The following article will appear in 31 University of Connecticut Law Review (summer 1999), which will publish papers presented at the 1999 Wiggin & Dana Symposium on U.S. v. Microsoft, March 26, 1999. Additional papers will be authored by William Kovacic, Robert Levy, Leonard Orland, William Page, Charles Rule, Will Rodger, and Jonathan Zittrain. Produced here with permission.
The Importance of the Microsoft Case
Albert A. Foer1
As I write, there is a recess in the trial of the Microsoft case. Both sides have presented evidence to the U.S. District Court, but there has been no rebuttal and of course no decision on the merits, no remedy, and no appeal. To pontificate on the importance of the Microsoft case, therefore, requires projecting ourselves well into the future.
Futurists tell us that it is a mistake to predict the future, as if there were only one future. Rather, they say, there are various potential futures, each of which can be developed from specified assumptions and assigned some probability. The nice thing about this type of predicting is that you can only be wrong if you are dumb enough to assign a 100% probability to a given scenario of the future. (Is the weather forecast wrong when it predicts a 50% chance of rain and the sun shines all day?) On the other hand, there is a value to this exercise, because developing the scenarios should help us understand the dynamics underlying our forecast, in this case the relationship of the Microsoft case to the evolution of high technology industries and the antitrust laws.
So, let us construct the following six scenarios of the future and see what happens. (And let us humbly recall a happy Harry Truman holding up his copy of the Chicago Tribune the morning after it declared Dewey’s victory.) The alternative headlines some ten years from now might read:
(1) Microsoft Won! Competition Reigns!
or (2) Microsoft Won! The Monopoly from Hell Entrenched!
or (3) Microsoft Won! Now We Have the Oligopoly from Heck!
or (4) The Government Won a Pyrrhic Victory!
or (5) The Government Prevailed! Remedy a Loser!
or (6) The Government Prevailed! Competition Reigns!
How history will view the Microsoft case will depend on which, if any, of these scenarios turns out to most closely match actual events.
Scenario I: Microsoft Won! Competition Reigns!
The plaintiffs (the United States and nineteen states) had alleged four separate claims, two under Section 2 and two under Section 1 of the Sherman Act.2 Handing Microsoft a total victory, the trial court found that the facts failed to support any of the claims: (a) Microsoft did not "willfully maintain its monopoly power in the market for PC operating systems," and (b) Microsoft did not "willfully attempt to monopolize, and to destroy effective competition in, the Internet browser market through a course of conduct that includes tying and unreasonably exclusionary agreements" and (c) Microsoft did not "enter into exclusionary agreements that unreasonably restrain trade," and (d) Microsoft did not "unlawfully tie its Internet browser product to its separate monopoly Windows operating system product."
Scenario I assumes that Microsoft is left free (a) to engage in agreements that its competitors (and many independent observers) view as exclusionary3 and (b) to incorporate what could be stand-alone applications software into its operating system4, and (c) to bundle its applications at what appears to be a zero price.5
There are two major variations in the outcome, one positive and one negative. In Scenario I we assume a salutary outcome, namely that the Government’s case was unnecessary because competition became healthier, even in the face of a Microsoft victory. The real or potential competitors who had been lurking in Bill Gates’ anxiety closet, with names like Linux and Java and AOL/Netscape, accomplished what analysts would later describe as a miraculous growth, imposing a "new paradigm" of the sort that had so worried Gates in the e-mails and testimony presented at trial. A vigorously competitive operating system market appeared within a few years after the trial concluded, innovation flourished, and prices went down.6 Microsoft suffered the IBM fate of falling back into the pack after its prolonged antitrust conflict ended successfully. The historians do not report whether Bill Gates auctioned off his lakefront home and went on welfare.
Scenario II: Microsoft Won! The Monopoly from Hell Entrenched!
But a Microsoft victory might lead in a different direction. In the negative outcome scenario, following its victory in court, Microsoft marched on from strength to strength. It quickly took over the enterprise market7 soon after introducing Windows 2000, replicating its success in the PC market by tying its applications to its operating system. It continued to learn from competitors’ software, and made its own versions that it bundled with the operating systems and priced at zero until the competitors were vanquished. A demoralized Justice Department, whose budget had been slashed by Congress amid charges of government waste and interference with the American technology juggernaut, did nothing.8 State Attorneys General were also quiescent, having learned that big but unsuccessful antitrust cases curry little favor with the local voters.
Microsoft also continued to build its content-providing business. By the year 2005, it dominated operating systems and applications, and was in control or gaining control over new delivery technologies such as hand-held computers, TV-internet interfaces, and voice recognition. When computers all over the world were turned on, Microsoft controlled what went onto their screens. It controlled the order in which browsers reported information. It supplied content for most of the leading websites, many of which it had purchased with its enormous cash reserves. Bill Gates had come to dominate the world’s flow of information, becoming what Time Magazine called "the most powerful individual in the history of the universe."
By this time, of course, financiers had stopped investing in companies or products, which were not fully sanctioned by Microsoft. Innovation became completely channeled into products that Microsoft wanted to introduce. Prices of Microsoft products went up and as e-commerce (under the control of Microsoft, of course) became the driving force in the world’s economy, Microsoft charged a small transaction fee on a substantial and growing portion of the world’s commerce. Finally, a political reaction occurred around the world. In the U.S., an effort was made to regulate Microsoft as a natural monopoly, but, according to political scientists whose books were extremely difficult to find on the Internet, Microsoft’s domination of the political system, through its unprecedented financial contributions to charities and politicians, rendered reform an impossibility. It was only through the initiative of international organizations and their member states that a substantial challenge was finally beginning to take form in 2009.
Scenario III: Microsoft Won! Now We Have the Oligopoly from Heck!
Scenario III builds on the reaction of the non-Microsoft part of the computer industry after Microsoft had won its victory. A coalition of closely coordinated companies, which had come to know each other in the course of their mutual opposition to Microsoft during the trial, became increasingly formalized. (The Justice Department quietly cheered this process on.) While it could not displace Microsoft from its OS monopoly, the coalition opposed Microsoft’s expansion wherever it could in area after area.
For the public, this offered limited options, but was an improvement over the monopoly from hell model. Eventually, the companies that were not part of this coalition formed a second coalition, and the computer world came to resemble the airline industry, with a small number of alliances competing around the edges but rarely in a direct way. As Ralph Nader revealed to Ted Koppel on Night Line, "This is only the illusion of competition".
Scenario IV: The Government Won a Pyrrhic Victory!
We turn now to scenarios in which the Government prevails in the trial. In the first of these scenarios, the court found violations of the antitrust laws, but wimped out9 on the remedy. In fairness, we can’t blame it all on Judge Jackson. While the trial itself was handled expeditiously, the appellate courts bounced the case back and forth several times. Years went by before there was a final judgment. By that time, the industry had changed so much since the case had begun that there was no justification for a strong remedy and in any event it had become clear that no more than a slap on the wrist was going to be upheld by the appeals courts. Microsoft lost the battle but it won the war.
Scenario V: The Government Prevailed! Remedy a Loser!
In our fifth scenario, various conduct remedies were imposed and implemented. In retrospect, the remedies were unduly regulatory in nature. Judge Jackson found himself pulled deeper and deeper into the role of information industry czar, and was constantly second-guessed by an appeals court with different ideas. Definitional issues kept cropping up as innovations and ambiguities had to be accounted for and the court frequently had to draw lines. Microsoft, competitors, customers, and consumer organizations were continually coming to the court with motions and objections to every conceivable issue.10
Because of a poorly conceived remedy, Microsoft itself became so bogged down in a regulatory climate that its most energetic and creative people grew restless and eventually moved on to other, more entrepreneurial companies. Leadership of the industry shifted to others. Microsoft’s stock entered into a long decline. Some scholars, looking back, felt that the remediation of Microsoft was the turning point that led surprisingly quickly to the loss of American economic dynamism. "Anti-antitrust" organizations in Chicago and Los Angeles burned effigies of Tom Jackson, Joel Klein, and Bob Bork. On the other hand, some scholars concluded that although the remedy had been too regulatory, the principal victim was Microsoft. While its hands were semi-tied, the rest of the industry had an opportunity to flourish.
Scenario VI: The Government Prevailed! Competition Reigns!
In the last scenario, the porridge is neither too hot nor too cold. Microsoft was found to have violated both Section 1 and Section 2, and historians tell us that the Justice Department, Judge Jackson, and the Supreme Court all got the remedy just about right. Using a combination of conduct and structural remedies, which we need not11 describe here, they constructed a regime that was largely self-executing. By creating the incentives for Microsoft and its structural offspring to act in a competitive manner, they succeeded in generating an industry in which a multitude of software companies thrived, innovation continually brought on new products, and even the shareholders of Microsoft walked around with [less-arrogant] smiles.12
What Is Probable?
So let us ask again, what is the importance of the Microsoft case? From the evidence adduced to date, as reported in our major newspapers, I assign my own probabilities for the various scenarios.
Microsoft wins! I place the probability of a Microsoft exoneration-victory at 1 in 3. The odds that the trial court will rule against Microsoft are higher than 2 in 3, but the appellate courts will be temperamentally inclined against the Government. Nonetheless, my sense is that the evidence will rest largely on findings of credibility that will not be overturned, and that the legal theories of the case are not going to require much groundbreaking.
To the positive scenario –Microsoft wins and competition asserts itself—I give only a 1 in 12 probability. In the absence of antitrust, it seems to me rather likely that Microsoft will become more completely entrenched. What inklings of competition we see today are to some extent the result of an easing-off that Microsoft has manifested under the pressure of litigation. If Microsoft wins, it is likely that antitrust will turn away from Microsoft and more generally from large or controversial undertakings. Congress will attack the Justice Department and the agency’s morale will suffer a major downturn. Aggressive antitrust will disappear for a period of time. As the rest of the computer industry adjusts to the fact that there is no antitrust shield, potential rivals (and their investors) will back away from direct competition with Microsoft.
This leaves a 1 in 4 probability of the "monopoly from hell" scenario or the "oligopoly from heck" scenario.
The three "Government Prevails!" scenarios together get a 2 out of 3 probability. But what are the chances of a workable remedy versus a failed remedy? Actually, they are fairly good, for several reasons that may not be obvious.
First, the closer the case moves toward the inevitability of a remedy being imposed, the more attention will be given to the remedy’s structuring, with an increasing involvement of the industry and Microsoft itself. AT&T ended up playing a large role in the design of its divestiture, and this worked out reasonably well both for the public, the industry, and for AT&T’s shareholders. There is virtue in having the company participate in the design of a remedy that solves the legal problem. For this reason, I would like to see Judge Jackson rule only on liability and propose a general rather than detailed remedy. At that point, he could certify his opinion to the appellate courts for an interlocutory decision.13 Only after there is a final holding of liability (which spells out what wrongs must be righted) does it make sense to focus in detail on the remedy. Microsoft, which has battled tooth and nail on every issue, will seriously turn its attention to a cooperative crafting of a remedy only when it perceives that a serious remedy is inevitable.
Second, a remedy will necessarily be subject to continuing oversight by the court for a number of years, whether the remedy is conduct-oriented or structural or both. If the remedy is not working, it can and, indeed, must be modified. In the United Shoe Machinery case,14 the District Court found an antitrust violation and applied a conduct15 remedy. Years later the Supreme Court found that this wasn’t enough and held that the lower court had the "inescapable responsibility" to fully remedy the illegal conduct. (At that point, United agreed to divest sufficient assets to bring its market share to below one-third.16) There’s no reason we should be stuck with a remedy that is a disaster.
Third, although neither judges nor bureaucrats may be especially talented in designing or redesigning markets, the remedy phase of the Microsoft case can bring in the best thinking not only of Microsoft and the plaintiffs, and their economic counselors, but also the other interests, public and private, which have a stake in the outcome. It is entirely possible that a reasonable consensus will arise, once the issue is narrowed by a set of authoritatively stated facts and legal principles.
I set the probabilities as 1 out of 6 that the Government will win but the remedy will be a loser and 1 out of 2 that the Government will win and the remedy achieves a substantial increase in competition.17 Note that in my reading of the future, if the Government wins, the probabilities of getting a satisfactory remedy are 75%, whereas if Microsoft wins, the probability of competition asserting itself is only 25%. Or put differently, if Microsoft wins, there is a 75% chance of a really negative scenario, whereas if the Government prevails, this chance is only 25%. Others are of course welcome to set up their own odds, but they should discuss how their reasoning differs from mine.
Returning to the question of the importance of the Microsoft case, it seems to me that history will look back on the following points:
Major Areas of Importance
Less Important But Also Noteworthy
Finally, the Microsoft case will be deemed important simply because it was perceived to be important. For the first time in more than a generation,21 antitrust is discussed in homes and at cocktail parties. In part, this salience is the result of Bill Gates’ extraordinary success and his talent for making enemies. More importantly, it grows from the recognition that the nation is moving into a new era that will be dominated by the computer and the Internet, and that a great deal depends upon whether the new era unfolds along lines of centralization or diversity. Competition policy, for so long ignored by the man on the street, has once again become relevant.
1Albert A. Foer is President of the American Antitrust Institute, an independent non-profit think tank described at www.antitrustinstitute.org . The author gratefully acknowledges insights and comments from Donald Falk, Robert Lande, Robert Litan, Roger Noll, Steven Salop, and Matthew Siegel. 2See United States' and Plaintiff States' Joint Pretrial Statement in United States of America v. Microsoft Corporation, Civil Action 98-1232 and 98-1233 (USDC DC) Oct. 6, 1998. The Sherman Act is 15 U.S.C. sections 1,2. Originally, there were twenty plaintiff States, but one dropped out during the trial. 3In the trial, the Government presented evidence, for example, that Microsoft entered agreements with AOL and Intuit whereby their products would be well-placed on the Microsoft desktop (a strategically important consideration) in return for their using the Microsoft Internet Explorer ("IE", its browser product) to the exclusion of competitors (i.e., Netscape), and with Apple whereby Microsoft would continue to develop Microsoft Office for Mac (Apple's major product) in return for Apple's agreement to make the Internet Explorer the default browser on all Mac operating systems. 4In the trial, the Government presented evidence that the Internet Explorer and Windows (Microsoft's operating system, or "OS") are separate products, that sale of the OS, a product with appreciable economic power in the market, was conditioned on sale of the IB, and that tying them together offered no benefit that could not be achieved by customers putting the two together themselves (or by combining Windows with some other browser). 5In the trial, the Government presented evidence that Microsoft, as operating system monopolist, sold the OS and the IB together -bundled-at a single price that included the software for free, recovering its costs through a high price for the operating system. Since a competing software producer needs to earn a positive return on each unit of software it sells, the argument is giving away the IB effectively prevents entry into the browser market and therefore illegally maintains the OS monopoly. 6At the trial, the Government presented as evidence of Microsoft's monopoly power that while prices for most software and hardware had been falling for years, the price of the Windows OS had been increasing. 7While the Microsoft case has focused primarily on the desktop personal computer, it also has major implications for the enterprise market that includes the networks serving corporations and other organizations. Companies currently in the enterprise market predict that Microsoft, by the planned merger of its desktop operating system (Windows 95 and Windows 98) with its server operating system (Windows NT) into a new product, Windows 2000, will leverage its dominance in PC operating systems into the enterprise market - a market growing faster than the desktop software market. Enterprise servers use various operating systems including several based on an open-source product, Unix. The fear is that if Microsoft is not substantially leashed as a result of the antitrust case, independent software vendors will be precluded from selling competitive products. Microsoft alone would dictate how non-Microsoft software applications will need to be written in order to function properly. 8For a history of the Government's failure in its IBM case and the long-term subsequent demoralization of efforts to deal with concentrated industries, see William E. Kovacic, "Failed Expectations: The Troubled Past and Uncertain Future of the Sherman Act as a Tool for Deconcentration," 74 Iowa L. Rev.1105 (1989). 9A legal term of art, for some reason not found in Black's Law Dictionary. It should perhaps be pointed out that the Microsoft case could be settled instead of litigated to a conclusion. Which scenario would best describe a settlement, of course, would depend upon the terms of the settlement. 10Alternatively in this scenario, Microsoft was subjected to a mindless breakup into pieces that were too small or too ill-constructed to function efficiently. In either event, we should also note that Microsoft was for many years bogged down in dozens of private lawsuits and State lawsuits taking advantage of the judicial finding that Microsoft was a monopoly. 11Or dare not. 12Apparently, John D. Rockefeller's petroleum company stocks doubled in value after the dissolution of Standard Oil of N.J., and AT&T's stockholders also came out as winners after the AT&T breakup. 13The precedent for this is Brown Shoe Co. v. United States, 370 U.S. 294 (1962), in which the Supreme Court affirmed an injunction of a merger on direct appeal from the District Court under the Expediting Act of February 11, 1903, 32 Stat. 823, as amended, 15 U.S.C. 29. The Court held that the injunction represented a "final" decision even though the details of a divestiture plan had not yet been spelled out. "We note, first," said the Court, "that the District Court disposed of the entire complaint filed by the Government. Every prayer for relief was passed upon. Full divestiture by Brown of Kinney's stock and assets was expressly required. Appellant was permanently enjoined from acquiring or having any further interest in the business, stock or assets of the other defendant in the suit. The single provision of the judgment by which its finality may be questioned is the one requiring appellant to propose in the immediate future a plan for carrying into effect the court's order of divestiture. However, when we reach the merits of, and affirm, the judgment below, the sole remaining task for the District Court will be its acceptance of a plan for full divestiture, and the supervision of the plan so accepted." If Judge Jackson could obtain expedited interlocutory review by the Supreme Court, the case could reach a relatively quick conclusion. 14United States v. United Shoe Machinery Corp., 110 F.Supp. 295 (1953). See David A. Balto, "United Shoe: a reminder about relief in monopolization cases," FTC:WATCH No. 517, March 1, 1999. 15United States v. United Shoe Machinery, 391 U.S. 244 (1968). Justice Fortas, writing for a unanimous Court emphasized the court's "inescapable responsibility" to fully remedy the illegal conduct. In a monopolization case, it is the court's "duty" to secure three goals: (1) terminate the illegal monopoly, (2) deny to the defendant the fruits of its statutory violation, and (3) ensure that there remain no practices likely to result in monopolization in the future. 16The Supreme Court held that additional relief was necessary because the decree had not restored full competition. The Court observed that "if the decree has not, after 10 years, achieved its principal objects, namely, 'to extirpate practices that have caused or may hereafter cause monopolization, and to restore workable competition in the market,' the time has come to prescribe other, and if necessary more definitive, means to achieve the result. A decade is enough." Balto, op. cit. 17U.S. v. Microsoft Corp., No. 98-5012 (D.C. Cir., 1998). The Court of Appeals of the District of Columbia, in overturning an injunction based on Microsoft's alleged non-compliance with a consent order prohibiting tying, said, "We suggest here only that the limited competence of courts to evaluate high-tech product designs and the high cost of error should make them wary of second-guessing the claimed benefits of a particular design decision…The question is not whether the integration is a net plus but merely whether there is a plausible claim that it brings some advantage." 18We obviously do not know at this writing how long the remedy phase or the appellate phases will take. Another procedural point of some importance is the role played by the States in the Microsoft case. Some observers say that the Justice Department would not have brought the case at all, or would have brought a weaker case, or would be willing to settle on less aggressive terms, had not so many Attorneys General coalesced to bring a parallel law suit (quickly joined into one suit with the Government's). How the Microsoft case comes out may determine the degree to which the States become involved in future national-scale antitrust cases. 19Microsoft is a case that was argued by the parties daily on the courthouse steps. It was covered by the major newspapers with a remarkable degree of detail by reporters who became quite expert in most aspects of the case and who recorded their journalistic impressions as well as the objective events. Microsoft was also the first case that was extensively reported on the Internet (for obvious reasons), with transcripts, commentaries, and debates shared in an unprecedented way. For example, The New York Times website forum on the Microsoft case had threaded 16,744 messages as of March 8, 1999. 20Microsoft Corp. undoubtedly suffered a public relations loss even as its stock price rose throughout the period of the investigation and trial. There followed numerous conversations in the p.r. industry and corporate boardrooms about the relative merits of Bill Gates' take-no-prisoners, fight-every-point approach versus Intel's simultaneous strategy of gentlemanly dealings and pre-trial settlement with the FTC. 21Antitrust currently has a salience not achieved since the AT&T breakup and possibly since the election of 1912.