banner ad


The landmark Microsoft case raises challenging questions concerning antitrust remedies. In this Article, we propose a framework for assessing the costs and benefits of different remedies, particularly divestiture, in monopolization cases involving network industries. Our approach can assist a court or enforcement agency not only in analyzing the welfare effects of divestiture, but also in choosing more generally among alternative kinds of remedies. The framework would, for example, apply to a court's choice between damages and injunctive remedies or between behavioral injunctions and structural injunctions. After developing our framework, we apply it to the divestiture proposals made by the government and others in the Microsoft case. We argue that those proposals leave open important questions that must be answered before divestiture can be shown to be either the best remedial alternative or to create likely net gains in economic welfare.


The late William F. Baxter went to Washington to fight monopolies in 1981. By the time he returned to teaching at Stanford University three years later, the Bell System had been restructured from one company into eight.1 Baxter's handiwork was the most ambitious antitrust divestiture since the government's breakup of Standard Oil in 1911. 2 As architect of the AT&T divestiture, Baxter believed that a theory of antitrust liability should map coherently on to a proposed remedy. The remedy should end the conduct that is alleged to have harmed consumer welfare and that forms the basis for a finding of liability. The remedy in a public antitrust action should do no more and no less. Regardless of liability, if "there is no assurance that appropriate relief could be obtained," then the government must question the value to consumers of prosecuting the antitrust case at issue.3 Accordingly, on the same day that he announced the AT&T divestiture, Assistant Attorney General Baxter terminated the government's other major monopolization case-the one against America's second titan of information technologies, IBM.4 Elegantly simple, Baxter's principle concerning the efficacy of antitrust remedies deserves the eponym "Baxter's axiom."5 It would serve well as a Hippocratic oath for antitrust enforcers and jurists. In fact, Baxter's insight is really an application of basic principles of welfare economics to the questions of when to bring antitrust cases and how to resolve them in a socially beneficial manner.

About a year into the Department of Justice's pursuit of AT&T, Bill Gates and Paul Allen founded the company that became Microsoft Corporation. 6 Since that time, Microsoft has grown to a market capitalization of approximately $340 billion7 and today symbolizes how a "New Economy" has risen from the advent of affordable, ubiquitous personal computing and the phenomenal growth of the Internet. The company is also a post-industrial giant that has been alternately lionized, vilified, and, ultimately, investigated and prosecuted by the Antitrust Division of the U.S. Department of Justice. It took less than fifteen years for a startup from the West to replace the century-old Bell System as the principal target of public antitrust scrutiny. It remains to be seen whether Microsoft will become the government's trophy for wise enforcement, like the Bell System perhaps, or its haunting nemesis, like IBM. Baxter taught us that the government's proof of liability does not suffice to predict its success in crafting a remedy.

On November 5, 1999, Judge Thomas Penfield Jackson issued his findings of fact in the civil antitrust case of United States v Microsoft Corporation. 8 On November 19, 1999, he appointed as a mediator in the case Chief Judge Richard Posner of the U.S. Court of Appeals for the Seventh Circuit.9 Following weeks of settlement discussions, Microsoft and the government returned to the courtroom on February 22, 2000, to present closing arguments.10 Judge Jackson likened Microsoft to the Standard Oil trust, and one state attorney general said that any remedies ordered in the case must be "drastic and far reaching."11 Judge Jackson issued his conclusions of law on April 3, 2000, finding Microsoft liable for violating the Sherman Antitrust Act.

When Judge Jackson ruled for the government, the task before the trial court changed from determining liability to identifying a suitable remedy. Following Baxter's axiom, any remedy should address the conduct for which Microsoft was found liable and advance economic welfare at the lowest possible social cost. The problem is a challenging one. As expressed by Timothy Bresnahan, the Stanford economist then serving as the Antitrust Division's chief economist, the government's case against Microsoft can be likened to a dog chasing a fire truck: what is he supposed to do once he catches it?13 Well, the dog caught the truck, and the question of what to do was no longer hypothetical.

On April 28, 2000, the government offered its answer: separate Microsoft's operating systems business from its applications business and, among other things, order a divestiture of the firm into two independent companies. Four distinguished economists-Robert E. Litan of the Brookings Institution, Roger G. Noll of Stanford University, William D. Nordhaus of Yale University, and Frederic Scherer of Harvard University-filed an amicus brief the same week which proposed an alternative divestiture remedy and argued that the government's proposed remedy was inadequate and would be hard to administer. 14 They observed that the Microsoft case presents important and novel questions in terms of fashioning a remedy:

[T]his Court will establish in the process of setting a remedy in this matter the contours of relief in monopolization cases where the defendant's value arises primarily from intangible assets in the form of intellectual property rather than the tangible capital assets characteristic of such prior major monopolization cases as Standard Oil, Alcoa, and AT&T. In essence, this case provides an important test of how antitrust law and remedies should be applied in the "New Economy," where informational capital is the scarce and precious asset and physical assets are relatively minor and hardly unique.

These economists argued that aggressive divestiture remedies are more justified in markets characterized by intellectual rather than physical capital. Microsoft predictably responded that a lesser set of remedies would suffice, 16 and other equally eminent economists, including Paul Krugman of MIT, warned of the unintended consequences of a divestiture remedy. 17 On May 17, 2000, the government stated in reply to Microsoft's filing that divestiture was the only practical remedy. 18 The trial court accepted the government's arguments and ordered Microsoft broken up into two companies.

The purpose of this Article is to establish principles for answering the remedial question faced by the court and for assessing Judge Jackson's decision to order a divestiture of Microsoft. In Part I, we explore differences between forms of economic competition-particularly network competition and Schumpeterian rivalry-relevant to antitrust analysis in dynamic industries. Those two concepts of competition are important to understanding conflicting views of the Microsoft case.

Microsoft's opponents have argued that the existence of "network externalities" creates market conditions that justify antitrust intervention against aspects of Microsoft's pricing, product introduction, product integration, and acquisition strategies. A network externality, or "network effect," exists when the value of a product or service increases with the breadth of demand for that product or service. The typical example is the telephone system, which becomes more valuable to any given subscriber as other people subscribe and become available to communicate with the subscriber. As the benefits offered by one network grow, so too do the costs to consumers of choosing, or switching to, a rival offering. Competition in network markets can therefore take on a winner-take-all dynamic with competitive strategies geared towards gaining an early lead in market penetration.

"Schumpeterian rivalry" is a distinct vision of competition that, though not mutually exclusive of network competition, may have implications for the durability of network monopolies and for antitrust enforcement in network markets. In this view, which some critics of the government's case against Microsoft contend is applicable to software markets, firms compete through technological innovation to achieve market dominance, but dominance that is continually challenged and subject to displacement by subsequent innovations. As with network competition, this form of rivalry may have an all-or-nothing flavor. Winners enjoy a period of dominance, during which they receive above-cost prices that include the returns necessary to induce risky investment in product innovation, but are subject to being supplanted by rivals in a later innovation cycle.

In Part II, we draw from principles of antitrust jurisprudence and microeconomics to propose an approach for choosing appropriate remedies in monopolization cases involving network industries. We present a three-step test for assessing the welfare effects of a remedy, which can also be used to compare the relative costs and benefits of available remedies. Step one is to evaluate whether the static (short-term, holding technology constant) efficiency consequences of a proposed remedy will yield a net gain. Do the gains in allocative efficiency (that is, reductions in price and increases in output) exceed the losses in productive efficiency (that is, ability to reduce production costs), if any, associated with a particular remedy? If so, then step two is to compare the static efficiency gains from the first step with any effects that the remedy is likely to have on dynamic (long-term, with technological change) efficiency. Examples of dynamic efficiency include innovation that reduces production costs or develops new products and services for consumers. If the net gain is positive, then step three is to evaluate the remedy in terms of its enforcement costs, broadly defined. The optimal remedy is the one that produces the greatest overall efficiency gains net of enforcement and administrative costs.

In Part III, we describe the government's basic theory of liability in the Microsoft case. We then examine Judge Jackson's findings of fact and his conclusions of law.

In Part IV, we use the axiomatic approach developed in Part II to evaluate the structural remedies proposed to the court in the Microsoft case. We focus our analysis on the vertical and horizontal divestiture remedies requested by the government and by some amici curiae. We also discuss how the analysis would extend to other structural and behavioral remedies such as compulsory licensing, line-of-business restrictions, prohibitions on product integration, disclosure of the application programming interfaces ("APIs"), and limitations on contractual terms with customers. We conclude that the divestiture proposals before the court do not contain the elements necessary to show either that divestiture is likely to create net social benefits or, even assuming it would, that it would do so at lower cost than alternative forms of relief.

In Part V, we examine whether, as has been widely suggested, the 1984 divestiture of AT&T provides the proper blueprint for formulating remedies in the Microsoft case. We conclude that it does not on multiple grounds.

In Part VI, we pose, but leave for others to answer, two more general questions concerning the process of selecting a remedy in the Microsoft case.


Read entire article (PDF).

J. Gregory Sidak is an Expert Economist in the fields of Antitrust, Telecommunications Regulation, Commercial and Investment Arbitration, and Intellectual Property Law. Prof. Sidak is the Ronald Coase Professor of Law & Economics at Tilburg University and the Chief Economic Expert at Criterion Economics in Washington, DC. The focus of his research has been regulation of network industries, antitrust policy, the Internet and electronic commerce, intellectual property, and constitutional law issues concerning economic regulation.

©Copyright - All Rights Reserved