the mouse." After a new PC's very first "boot up" is complete, the machine's owner can configure the machine to boot to a different first screen thereafter. Similarly, computer manufacturers are free to load other third party applications and utilities - the icons fee which computer manufacturers are free to place on the first screen of Windows - even if those applications and utilities compete with features of Windows. Indeed, if, as Microsoft's critic claim, the Windows first screen is the world's most valuable commercial real estate, more than three-quarters of that screen is "vacant" and freely available to a PC manufacturer to accommodate the icons of applications and utilities provided by the manufacturer and third parties.
In order to succeed, an antitrust challenge based on such an argument would require a dramatic change in the way the courts have historically interpreted section 2 of the Sherman Act. As explained above, in section 2 cases, the courts historically have avoided even toying to balance legitimate business justifications against perceived anticompetitive effects - antitrust courts are simply not capable of engaging in such complex and subjective balancing. Moreover, Microsoft's conduct that some have characterized as "Tipping" can just as easily and more accurately be characterized as competition on the merits. Microsoft, no less than Netscape, Sun, IBM, or any other competitor, should be free to convince consumers that its vision of computing after the next paradigm shift is the desirable. Microsoft should also be free to be free to bring that vision to market as quickly and efficiently as possible. II. Remedies and The Inescapable Threat of Regulation If Microsoft has not violated the law, then - no matter how much the lawyer at the Department would like to change or restrict Microsoft's behavior - they have no power to impose any remedies. Thus, unless and until there is some indication that the Department has meritoricus claims against Microsoft, it is inappropriate for the Department even to contemplate remedies. Nevertheless, if and when the Department (appropriately or not) reaches the issue of remedies, it is a virtual certainty that any remedy the Department might consider would enmesh the Department in broad regulation of the software industry. Even in the best of circumstances when it has explicity tried to avoid regulation, the Department has almost always found itself acting as a regulator under section 2 decress. The examples of regulatory decrees abound. However, one need go no further back than 1982 and the AT&T decree to find a striking example of how difficult it is to fashion a non-regulatory remedy when the Goverment seeks to constrain or manipulate the ability of a company or companies to compete. The AT&T consent decree (also referred to as the "Modified Final Judgement" or "MFJ") marked the end of the Department's section 2 case against the nation's telephone monopoly. In addition to beaking up "Ma Bell", the decree also restricted the ability of the divested local telephone companies (the Reginal Bell Operating Companies or "RBOCs") to enter competitive, non-regulated businesses. Though the Department believed that it had crafted decree restrictions that were simple, self-enforcing, and non-regulatory, those restrictions very quickly mired the Department in a regulatory quagmire that was only eliminated with the enactment of the Telecommunications Act of 1996. Within a year of the beak-up of AT&T, there was a small staff of Deaprtment lawyers doing nothing but applying the decree restrictions to the day-to-day operations of the RBOCs. Decisions that most businesses make in a matter of days and weeks took the RBOCs months or years because they had to be cleared with the Department and, in some cases, even the decree court. Moreover, because of the evolution of technology, definitions that made sense in 1982 mandated inefficient conduct only a few years later. The Department's experience with the AT&T decree bodes ill for any effort to find "non-regulatory" remedies to impose on Microsoft, particularly if the concern is "tipping." Remember the AT&T experience is not ancient history; it is recent and this Administration was every bit as responsible as its predecessors were for implementing the decree in a regulatory manner. Moreover, no one can doubt that Bill Baxter, who authored the decree, was sincere in his search for a non-regulatory solution to the competitive problems posed by the AT&T monopoly. While reports indicate that the Department recognizes that a "break-up" of Microsoft would not be warranted (and would make no sense in any event) the Department might fool itself into believing that it can devise modest but effective "remedies" that are non-regulatory. It has been suggested, for example, that one such remedy could be found in a decree provision that simply requires Microsoft to make any new operating system feature optional to computer manufacturers. Though it may sound modest, such a remedy would in fact be highly regulatory. First, such a provision would require Microsoft to consult with and defer to the Department in designing its operating systems, complicating and slowing the process to an intolerable extent. Second, giving each computer manufacturer the option of picking and choosing features of the operating system would defeat tie efficiencies and consumer benefits associated with ensuring thet Microsoft's operating systems are consistent throughout the market. Currency, that consistency ensures software developers that, if they write their applications in a way that uses (and depends upon) certain functionality in (i.e., system services provided by) Windows, such functionality will be present in every copy of Windows that is installed on a PC. That consistency also ensures consumer that the features they associate with and expect from Windows will be present on any PC with Windows, regardless of the manufacturer of the PC. Not only is it more efficient to market Windows nationally and internationally if it has a consistent feature-set, but it is also makes using computers easier (i.e., consumers don't have to releam how to use a computers features every time they move from one brand of PC to another). Third, because of the integrated nature of the various functionalities in an operating system, it is no simple matter to determine the impact that removing a given function will have on the rest of the operating system. The cost and delay associated with testing and "de-bugging" the numerous resulting varieties of Windows would be huge. Moreover, if any new feature of Windows must be made optional to computer manufacturers, the number of potential varieties of Windows that Microsoft would have to support (for example, by providing customer service) would be virtually infinite. The enormous cost and confusion would largely be borne by unhappy consumers. Finally, such a regulatary remedy would undoubtedly beget even more regulation. For example, the Department might decide that such an option would only be meaningful if Microsoft charged more for a fully functional operating system than for an operating system from which certain features had been removed. That step, however, would inevitably involve the Department in regulating Microsoft's pricing decisions, a truty radical interference with the operation of the free market * * * Based on the available information, it would appear that there is simply no viable section 2 case against Microsoft. That should be the end of matter - there is no need to consider remedies. However, if the Department launches a quixotic campaign to persuade the courts to overturn the accumulated lessons of over one hundred years of section 2 case law, the cost to innovation and consumers will be enormous.
Microsoft VP Steve Ballmer speaks Steve Ballmer has opinions-lots of them. As the Microsoft executive vice president of sales and support, he has to convince the buying public that the software giant has the goods to be a leader in the Internet-domained world. Ballmer held court on Microsoft and the Web during a visit to CNET earlier thisweek. CINET: Have you seen any significant improvement in market share for Internet Explorer? BALLMER: Some improvement We're still at the level where even that improvement isn't something where you go. "Oh, oh, oh!" On the intranet, I think we're starting to make some real inroads. It's not clear to a lot of people why they ought to go out and buy the Netscape browser when our stuff will be very well integrated with Windows and will be very integrated with our Office stuff. So certainly in that [intranet] environment we've seen some great progress, but it's a little harder to measure. But I sure feel like there's a lot of momentum in the customers I visit. I see people moving fairly nicely. CINET; Do you have a specific expectation on the Internet? Do you think you'll get 20 percent by the end of the year? BALLMER: I don't think we have a stated target. We just know we have to put our heads down and get everything we can, as fast as we can. AOL itself, if that all works out right, should be a significant bump, almost independent of everything else, but we just have to keep our beads down. We've got to ship IE 3.0-that's actually a very important milestone. We've got to get content that really takes advantage of IE 3.0. IE 3.0 is very customizable which from the number of provider standpoints, is very attractive. The best way to become number one is to first become number two! And the best way to never be number two is to seize number three and number four! When we started building marked share in word processing, we didn't take it away from WordPerfect at first. Eventually we started cutting into them, CINET: Are there going to be just two horses in this browser race? BALLMER: You tell me, I think so. I have not seen significant momentum, at least in terms of what ISVs [independent software vendors] or ICPs [independent content providers] are talking about. Other than our stuff and Netscape's stuff, but maybe I'm fooling myself. But I think that's pretty true. CINET: Bill Gates at Internet World said he thinks the Web browser is the fastest-growing piece of software out there and that if we don't watch out we're going to have another OS an our hands. Many people see Netscape Navigator as an alternative, on some level, to the traditional OS. BALLMER: Yeah, [Netscape] wants to be an OS. That's right. CINET: Is Internet Explorer 3.0 not headed for the same goal? BALLMER: No, IE 3.0 doesn't need to reproduce that. We have to make sure that we share well with the rest of Windows. That's basically the difference. So it doesn't need to be another OS; it just needs to add on certain things that you don't gel today in Windows. We need to integrate in. There doesn't need to be an Explorer and an Internet Explorer. There just needs to be one thing. We'll do that later this year. We don't need to have multiple ways of letting people view email. That can be integrated and not done twice. In other words, we don't need two programming models; ActiveX attempts to unify the programming model. CINET; How seriously is Microsoft going to get into providing content? BALLMER: Quite seriously. The content field is vast, so it's not clear what it means to say "seriously." But we clearly have plans to try to add value, to do some exciting things in more than one area. Slate [an online political magazine set to launch this summer] is a good example. What we're doing with MSN is a good example. I read that rumor about old CityScape-who knows whether that's a good example or not, but we've done some stuff in health already. I think our pregnancy stuff is available today. I guess "health guides" is the best way to characterize what we've done there. Car Source, of course, will be 100 percent Web based. You can call it content or not--it's more of a stopping and information guide. It's a form of content, absolutely. The work we're doing with NBC certainly falls into that category. So we'll have a number of things that we try to do. Hopefully, some of those will be viewed positively. I'm sure we'll get some bumps and bruises from some of them. CINET: And your business model is to have users pay for the content? BALLMER; Well it's a mix, like everybody else. Where appropriate, each of the- properties will have sort of a different mix of zero subscription fees, all advertising-funded, to exactly the reverse. We know there's definitely a subscription fee for Slate. Not only might that be important in having it make money, but it's very important to [Slate editor Michael] Kinsley because he doesn't come from the same kind of world that maybe some of us do. He doesn't think anything is worth owning if you didn't have to pay for it. That's the magazine world. CINET: A year and a half ago the platform for most users was Windows, and today on a PC, for a lot of people the platform is Windows plus Office. Very soon it will be Windows plus Java, Office, a built-in browser, and ActiveX components. What does Windows take on next? BALLMER: Telephony, video, MPEG, 3D. There are a lot of things that relate to the way people communicate to the platform. One of the key messages of the whole browser thing isn't so much the Internet, because we always bet that communication prices would come down and people would communicate. But people like hypertext links. If you ask me what's the number-one lesson we should have learned, it's kind of a shame on us-we had this very tree-oriented metaphor. People turned out to like this hypertext link mode. Now, I think there's sometimes people who actually like to go back to the old tree metaphor. Being able to mix those two modes well is important, but there's always something on the frontier of what people want to do in the way they organize information to express themselves, to communicate. CINET: Are Microsoft's expectations in line with Windows 95's penetration into the corporate market? BALLMER: Right now, we've exceeded our plans reasonably significantly. CINET: If you're taking market share with Windows 95, who are you taking it from? BALLMER: OS/2 has kind of faded, so primarily we take it from Windows 3.1. The battle for the operating systems isn't about Windows. It's really about what's the future for Windows. Is the future of Windows Windows, or is the future of Windows Navigator? Windows' number-one competitor is Netscape today much more than it is OS/2. CINET: Why do you say that, even if Navigator isn't a fully fledged operating system? BALLMER: Oh, because [Netscape's Marc] Andreessen says it's going to be. CINET: Really? BALLMER: Whatever you want to call it-an operating environment. Aren't you fined a quarter for calling Navigator a "browser" at Netscape? You're supposed to call it a "client." CINBT: Do you take that seriously? BALLMER: Sure! Let me tell you a story. Once upon a time there was a piece of software that was an extension of an operating system, and it had a nice little user interface and it had some programming interfaces and people kind of liked it, and over time they built on top of it. One day, the thing that it was built on top of wasn't all that important anymore, and it kind of got subsumed inside of the thing that was originally an extension. OK, well I'm telling you, of course, the story of Windows 95, Windows, and DOS. And when we tell the story about what's happening today wilh browsers ten years from now, I want the thing that replaces Windows to be Windows. I don't want to wake up in a position one day where the guys at Netscape say, "Isn't Windows just that little thing that we use to put up menus and draw lines? Let's just write our own and suck it up into our client." Is that a real competitive risk or not? It is if we screw up, you bet! That's the way we will get smacked over the head. If I wanted to compete with Microsoft, I wouldn't say, "Oh yeah, left try to write a better memory manager, and we'll have lighter-weight threads than they do." In this day and age it's not sexy enough, and there's no user interface. The OS/2 experience proves full well the value of that. I would do what Netscape is doing. I'd say I'll build on top of [Windows], and I'll take their future away from them. So yeah, we look at [Navigator] as a very direct threat for Windows if we don't do a good job. That's not a threat a month from now or five months, from now, It's a longer-term threat, but it's a clear threat. It's sort of an everyday issue, but if we screw up, you won't know for a few years. You'll know day-by-day what our browser share is, but that doesn't really speak. So the issue, doesn't threaten the fundamental health of Windows. I'm not sure any of us will know which day it was when we got it salted away. So we're just going to keep working very, very, very hard-at least the rest of my working life! [Windows] is our core franchise. What else is worth working hard on? Back to Front Door Copyright ©1995-98 CNET, Inc. All rights
FOOTNOTES 1. See Sibley Appendix to this Declaration (hereinafter "App.") Exhibit 1. 2. See International Data Corporation (1996), Internet Commerce IDC Internet Fact Book; First Half of 1996 Edition, p. 8. According to International Data Corporation, the latest tabulations from The Internet Society indicate that there are ten million Internet hosts, or machines connected to the Internet, a number that is predicted to grow to 120 million by the year 2000,, The potential impact of electronic commerce on competition is dramatic, as described in a recent White House Report entitled A Framework for Global Electronic Commerce, released July 1,1997. 3. U.S. Department of Justice and Federal Trade Commission, 1992 Horizontal MERGER Guidelines, Section 13. 4. The relevant geographic market is the area that encompasses firms that produce the relevant product such that the hypothetical monopolist of that product would raise the price by a small but significant amount for a nontransitory period. All firms that produce the relevant product are located in the United States. 5. This is the per system royalty for Windows 95, MS-DOS and Windows 3,11, and MS-DOS and Windows for Workgroups, The amount does not take into account the approximately $20 discounts the OEM could receive for meeting all of the conditions specified in Microsoft's Market Development Agreement. See, for example, Sibley Confidential Appendix to this Declaration (hereinafter "Conf. App,") Exhibit 2, MSV 4274-4288 at 4277, Amendment Number 4 to Microsoft OEM License Agreement for Desktop and Portable Operating Systems (dated May 24, 1996) between Microsoft Corporation and Gateway 2000, Inc., Exhibit C I note that the per system royalty fee for Windows NT Workstation is generally greater than $65 and is in the neighborhood of $130 per system. Using a $130 operating system in my analysis would not alter the conclusions presented herein. 6. Using Microsoft's $65 licensing fee in this calculation produces a conservative (or "upper bound'3) estimate of the dollar increase in the price of a PC that would result from a ten percent increase in the price of an OS. That is, a competitive baseline price of less than $65 when increased by ten percent would produce a smaller dollar increase in the price of a PC. REDACTED REDACTED 8. For simplicity, this discussion has ignored retail sales of the operating system, a relatively minor distribution channel. 9. This definition excludes so-called shell browsers, which rely on third party software that itself is marketed as a browser, typically Internet Explorer, It would be inappropriate to include shell browsers in the relevant product market because they are not substitutes for browsers such as Internet Explorer or Netscape Navigator, but, instead, rely on them. 10. For example, a browser acquisition study performed by Microsoft indicates that fourteen percent of browsers are acquired through the OEM channel; twenty-nine percent of browsers through ISPs; twenty-five percent through "work"; thirteen percent downloaded; four percent through retail; and two percent through "s/w bundle.'* See Conf. App. Exhibit 6, MSV 10540-10577 at 10551, Kumar Mehta, IE Market Review, April 1997. In an email with the subject heading, "le data," Kumar Mehta stated that twenty percent of home users of IE obtained it bundled with their computer, while twenty-four percent of those using it at work obtained IE bundled with their computer. See Conf, App. Exhibit 7, MS7 6062-6064 at 6063, Kumar Mehta, "ie data," email to Bob Foulon on March 27,1997. REDACTED In addition, Brad Chase, Vice President of Marketing and Development Relations in the Personal Businesses System Group at Microsoft, acknowledged that "some companies who don't want to have their users go on the Internet** have expressed a desire not to receive a browser with Windows. See Conf. App. Exhibit 9, Deposition of Brad Chase, dated March 25, 1998, p. 80. Set also, Conf. App. Exhibit 10, Deposition of David W. Cole, dated January 9, 1998, p. 49. Mr. Cole, Vice President of Microsoft responsible for overseeing the development of Internet technologies and the Windows users interface, stated: "we had feedback from corporate customers that wanted to prevent access to the Internet, so that when they - they buy a new machine from a PC manufacturer they want the ability to remove easy access to the Internet so their employees, you know, aren't spending their time out on the Web doing whatever," REDACTED REDACTED 13. See Coat App, Exhibit 13, Deposition of Christopher Jones, dated April 8,1998, pp. 77 and 86-88. Jones is Product Manager of the Internet Explorer Team, Microsoft. 14. Increasing returns to consumption is often discussed as an important consequence of network effects. First formalized by Rohlfs, there is a network effect whenever the value to existing users of a network increases as the network expands with new users. See Jeffrey H. Rohlfs (1974), "A Theory of Interdependent Demands for Communications Service," Bell Journal of Economics and Management Science, Vol, 5., No.l, pp. 16-37. See also Michael Katz and Carl Shapiro (1985), "Network Externalities, Competition and Compatibility," American Economic Review, Vol. 75, No. 3, pp. 424-440, and Michael Katz and Carl Shapiro (1986), "Technology Adoption in the Presence of Network Externalities, " Journal of Political Economy, Vol. 94, No. 4, pp. 822-841. 15. See, for example, Joseph Farrell and Garth Saloner (1986), "Installed Base and Compatibility: Innovation, Product Differentiation, and Predation," American Economic Review, Vol 76, No, 5, pp. 940-955, and Joseph Farrell and Garth Saloner (1985), "Standardization, Compatibility, and Innovation," Rand Journal of Economics, Vol. 16, No. 1, pp. 70-83. 16 Conf. App. Exhibit 14, Confidential Submission of Microsoft Corporation to the Staff of the Antitrust Division of the United States Department of Justice, September 19, 1993 (hereinafter "Microsoft Confidential Submission"), p. 61. 17 Conf. App. Exhibit 15, MS 154265-4279 at 154268, Nathan Myhrvold, file attachment (interOffice Memo) to email to Bill Gates, Peter Rinearson, and Jonathan Lazarus, July 24,1993. 19. Shortly after the introduction of MS-DOS, IBM marketed a version of its popular electronic worksheet program, VisiCalc, which ran on MS-DOS. This applications product, together with a handful of applications software products developed for MS-DOS by Microsoft, was sufficiently popular to sustain the operating system until ISVs could bring other applications software to the market. By 1993, it was estimated that 20,000 applications software programs of various types could run on MS-DOS (see Con£ App. Exhibit 14, Microsoft Confidential Submission, p. 62, citing Fax from International Data Corporation to Jennifer Choate, September 10, 1993, p, 2), In 1993, there were also an estimated 5,250 applications software programs available for Windows, the operating system introduced by Microsoft in 1985 (Id at 67/ For the most part, those programs that operated on MS-DOS were compatible with Windows. 20. If the product market were defined to include Apple and non-Intel-based UNIX operating systems, Microsoft's market share would have been 92% (1991), 82% (1992), 82% (1993), 84% (1994), 83% (1995), and 86% (1996). The market share data contained in Table One were obtained from reports published by International Data Corporation, a leading provider of market data in the computer industry. Microsoft's internal documents contain forecasts of its share of licensed PC operating systems that are in the range of 96% to 99% for the years 1998-2001. REDACTED REDACTED REDACTED 24. Conf. App. Exhibit 17, MS7 7193-71% at 7196, Joachim Kempin, email to Bill Gates, Steve Ballmer, and Paul Maritz, December 16, 1997. 25. David S. Evans and Richard Schmalensee (1996), "A Guide to the Antitrust Economics of Networks," Antitrust, Vol. 10, No. 2, pp. 36-40. 26. Conf. App. Exhibit 18, MS6 5004550-4558 at 5004553, Bill Gates, email to the Microsoft Executive Committee, May 26, 1995. 27. Pursuant to the Stipulation and Order, filed January 22, 1998 in the United States District Court for the District of Columbia, Supplemental to Civil Action No. 94-1564 (TPJ), Microsoft agreed to allow OEMs to remove IE using a program included with Windows 95 known as the Add/Remove utility. The Stipulation and Order will remain in effect for the duration of any existing or new OEM license to install Windows 95. REDACTED 29. See, for example, Conf. App. Exhibit 11, Decker Dep., pp. 15-17. REDACTED 31. ActiveX controls are software components that can be embedded in a web page and other applications that allow a programmer to re-use the packaged functionality programmed by others. REDACTED 33. See Conf. App. Exhibit 22, Deposition of Brad Silverberg, dated April 14, 1998 (hereinafter "Silverberg Dep.")5 p, 143, Silverberg is former Senior Vice President of the Applications and Internet Client Group at Microsoft. 34. See Conf. App, Exhibit 23, MS6 6009919, Top North American Internet Access Providers. 35. Included in the providers who siged minimum distribution commitments with Microsoft is AOL, which, according to Microsoft, provided access to over 45 percent of dial-up users. 36. See Conf. App. Exhibit 22, Silverberg Dep., pp. 143-145, Agreements with a minimum commitment of greater than fifty percent are included in the tabulation of "IE preferred." REDACTED REDACTED 38. Conf. App, Exhibit 27, MS7 005527-5528, Jim Allchin, email to Paul Maritz, January 2, 1996. 39. Having said this, it is not obvious that ISVs would have to rewrite their applications entirely, ISVs who want to rely on particular code provided by other software -- referred to as "shared-program libraries'* - typically bundle the needed shared-program libraries with their applications, because Microsoft continually releases new versions of its shared-program libraries included with the Windows operating system. See App. Exhibit 28, Transcript of Proceedings Before the Honorable Thomas P. Jackson, Washington D.C, on January 13, 1998 (P.M. Session), in United States v. Microsoft Corporation, et al, CA No. 94-1564, Testimony of Glenn Weadock, pp. 14-17. 40. In fact, Microsoft has published articles in its "Knowledge Base," a collection of technical notes and advisories available on its website, that describe the method to remove IE from Windows 95. See App. Exhibit 29, Transcript of Proceedings Before the Honorable Thomas P. Jackson, Washington D.C., on January I3S1998 (A.M. Session), in United States v. Microsoft Corporation, et al, CA, No. 94-1564, Testimony of Glenn Weadock, pp. 52-58, See also, App. Exhibit 30, Microsoft Knowledge Base Articles, "How to Uninstall Internet Explorer 4.0/' "Removing Internet Explorer 4.0 for Windows 95 Using Iercmove.exe,1" "How to Manually Uninstall Internet Explorer 4.0," and "Solving Problems Installing or Uninstalling Internet Explorer 4.0'\ http://premium.microsoft.com/isapi/support/apgts/apgts.dll. 41. For example, Cameron Myhrvold, Vice President of the Internet Customer Unit at Microsoft, stated that Microsoft's "Referral server doesn't even pay for itself, much less generate any profits , ., ." See, Conf, App. Exhibit 31, Deposition of Cameron Myhrvold, dated April 24J998, p. 137. REDACTED Conf, App, Exhibit 34, Deposition of Joachim Kempin, dated March IS, 1998, pp. 59-6L See also, Con£ App. Exhibit 35, MSV 0000726-744 at MSV 0000738, Microsoft OEM License Agreement for Operating Systems #2811-5180, dated July 1, 1995, with Dell Computer Corporation; Conf. App. Exhibit 36, MSV 0000411-413 at MSV 0000412, Microsoft OEM License Agreement for Desktop Operating Systems with Packard Bell Electronics, Inc., Amendment 1, Addendum to Exhibit C1, May 1,1996. 43. Ex-post opportunism arises when one party to a proposed transaction must make an investment with no alternative use for the transaction to occur. That party might be concerned that the other party will behave opportunistically once the investment is made and refuse to share the cost of that investment In such a setting,, the first party might reasonably require a highly restrictive agreement governing the other party's ex-post behavior. See Oliver E. Williamson (1985), The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting, Free Press. 44. See Con£ App. Exhibit 37, Microsoft's Answers to Interrogatories, Civil Investigative Demand No. 1S140, Response to Interrogatory No. 1, p. 5; App. Exhibit 38, "Overview of Section 2 of the Sherman Act and its Application to Microsoft," Charles F. Rule, http://www.microsoft.com/corpinfo/doj/rickrulewp.htm. 45. For a discussion of the monopolist's incentive to exclude a competitor's product in a complementary market (and the efficiency justifications for tying) see, for example* Aaron Director and Edward Levi (1956), "Law and the Future: Trade Regulation," Northwestern University Law Review, Vol. 51, pp. 281-296; Robert H. Bork (1978), THE ANTITRUST Paradox: A Policy at War with Itself, Basic Books: New York; and Richard A, Posner (1976), Antitrust Law: An Economic Perspective, Chicago, IL: University of Chicago Press. For a discussion of the case in which the anticompetitive tying can be profitable, see Michael Whinston (1990), "Tying, Foreclosure, and Exclusion," American Economic Review, Vol. 80, No. 4, pp. 837-859. 47. See Conf. App. Exhibit 37, Microsoft's Answers to Interrogatories, Civil Investigative Demand No. 18140, Response to Interrogatory No. 6, pp, 22-23. REDACTED 48. See App. Exhibit 40, Microsoft VP Steve Ballmer Speaks, C/NET News, May 10, 1996 (http://www.news.com/SpecialFeatures/o,5,1300,00.html). 49. See Conf. App. Exhibit 41, Deposition of lames Allchin, dated March 19,1998, p. 120. 51 The Java technology allows software written in the Java programming language to run on any system* regardless of the operating system and the underlying hardware. Browsers that are Java enabled can execute programs written in Java. In this way, a browser can serve as an effective software platform for Java applications, regardless of the underlying operating system. 52. See Michael Whinston (1990), "Tying, Foreclosure, and Exclusion," American Economic Review, Vol. 80, No. 4, pp. 837-859. 53. See Irwin Stelzer, "Why Janet Reno vs. Bill Gates is Good for Capitalism; In Praise of Antitrust Law," The Weekly Standard* December 1,1997. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||