Trial and Error:
The Government’s Case
Against Microsoft

Introduction | Microsoft’s Fading Fortune | Confused Claims | Death By Government | Conclusion | About The Author | Notes

 

Introduction

In the ongoing antitrust case, the Department of Justice (DOJ) claims that Microsoft is a harm to society, with prices high enough to be called monopolistic, low enough to be called predatory, and innovative products which, though a benefit to consumers, are illegal under antitrust law. But the case against Microsoft is not about protecting consumers or creating healthy competition in the market. It’s about helping frustrated competitors in Silicon Valley and elsewhere to attain market share that they have been unable to win from Microsoft on their own merits.

For the past six years, rival firms have been working to convince the government to help them to collapse the powerful software giant the easy way. With the lens of antitrust law as a guide, U.S. District Judge Thomas Penfield Jackson has laid down a harsh ruling against Microsoft. But the antiquated notions of antitrust law are wholly irrelevant in the self-regulating software market, in which companies must either follow the rapid change of consumer preference and innovation or lose out to rivals.

 

Microsoft’s Fading Fortune

Antitrust policy has changed little since the Sherman Act was passed in the late nineteenth century, but the nature of competition has been remade. Competition in the past century was restricted by numerous factors, including the large investment costs associated with creating a new product, switching product lines, and changing geographic location. In the absence of healthy competition, firms that were able to surmount the high barriers to entry often ended up long-standing monopolists, with the power to generate high profits, even while inflating prices, holding back output, and slowing innovation. The monopolies of today’s fiercely competitive high-tech industry are a different story.

A software company with minimal resources can spin out a desirable product and quickly gain a monopolist’s share of the market. But market competition ensures that the monopoly is only temporary. A new competitor with a better gadget may steal the market, causing the dominant firm a sudden fall. Thus, software firms must compete aggressively on price and product to keep a hold of the market. "A monopoly position is in general no cushion to sleep on," noted the famous economist, Joseph Schumpeter. "As it can be gained, so it can be retained only by alertness and energy."1

Consider Lotus, a company whose market share fell from a whopping 70 percent in 1984 to 40 percent in 1988. By 1993, Lotus had plunged to a meager 15 percent as Excel emerged to dominate the market. In the same way, WordStar, the leading word-processing program until 1986, gave way to WordPerfect, which was pummeled by Word in 1992.2 Such is the nature of the high-tech market—one temporary monopoly after another. Unlike monopolists of the past century, software monopolists must compete aggressively on price, service, and product to avoid being overturned.

Microsoft currently holds roughly 90 percent of the market in operating systems, yet the company does not feature the rising prices, poor service, and halts in technology that typify a traditional monopoly.3 Microsoft offers consumers reasonable prices. According to the National Economic Research Associates (NERA), Microsoft’s $50 price for Windows is far closer to the price that results from perfect competition—in which numerous firms hold roughly the same market share and sell at the same price—than the price emerging from a traditional monopoly. Based on the NERA’s calculations, Microsoft would charge a price of $1,000 if it held the power of a traditional monopoly.4 And Stan Liebowitz, a professor at the University of Texas at Dallas, reports in the Wall Street Journal that price has dropped more dramatically in markets in which Microsoft competes (65 percent) than in markets where it does not (15 percent).5 Microsoft has also offered consumers exceptional service.

In April 1998, Computer Reseller News conducted a survey of high-tech firms, and 46 percent of the respondents ranked Microsoft as number one in customer technology training. IBM came in second at 14 percent, with Novell and Sun Microsystems trailing behind. Netscape, Microsoft’s competitor in the browser market, was ranked so poorly that it didn’t even make the list.6 Microsoft is also the leader in Internet service consumer satisfaction—bypassing even AOL, the Internet Service Provider (ISP) with the largest user base according to a November 1998 survey by J.D. Power & Associates.7 In terms of technological innovation, Microsoft is spinning out new products and features at every turn.

Modem support, fax options, and CD-ROM drivers are now standard parts of Windows. And according to the NERA, some of those features were more expensive than all of Windows is today. The cost of Windows 98 is less than one-fifth of the price of roughly comparable software in 1989.8 Microsoft now faces a host of able competitors in the operating systems market.

If Microsoft were to price Windows Operating System (OS) at an unreasonable rate or offer an unsatisfying Windows upgrade, Linux would be the first heir to the desktop computer market. Whereas Linux was once run mainly on servers, the OS is now available on an Intel platform through heavy hitters such as IBM, Dell, and Hewlett-Packard. The International Data Corporation expects that commercial shipments of Linux will jump by a compound annual rate of 25 percent from 1999 to 2003.9 Just behind Linux are Sun’s Solaris, which featured a 30-percent hike in sales in 1998, and Apple, which garnered 10 percent of the market during the fourth quarter of 1998 with its popular iMac.

If existing competition from the various operating systems doesn’t crush Microsoft’s monopoly, then potential competitors certainly will. Two technologies emerging in the software market are threatening to render the Windows platform obsolete. First, Sun’s Java "write once, write anywhere" language promises that software developers will soon be able to write applications that run on any operating system. Currently, Java programmers are only able to offer functionality of the lowest common denominator. In order to provide the state-of-the-art interactive and multimedia features that consumers desire in today’s applications, Java programmers must rely on the "native code" of an operating system, such as MacOS, Sun Solaris, or Windows. Once Java is perfected, however, cross-platform Java-based web-applications will include all of the features of the most sophisticated software program without compatibility problems.

Second, the advent of web-based software means that software applications are no longer exclusively linked to operating systems. Browsers can run applications, just as operating systems run them. As web-based software grows in popularity and PC-based software becomes obsolete, customers will focus on browsers and the features they offer, while operating systems take a back seat. In such a market, consumers will likely abandon the Windows platform for a cheaper alternative. Here’s how the Wall Street Journal described the situation: "If users don’t need PCs with Microsoft’s Windows operating system or Intel Corp. chips to use Web-based software, the vaunted market power of the duo called Wintel doesn’t seem so unshakable."10

Web-based software is already gaining wide acceptance among businesses. Many companies today spend large sums of money to train employees on word-processing and spreadsheet programs, such as Word and Excel. But the majority of employees uses only a fraction of the features that those programs offer. In order to cut costs, companies are choosing to train employees on streamlined web-based applications that fit the needs of the user. Software companies are jumping into the market to meet the demand. For instance, eAlity, a company headed up by one-time Oracle and Apple executives, is offering server-based accounting and sales applications to small companies like Williams-Sonoma, eLoan, and one subsidiary of eToys.

The DOJ and other commentators argue that "network effects" of the operating system market give Windows—the dominant "networked" good—an unfair advantage over rivaling systems because consumers and software developers will be unwilling to expend the effort and cost involved in switching to a superior operating system.11 Research by Stan Liebowitz, professor at the University of Texas at Dallas, and Stephen Margolis, professor at North Carolina State University, however, reveals that the market does not feature such disasters. The authors debunk several myths of superior products supposedly failing because of network effects and reveal examples of more favorable networked goods winning out over dominant networked goods. For instance, VHS triumphed over Sony’s Beta, which held a virtual monopoly over the market for two years.12 In the same way, compact disks replaced vinyl LPs. VHS camcorders gave way to 8 mm camcorders. And 31/2-inch floppy drives won out over 51/4-inch floppy-disk drives.

If it is the government’s intention to move consumers out of the Windows platform and into another standard, the market is already taking care of that. Web-based software and Java applications will eventually render the standardized Windows platform obsolete.

Indeed, Marc Andreessen, Netscape’s co-founder also recognized the threat of new technologies upon Microsoft’s monopoly position. Windows, Andreesen said, would become nothing more than a "mundane collection of not entirely debugged device drivers."13 Netscape planned to make the Navigator the new standard—not only for the Web, but for all applications.

Holding 90 percent of the market in 1995, Netscape was in a prime position to dominate. Microsoft responded to the threat in the way that all companies should—with low prices and innovative products that benefit consumers. And yet, the company became the target of antitrust claims which are senseless, unless, of course, the government’s goal was not to better competition in the market, but to safeguard Netscape’s browser monopoly.

 

Confused Claims

The government’s wildly imaginative antitrust case revolves around one central claim—that Microsoft attempted to push Navigator from the browser market by "tying" its Internet Explorer (IE) web-browser with Windows 95 and Windows 98, thereby offering IE for free. But Microsoft was hardly the malicious bandit that the government depicts. The company had solid commercial reasons for bundling the free browser with Windows. After all, product integration is Microsoft’s tried-and-true method of luring consumers into buying upgrades.

The first Windows program was an integration of the MS-DOS personal operating system with a graphical interface which allowed users to "point and click" instead of having to type in DOS commands. Windows 95 and Window 98 simply took it a step further, with the introduction of browsing capabilities. The integrated platform effectively optimized internal applications for the Net. For example, Word 97 automatically converts any typed-in web address (URL) or e-mail address into a live link by accessing the seamlessly integrated IE. Quicken and Lotus Notes now allow users with Windows 98 to access current interest rates and stock prices from the Internet without ever having to leave the spreadsheet application. And MGI PhotoSuite 2, an image-editing program, lets users load pictures from Web-based sources onto the software with ease.

The meshing of Internet Explorer into Windows features has changed the Windows OS in more subtle ways as well. Hyperlinks can now be inserted into applications folders. Web shortcuts can be created on the desktop. HTML-based images can take the place of the standard wallpaper that Windows offers. And users can access the "find" option on the Start menu to search for people through a variety of different web directories, such as Bigfoot, Four11, and Who Where? Numerous reviews reveal the consumer benefits of integration.

According to Business Week, "the incorporation of browsing and other Internet functions into Windows is a powerful innovation. It may be very inconvenient for Microsoft’s competitors, but it’s a big gain for consumers, who should be allowed to enjoy the benefits."14 PC Computing reported, "once you experience the improvements [IE 4] makes to your Windows desktop, you’ll find it hard to go back…"15 And PC Magazine noted that "this integration has big implications even when you are not looking at Internet pages. Because IE is now effectively the user interface, everything works the way it does in a browser."16

With a blind eye to the benefits, the court has mused endlessly over the allegedly negative effects of Microsoft’s product fusion. For one, antitrust enforcers say that Microsoft made it difficult for users to completely remove IE from their computers. As a result, disk drives are full of unwanted browser software, companies are concerned that employees are shopping online instead of working, and parents are worried that their kids are calling up porn sites. IE takes up less than one percent of the disk drive. And if employers want to bar employees from surfing, they should just make sure that the employee doesn’t receive a modem along with his computer. Parents can slam shut the Pandora’s Box of the Internet by doing the same.

The only relevant question that government could ask of Microsoft is whether the company behaved unfairly while it was playing catch-up with Netscape. Microsoft could indeed be guilty of an antitrust violation if it had designed its system to repel the Navigator. But IE’s initial lack of success upon its release in August 1995 and the many millions of people who operate Windows and Navigator reveals that Microsoft did not impede the Navigator from hooking onto Windows. By 1998, two years after Microsoft allegedly pushed Netscape from the market, Netscape sold 15 million copies of Navigator and doubled its installed base—from 15 million users in 1996 to 33 million users in December 1998 in the United States alone.17 Almost one-fourth of all new computers sold in 1998 had Netscape on board.18 And even though every computer with Windows came with IE, a whopping 46 percent of people that bought PCs during the first eight months of 1998 were using Navigator, according to survey results.19

Netscape is now a part of AOL—a "Fortune 500" company that is soon to become part of Time Warner, one of the 10 most highly-valued companies in the country. AOL found Netscape such an appealing prospect that it was willing to buy it for $10 billion—twice what the market said it was worth. Currently, one-third of the market is under the thumb of Netscape, and that market share could balloon to more than 50 percent if AOL chooses Netscape over Internet Explorer next year, when its contract with Microsoft presents the option.20

The court’s shaky tying claim will not likely hold up in appeals. Already, the appellate court has shown its skepticism, having previously thwarted a tying claim in a closely related case, by ruling that IE and Windows were integrated, not separate products. In that case, Microsoft argued that if the consent decree of 1995 recognized that Windows 95, which integrated features of MS-DOS and Windows 3.11, was a good thing, then how could the integration of browsing capabilities into Windows be a bad thing?

The D.C. Circuit sided with Microsoft, agreeing that integrating browsing capabilities into Windows was analogous to linking a graphical user interface with an operating system. To win the case, Microsoft had only to prove that integration was a benefit to users. With solid testimonies, Microsoft convinced the court of the increased functionality that the integrated product offered. The government’s predatory pricing claim is equally weak, even on the government’s terms.

Predators generally lower prices below the market rate to drive competitors out of businesses, only to later raise price and recoup short-term losses. But clearly, Microsoft did not succeed in pushing Navigator out of the market. Nor did Microsoft accumulate short-term losses. Instead, the company was raking in immediate profits from its expanded install base. Various portal sites—the gateways into the Internet which sell advertising space and earn commissions based on the amount of traffic flow through their sites—were paying Microsoft in order to be set up as the default home page for IE. Microsoft chose to compete aggressively by taking advantage of the commercial option in order to offer IE for free as early as July, 1995.21 Netscape waited almost three years—until January, 1998—to do the same.22

The cost to integrate the Internet Explorer technologies with Windows has already paid for itself several fold. Microsoft’s revenues for Windows 98 were about $3 billion annually. That’s 30 times what Microsoft paid annually ($100 million) to integrate IE into the operating system.23 Thus, to break even in 1998, Microsoft needed only to license about four percent more units of Windows than it would have otherwise. Given the increased possibilities the integrated product offers consumers, it seems that Microsoft’s decision to integrate was beneficial—both for consumers and for Microsoft.

Instead of battling the serious issues of horizontal price fixing as it did in the past century, the government has moved into quibbling over petty claims such as tying and predatory pricing. Could it be that tying is in reality a competitive way for companies to offer consumers more product for their money? And isn’t predation better explained as competitive pricing that benefits consumers?

Unfortunately, the DOJ’s reading of the case against Microsoft leaves no room for such niceties as consumer interest. With antitrust law turned on its head, the DOJ’s antitrust claims are merely sticks to prop up the collapsing monopoly of the government’s preferred client.

 

Death By Government

The DOJ is floating a variety of conduct-restraining and structural "remedies," all of which are guaranteed to harm consumers. A conduct-restraining remedy could force Microsoft to license its Windows products to competitors, relinquish trade secrets, and abstain from folding products into Windows. But it seems unlikely that the government would be able to enforce such an agreement, considering that the DOJ’s previous attempt to control Microsoft’s practices under the consent decree of 1995 backfired.

Under the decree, the government asked that Microsoft stop tying the Internet Explorer with Windows. That request was voided when the appeals court ruled that Windows and IE were integrated products. Then, the government asked Microsoft to revise provisions of its purportedly "exclusionary" contracts with various companies, such as Internet Service Providers (ISPs) and Internet Content Providers (ICPs). But all exclusionary language had already been removed from those contracts. Next, the DOJ wanted Microsoft to give Original Equipment Manufacturers (OEMs)—the companies that manufacture PCs—more freedom to alter the opening screen of Windows. As it turned out, OEMs could easily alter most of the screen, and consumers could delete the screen completely.24

Another equally preposterous remedy popular among the DOJ and the attorneys general is a vertical or horizontal breakup of Microsoft. Under horizontal divestiture, Microsoft would be split along product lines, into two or three parts. One company would work on the operating system. Another company would take the software applications. And a possible third company would deal in Internet and e-commerce products. But what is the point of spending taxpayer dollars to split up Microsoft by horizontal divestiture when the plan leaves the company’s PC OS monopoly fully intact? What’s more, the plan will slow innovation by forcing Microsoft to petition the government each time it wants to create a new product.25

The second plan involves a vertical split-up of Microsoft into several separate clones—or "Baby Bills." Each of the Bills, with rights to Microsoft’s intellectual property, will produce the full range of products and compete fiercely against one another.26 The Windows companies may choose to create products that are compatible with each other in order to garner a larger portion of the market. But most likely, the Windows companies will create products that are not compatible. After all, a competitive firm usually favors having the whole pie to itself, rather than having to share a slightly larger pie with competitors.27

For example, Unix and Linux would have benefited from a larger consumer base and the attention of software developers if they had chosen to be compatible, but they chose incompatibility. And if AOL had given its competitors the right to use its chat standard, AOL would have been able to maximize its network effects. But instead, AOL refused to open the chat standard to other ISPs.28

If the vertical split-up does indeed destroy the standardized platform of Windows, consumers will suffer irreparable damage. Absent the Windows standard, which guarantees software developers high profits to cover the up-front investment costs, software developers will have less incentive to develop the roughly 70,000 Windows-based software programs currently available to consumers.29 Not only will the number of software products shrink under vertical divestiture, but the price for software will rise. According to one calculation, software development prices will jump an estimated $30 billion under a three-way vertical split-up.30 Software companies will have no choice but to pass that cost on to consumers.

The implications of antitrust enforcement are clear. In the software industry, a firm with a good idea produces a good product, and the best products in the market win. Government intervention in that process is highly destructive, both for consumers and for competitors in the market. The government’s proposed remedies may help competitors to gain a leg up on the software giant in the short run. But in the end, both consumers and software companies will suffer as the result of the dangerous precedent the case has set for the software industry—that for any firm with the lion’s share of the market, success is rewarded with nothing short of regulation or dismemberment.

 

Conclusion

The government needs to reconsider its approach to antitrust enforcement in the Microsoft case. Microsoft’s aggressive competition has consistently brought consumers low prices, superior service, and flourishing technology. The rapidly evolving software market is already moving consumers out of the Windows platform, without help from the United States Department of Justice. None of the government’s proposed remedies will lead to the low prices and highly advanced technologies that consumers currently enjoy. In short, antitrust enforcement in the self-regulating software industry is as obsolete as Windows will be in the near future.31

 

About the Author

As a public policy fellow in the Center for Freedom and Technology, Helen Chaney researches and writes on current technology issues, including Internet privacy, the digital divide, e-commerce, and telecommunications. She has appeared on ZDTV and her opinion editorials have been syndicated by Knight Ridder and Bridge News. Ms. Chaney holds a BA in International Studies from Trinity University.

---

Notes

1 Joseph Schumpeter, Capitalism, Socialism and Democracy (New York: Harper & Row, 1976), 102.

2 Stan J. Liebowitz and Stephen E. Margolis, Winners, Losers, & Microsoft (Oakland: The Independent Institute, 1999), 183.

3 Robert A. Levy, "Microsoft Redux: Anatomy of a Baseless Lawsuit," Policy Analysis 352, 30 September 1999, 2–3.

4 Bernard J. Reddy, David S. Evans, and Albert L. Nichols, "Why Does Microsoft Charge So Little for Windows?," (White Plains: The National Economic Research Associates, 7 January 1999), 3. Available: www.neramicrosoft.com.

5 Stan Liebowitz, "Bill Gates’s Secret? Better Products," Wall Street Journal, 20 October 1998, A22.

6 Reddy et al, 1.

7 Brad Skillman, "Polls show Internet users like MS but are leery of online shopping," Financial Post, 19 November 1998, C10.

8 Reddy et al, 3.

9 Richard McKenzie, Trust on Trial: How the Microsoft Case is Reframing the Rules of Competition, (Cambridge: Perseus, 2000), 51.

10 Don Clark, "Software Becomes an Online Service, Rattling Industry," Wall Street Journal, interactive, 21 July 1999. Available: www.wsj.com.

11 Software products that are compatible with other software products quickly gain market dominance because of the "network effects" operating in the market. In "networked" markets consumers tend to choose one single product over all others for the compatibility the good offers. A "networked" product is only as valuable as the number of consumers who use it. For example, an operating system is useless without compatible software products. But as consumers flock to one operating system, the value of the operating system increases, both for software developers who are drawn to writing a platform with a large number of users and for consumers who benefit from the large number of compatible applications that the operating system affords.

12 Liebowitz et al, 120–127.

13 Bob Metcalfe, "Netscape’s tools will give Blackbird reason to squawk," 18 September 1995, 111.

14 Stephen H. Wildstrom, "Why I’m Rooting for Microsoft," Business Week, 23 February 1998, 30.

15 Ed Bott, "Upgrade Windows and the Web; Microsoft’s Internet Explorer 4.0 Web browser," PC Computing (11) no. 1, January 1998, 117.

16 Michael Miller et al, "Memphis & IE 4.0," PC Magazine, 9 September 1997, 112.

17 Findings of Fact, 378.

18 "The Economics of the Microsoft Antitrust Case: A Post-Trial Primer," (White Plains: The National Economic Research Associates), 2.

19 "Battle of the Browsers—Don’t Count Netscape Out Yet!," Ziff-Davis News Network, 19 November 1998. Available: www.zdnet.com/zdnn.

20 "The Economics of the Microsoft Antitrust Case," 3.

21 Lisa Wirthman, "Microsoft Says It Always Had Web On Mind," Investor’s Business Daily, 11 August 1998.

22 Benjamin Klein, "An Economic Analysis of Microsoft’s Conduct," (White Plains: The National Economic Research Associates, 11 November 1999), 8–11. Available: www.neramicrosoft.com.

23 Summary of Written Testimony of Microsoft Witness Paul Maritz, PR Newswire, 22 January 1999.

24 Robert A. Levy, "Microsoft Redux: Anatomy of a Baseless Lawsuit," Policy Analysis 352, 30 September 1999, 13.

25 Levy, 13–14.

26 Ibid.

27 Stan J. Liebowitz, "A Fool’s Paradise: The Windows World After a Forced Break-up of Microsoft," (Washington D.C.: The Association for Competitive Technology, 25 February 2000), 10–11.

28 Ibid.

29 Findings of Fact, 40, 46–47, 49.

30 Liebowitz, 2.

31 Levy, 16.