New York Times
November 8, 1999Silicon Valley Cites Concern Amid Glee on Microsoft Case
By JOHN MARKOFF with STEVE LOHR
PALO ALTO, Calif. -- In Silicon Valley, the nation's high-technology capital, Judge Thomas Penfield Jackson's powerful findings in the Microsoft antitrust case have been received with a sense of vindication but also ambivalence.In issuing his findings of fact on Friday evening, Jackson delivered two implicit yet unmistakable messages to the industry. Yes, he found, there is indeed a "Microsoft problem" in the computer business, and the software monopolist's bullying tactics hurt consumers and undermine innovation. He also found that there is nothing special about the software industry: For purposes of antitrust, he found, this high-tech world is not much different from any bricks-and-mortar business.
So any satisfaction that the judge shares the prevailing Silicon Valley view of Microsoft is offset here by concern that his document is a step toward greater government intervention in the high-tech economy.
The judge's findings are not a final ruling in the case, but they will almost certainly serve as the framework for the outcome of the case -- either in court or in an out-of-court settlement between Microsoft Corp. and the government. In short, it seems increasingly likely that there will be a solution that involves mandated changes in Microsoft's corporate behavior, if not more sweeping measures that, in theory, could include breaking up the company.
"This document does change the landscape," said Jim Breyer, managing partner for Accel Partners, a venture capital firm here.
Much has changed in the computer business since the government filed its sweeping antitrust suit against Microsoft in May 1998. But industry executives here say the issues raised in the trial are still of central importance to the nation's high-tech economy.
It is true, they note, that most of the venture capital and entrepreneurial effort in Silicon Valley these days is focused on Internet startups. At these startups, the software and electronic commerce services are built to work on the Internet's World Wide Web rather than being bound to Microsoft's Windows operating system. Such an evolution has become known as a "post-PC era," which some experts say is beyond Microsoft's control as people increasingly tap into the Web using everything from hand-held devices to TV set-top boxes.
Still, the personal computer remains the primary avenue to the Web -- and that promises to continue for years. And the federal antitrust case focuses largely on how Microsoft allegedly used the market power from its Windows monopoly to try to stifle the challenge from Internet software.
Whether Microsoft can use its dominance in personal computing to give it the upper hand in the Internet era, industry executives say, will be an important policy issue for years.
"Some people think we're in some post-PC era, but we're not -- and this issue really matters," said John Doerr, a partner in Kleiner Perkins Caufield & Byers, a leading venture firm that has backed Internet startups, companies aligned with Microsoft and some of Microsoft's leading rivals, including Sun Microsystems and America Online.
There are aspects of Jackson's findings of fact that give people here second thoughts. The judge's sweeping castigation of Microsoft's business practices deals largely with how the company's restrictive contracts thwarted personal computer makers, Internet service providers and others from promoting, distributing or using the products of rivals.
But in his findings, Jackson made some fine-tuned judgments about the software business. He found that the computer operating system and Internet browsing software are separate products. Microsoft, however, chose to fold its browsing software into Windows -- and in a separate but related case, a federal appeals court said Microsoft was free to do that as long as it could make "any plausible claim" of business efficiency or consumer benefit from doing so.
Jackson went further. He found that by putting its browser in the operating system Microsoft made Windows 98 unnecessarily buggy. That product decision, he wrote, "unjustifiably jeopardized the stability and security of the operating system."
Just what does and does not belong in a computer operating system can make for fascinating and impassioned debate among software programmers. But having a federal court or the government second guess such decisions is something that makes many industry executives and experts leery, even those who are critical of Microsoft.
"Everyone in the industry is uncomfortable with the chokehold that Microsoft has," said Hal Varian, dean of the school of information management and systems at the University of California at Berkeley. "But they are also uncomfortable with having the government get involved in software design decisions."
Some suggest that while the post-PC era is still not here, its arrival is imminent and will bring an inevitable decline in Microsoft's authority. "The reality is that Microsoft today is very much a niche player in the non-desktop market," said Greg Galanos, a software strategist in the office of the president at Motorola Corp.
To be sure, industry executives are not shy about offering solutions for curbing Microsoft's power, and have done so since before the suit was filed. They run the gamut from restricting Microsoft to writing restrictive contracts to forcing it to publish its software protocols to breaking up the company.
Nearly all the proposed solutions are highly problematic. Breaking up the company by separating the Windows operating system business from Microsoft's Office productivity programs, for example, runs the risk of merely creating two monopolies instead of one. Microsoft's market share in productivity programs like word processing, spreadsheets and presentation software is even higher than its share of the operating system market.
Mandating that Microsoft openly publish the underlying "source" code for Windows -- another sweeping sanction mentioned -- runs the risk of reducing Microsoft's incentives to innovate because other software companies, at no cost, would get the technology Microsoft had invested to build.
"The fundamental challenge for the government and the courts is to come up with a set of remedies that really do help consumers instead of doing more harm than good," said David Yoffie, a professor at the Harvard business school.
Yet many industry executives agree on what the policy goal should be in dealing with Microsoft: keeping the basic technology standards an open, public resource in the Internet era. Common technology standards have been catalysts for the growth of industries for centuries. In the 19th century, for example, the national railway system could not develop until there was agreement on a common track width and gauge.
"The personal computer industry has benefited greatly by the establishment of technology standards, but they happen to be owned by Microsoft," observed Richard Shaffer, principal of Technologic Partners, a consulting firm. "And for the Internet and electronic commerce to flourish, we need standards for security, payment and software protocols. But the objective has to be that nobody -- no single company -- becomes the next Microsoft in this new world."
Esther Dyson, editor of Release 1.0, an industry newsletter, added: "From a policy standpoint, what you want to do is create an environment of dynamic instability where everyone, including Microsoft, has to innovate constantly to move forward without getting the government deeply involved. You don't want Microsoft controlling the industry, but you don't want the government controlling it, either."
Microsoft's top executives cringe at the suggestion that they control the key technology or the pace of innovation in computing -- in the PC era or any other.
"There is no control," said Steven Ballmer, the president of Microsoft. "If customers don't want to upgrade to the next version of Windows, they don't have to. And if anybody wants to innovate on top of us or alongside us, they can."
In an interview, Ballmer said that the rise of new technologies like Web-based software and the Linux operating system represented a "cataclysmic change" in the software business. "We have no guaranteed position in this new Internet era," he said. "To me, it sure doesn't feel like we control things."
More and more, Microsoft's challengers hope that Ballmer proves to be correct. In his findings of fact, Jackson found that Microsoft's grip on the industry would not be loosened anytime soon by Linux or another rival operating system, the BeOS. The judge called both "fringe" technologies and found that neither posed a challenge to Windows in the foreseeable future.
"I loathe Microsoft and there's no question it's a monopoly," said Eric Raymond, a programmer, author and Linux evangelist. "But I think Judge Jackson underestimated the competitive threat that Linux presents."
Jean-Louis Gassee, founder of Be Inc., maker of the operating system BeOS, said he founded his company because he felt Microsoft's grip on the industry was loosening. The Microsoft suit and the rise of Linux, he said, "made it possible to consider a fringe operating system" -- and, he added, to consider the possibility that it could make inroads against Microsoft.
After the judge's findings of fact, he says, his chances could improve. "It's a staggering document," he said.
Wall Street Journal
November 8, 1999
A Predatory Monopoly
By Robert H. Bork, a former federal judge and a senior fellow at the American Enterprise Institute. He has represented Netscape Corp. and currently is a consultant for ProComp, the Project to Promote Competition and Innovation in the Digital Age.The first leg of the Microsoft antitrust marathon is over, and Microsoft is so far behind as to be barely visible. Judge Thomas Penfield Jackson has made voluminous findings of fact vindicating the government's case. Given the trial record, he could hardly have done otherwise. He concluded that Microsoft used illegal business tactics to destroy threats to its monopoly in personal-computer operating systems. The determination that as a matter of law this amounts to a Sherman Act violation, which will come in the next stage, seems a foregone conclusion, and the case is so strong that any settlement will probably require Microsoft's capitulation to any terms the government may demand. The company's best hope in the relief phase probably lies with the judge, cold comfort though that may be.
On appeal, Microsoft's lawyers face a formidable task, since a trial court's findings of fact are not overturned unless they are clearly erroneous; these seem clearly correct. Microsoft will undoubtedly rely heavily on the D.C. Circuit Court of Appeals' prior ruling concerning the 1994 consent decree between the parties. The appellate panel in that case reversed the trial judge's grant of a preliminary injunction prohibiting Microsoft from incorporating its browser, Internet Explorer, into its Windows operating system.
Though the decision was based entirely on procedural grounds, two of the three appellate judges took the occasion to dilate upon the substantive issue that was not before them. But these judicial asides are clearly not binding, and even if they were they would be far from determinative. The majority opinion said that the court should not "embark on product design assessment. In antitrust law . . ., the courts have recognized the limits of their institutional competence and have on that ground rejected theories of 'technological tying.' "
That's much too strong, for if the courts declare themselves incompetent to judge technological issues, they will create an antitrust immunity for high-tech predators. Whatever the difficulties involved, and however carefully such issues should be approached, an immunity for illegal behavior is hardly a satisfactory solution. The appeals court majority recognized that problem, saying, "The factual conclusion [about joining the operating system and the browser] is, of course, subject to reexamination on a more complete record," and, "Manufacturers can stick products together . . . without the link serving any purpose but an anti-competitive one. The concept of integration should exclude a case where the manufacturer has done nothing more than to metaphorically 'bolt' two products together."
Judge Jackson's findings are that Microsoft has done precisely that, "bolting" products together not for purposes of efficiency but entirely to destroy competition. This question of pro-consumer efficiency versus anticonsumer predation is the pivot on which the case turns.
I have written in the past that tie-in law is mistaken and attacks a nonexistent problem to the detriment of consumers. That is true of the tie-ins I was then discussing. When two products are complementary, tying one to the other does no harm and often helps the consumer. If General Motors had a monopoly on automobiles and required purchasers to take the cars with GM's tires, there would be no consumer harm. The purchaser is buying transportation, and the full monopoly price for that can be extracted in the price of the car. If GM tried to charge a second monopoly price for the tires, it would be demanding two monopoly returns where only one was available and would make less money by pricing transportation too high. No rational profit maximizer would try that, and no one has ever accused Microsoft of not trying to make as much money as it could.
The primary charge against Microsoft in this case involves the tying of a product that is a potential substitute for the monopoly product, a case I had not previously discussed. If tires could evolve into automobiles, an automobile monopolist would have reason to tie in the tires in order to destroy a rival tire manufacturer, thus squelching incipient competition in cars. A Web browser can become an alternative platform for which applications are written, and then it would not matter what operating system underlay it.
Netscape's Navigator browser posed just such a threat. Microsoft, recognizing that it was unlikely to beat that browser on competitive merit (as a plethora of incriminating e-mails showed), saw the danger and responded by requiring those who needed its monopoly operating system to take its browser as well. Possessing a monopoly is not illegal, but preserving one artificially by such tactics is.
If Navigator had been allowed to compete freely and succeeded, application writers would no longer have an incentive to write only for Windows, and the applications barrier to smaller operating systems' competition with Windows would disappear. It was to avert that threat that Microsoft tied Explorer to Windows--not to gain a second monopoly profit (which would be impossible), but to preserve its existing monopoly (which is eminently rational).
There is, however, a way in which the operating system and browser monopolies together can generate additional monopoly revenues. The browser is essential to reach Web sites, and a browser monopoly can commandeer the enormous advertising revenues those Web sites generate by being the sole means of access. That is true whether Microsoft owns the sites or merely drains revenues from them. And if competition developed in operating systems, it would only hasten competition in the browser market.
Predation in this case was heavily reinforced by contracts that prevented Microsoft's customers and Internet service providers from selling Netscape's browsers or in some cases from even mentioning to consumers that the product existed. The Windows monopoly was also employed to stop potential competition from the likes of Apple, Intel and IBM.
Microsoft, whose legal position is untenable, is attempting to recover its losses politically through a massive lobbying and public relations campaign. It has deployed boatloads of lobbyists, made heavy donations to politicians and nonprofit groups that support it, courted journalists unceasingly and even lobbied Congress to cut the budget of the Justice Department's antitrust division. Nothing comparable in intensity, expense and mendacity has ever been seen with respect to antitrust litigation. Combining paranoia and self-righteous belief in its own virtue, Microsoft is using tactics that would have made the Robber Barons blush.
In the end, there are only three choices: restoration of competition by antitrust enforcement, comprehensive regulation of the industry or permitting an unlawful situation to continue and grow unabated. Antitrust is surely by far the preferable solution.
Wall Street Journal
November 8, 1999A Feeble Case
By George L. Priest, a professor of law and economics at Yale Law School.Thomas Penfield Jackson knows that the real power of a trial court comes from its findings of fact. When a district court judge applies the law, the best he can do is to try to persuade the court of appeals that he is right. But when he makes a finding of fact, he binds the appellate court. Those findings can be overturned only if they are "clearly erroneous" or if the judge has "abused" his discretion.
Judge Jackson's 206 pages of findings of fact, therefore, represent his best effort to bind a court of appeals that had earlier reversed him on some of these same issues to his view that Microsoft is guilty of antitrust violations. Don't bother waiting for his legal conclusions, expected by Jan. 1. Microsoft is guilty all right. All that's missing from this document is the case citations. Yet the question remains: How convincing are the judge's conclusions?
Judge Jackson's findings present a powerful narrative of rapacious monopoly. The story begins with the finding that Microsoft's development of computer operating systems, most recently Windows, has created what the judge calls the "applications barrier to entry." In the judge's view, consumers will only want to use an operating system for which there exists "a large and varied set of high-quality, full-featured applications" and for which it seems certain new applications will continue to be developed. Only Microsoft has succeeded in developing such an operating system.
Microsoft's repeated violations, Judge Jackson writes, come from its relentless efforts to preserve the applications barrier. To do so, the company actively tries to snuff out any product that might serve as a platform for other software use, though its efforts aren't always successful. Judge Jackson views Microsoft's behavior toward Netscape--whose Navigator browser is a potential threat to the applications barrier--as particularly nefarious. Microsoft first tries to make a deal with Netscape to control the market, then, after being rebuffed, develops its own browser, Internet Explorer. Microsoft's offense here is using monopoly power to force all companies it deals with to include Internet Explorer in their operations, usually in addition to (not in place of) Netscape.
Microsoft's greatest achievement was its deal with the largest and most important Internet service provider, America Online, to distribute along with its own software Internet Explorer, rather than Navigator. AOL agreed to the exclusivity not from any threats of Microsoft but, according to the court, because of "the attractiveness of Microsoft's componentized approach. Notably, Netscape had not yet developed a componentized version of Navigator."
It was this deal that sealed Explorer's success. As Judge Jackson ends the narrative in August 1998, Explorer has pulled roughly even with Navigator in market share, with Explorer continuing to gain "at Navigator's expense," again proving Microsoft's success in suppressing all potential competitors to its applications barrier.
Though compellingly written, this story has a lot of holes. Judge Jackson details repeated Microsoft threats against equipment manufacturers and software developers, but more times than not, the threats don't work. Microsoft wants Apple to abandon development of its multimedia playback software. Apple refuses. Microsoft imposes its great pressure on IBM to stop bundling its SmartSuite program with its PCs. IBM rejects the idea. Microsoft binds small fry like Concentric and EarthLink to stifling quotas, limiting them from shipping Navigator. Concentric and EarthLink violate the restrictions; Microsoft does nothing about it.
Although Judge Jackson emphasizes the restrictive agreements through which Microsoft ties the industry together, most of these agreements are reached by the company paying equipment manufacturers and software developers to include Microsoft products along with their own. These various co-marketing payments suggest less the stranglehold of a monopoly than the typical efforts of a company to promote new products.
Microsoft's deal with AOL is a giant hole in the narrative. Though the judge concludes that the deal "accomplished no efficiency," his own description strongly suggests that it was concluded because Microsoft offered a superior product.
More striking are the subsequent developments. In November 1998, AOL acquired Netscape for $4.3 billion. Under its deal with Microsoft, AOL had the right to terminate the exclusivity agreement regarding Explorer at the end of 1998 and might have been expected to do so. Instead AOL extended the agreement to 2001. Judge Jackson denounces this agreement, claiming that Microsoft once again is "extinguishing the threat that Navigator posed to the applications barrier to entry." The reality is that even the new owner of Navigator finds it advantageous to have its products bundled with Windows.
The biggest legal weakness of Judge Jackson's ruling is that his conception of the "applications barrier to entry" is surely not a fact but a legal conclusion that the appeals court will reconsider. What Judge Jackson and the Justice Department cite as a barrier to entry--that most consumers value Windows' many applications, especially its Internet connections--is also an efficiency. To break down the applications barrier is to reduce the efficiency and diminish the advantage of the network to the world's consumers.
Neither Judge Jackson nor any court has seriously worked out the implications of competition in the context of a network. A network is different from traditional markets, and the rules governing competition must be different. In a network, the more connections, the more all participants gain.
In a typical market, to tie one product to another may reduce consumer choice; in a network, it may expand choice. Thus, the tempest over whether a browser and an operating system are separate products--relevant to traditional tying theory--may be largely irrelevant in the context of a network. Microsoft's multiple agreements insisting that Explorer be included in various products appear to be efforts to ensure that the network keeps expanding.
Thus it is that, at the end, Judge Jackson is largely at a loss to describe who has been harmed by Microsoft's behavior. Obviously Navigator has lost market share, but that alone is not actionable. Who are the victims? To Judge Jackson, they are those consumers who don't want to be connected to the network: those who don't want Internet browsers on their operating systems. They may be confused, or they may lose some space on their hard drives, or their computers may run slightly slower because Microsoft has insisted on connecting everybody to the network.
I don't doubt that there are such consumers. But to compare their loss to the benefits all of us gain from network connection, or to the loss that all who are now connected might suffer if the expansion of the network is impaired, suggests the feeble nature of Judge Jackson's case.