Bork's 'white paper' (continued)
... But a monopolist is different. There is nowhere for its customers and suppliers to turn in order to avoid the onerous terms.
A monopolist cannot, of course, demand both a full monopoly price and also impose unwelcome restrictions on those with whom it deals. That would entail charging more than the maximizing price and would result in lower sales and profits. Although the monopolist holds the whip hand, it must pay in lower prices for any additional restraints it demands. The war Microsoft wages here is a form of predatory pricing through the payments it gives up in return for its exclusionary practices.
Such payments, however, may be worth the cost in order to block competitors and thereby preserve the monopoly. The record shows numerous instances of Microsoft paying or offering discounts in return for exclusionary agreements.
In the usual case, predatory pricing is unlikely to be attempted or, if attempted, is unlikely to succeed. The primary reasons are that the predator and the intended victim often have roughly proportional reserves (liquid assets plus available lines of credit) and that losses during the price war will be proportionately higher for the predator than for the victim. These factors mean that the victim can usually outlast the predator in a price-cutting campaign.
These constraints on predation do not apply in this case.
Microsoft and Netscape do not have roughly proportional reserves or anything close to it. Microsoft's reserves dwarf Netscape's by many orders of magnitude. The market capitalization of Microsoft is approximately $250 billion; that of Netscape is $2.7 billion. Bill Gates' personal fortune alone, which came from Microsoft, far exceeds the capital value of Netscape. The upshot is that Microsoft can easily outlast Netscape in a price war.
Microsoft also makes enormous profits on its monopoly operating system licenses while Netscape has no comparable source of income. As Bill Gates put it, "Our business model works even if all Internet software is free . . . We are still selling operating systems. What does Netscape's business model look like? Not very good."
Price predation also rarely succeeds because the predator must operate at increasingly higher costs while the victim need not. It will often, in fact, be possible for the victim to lower its costs in response to the attack. Before the price war begins, both the predator and victim will produce at rates where their marginal costs are increasing. In order to drive prices down, the predator must increase its rate of output, thus operating at a still higher marginal cost. As market prices decline, costs rise, and the predator suffers losses. The victim labors under no such disadvantage. It may suffer losses, but those losses will not be nearly as large as the predator's because its costs will not have increased. The victim may, in fact, be able to decrease its rate of output, thus moving to a lower marginal cost. This means that the predator will exhaust its reserves not only faster but disproportionately faster than the victim. That is the reason price predation is usually a losing game.
This analysis does not apply to software, however. The major elements of cost there are the fixed costs of physical facilities and compensation of personnel. The costs of production and distribution, once the software is perfected, show little variation with respect to the rate of output. The marginal cost curve is either flat or very nearly so.
Thus, Microsoft need not bear significantly disproportionate costs when both it and Netscape price their browsers at zero. Microsoft, moreover, need not even greatly increase its rate of output in attacking Netscape's Navigator because Microsoft already ships 97% of all operating systems pre-installed on personal computers. The only expansion needed is the rate of browser output and the browser is now, in any event, tightly tied into the operating system. The marginal cost curve with respect to browser distribution is flat. This fact, plus the enormous disparity in reserves, makes the predatory exclusion of the Navigator from the market entirely feasible.
To preserve its monopoly in operating systems, moreover, Microsoft need not drive Navigator completely from the market. It need only reduce Navigator to a size at which most application writers find it unprofitable to write for the Navigator. That would guarantee that Netscape would indeed be a "fringe" firm with regard to browsers.
It is axiomatic that no firm attempts monopolizing unless it has a clear prospect of recouping the costs of its predatory tactics and enjoying a monopolistic return. Brooke Group, Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 224 (1993) ("` For the investment to be rational, the [predator] must have a reasonable expectation of recovering, in the form of monopoly profits, more than the losses suffered.'")(quoting Matsushita Elec. Indus. Co., Ltd. v. Zenity Radio Corp., 475 U.S. 574, 588-89 (1986)). Microsoft, however, enjoys a monopoly return even while engaged in exclusionary practices.
Its present market size requires little expansion of output and the concessions to customers and suppliers in return for restrictive agreements are relatively inexpensive. If the Navigator is destroyed or marginalized, and if Java is altered to work only on Microsoft technology, Microsoft will not even have to deduct these minimal amounts from its monopoly profit.
3. The Court of Appeals Consent Decree Decision In No Way Forecloses the Separate Government Suit Against Microsoft Under Sections 1 and 2 of the Sherman Act
The conclusions of this memorandum are not disturbed by the recent decision of the U.S. Court of Appeals for the D.C. Circuit.
In a decision concerning the meaning of a 1994 consent decree, a three-judge panel of the U.S. Court of Appeals for the D.C. Circuit on June 23, dividing two to one, reversed the District Court's grant of a preliminary injunction prohibiting Microsoft from tying its operating system and its browser in licenses to computer manufacturers. The preliminary injunction was vacated on procedural grounds: the injunction was entered without notice to Microsoft in violation of the Rules of Civil Procedure. A divided panel also said that the District Court misconstrued the consent decree. The procedural ruling obviously has no implications for the separate Sherman Act suit filed by the government in May. Many commentators, however, quickly concluded that the Court of Appeals' substantive discussion of the consent decree's meaning harms the prospects of the government's Sherman Act case.
That conclusion is not justified.
Both the decree and the Sherman Act suit involve, among other issues, the question of permissible and impermissible bundling, the tying of Microsoft's Internet Explorer and its Windows operating system. The decree forbade tying products together unless they add up to one "integrated product." "Integration" occurs, the majority said, when combining two products produces benefits greater than selling them separately and the producer can combine them though the purchaser cannot. The Court said that Microsoft had ascribed "facially plausible benefits to its integrated design." This facial plausibility was apparently enough, under the consent decree, to justify tying the browser and the operating system. That conclusion has less impact upon the government's separate Sherman Act suit than some commentary supposes.
In the first place, the discussion of this subject by the Court of Appeals majority was unnecessary to its decision about the preliminary injunction. The procedural ground was sufficient, and all three members of the panel agreed on that point. Furthermore, the District Court had not held hearings on or decided the bundling issue. Thus, there were no findings of fact, no record, and no briefing of the question. The panel majority's observations are, in the strictest sense, dicta. As such, they are not binding on any future court or even on the two judges who entertained the issue in the consent decree context.
The non-binding character of the panel majority's remarks is clear for a much more important reason: the decree litigation did not bring before the Court the many predatory contracts required by Microsoft or the internal documents that make clear the predatory intent underlying both those contracts and the bundling of the browser and the operating system. That evidence will be central in the upcoming Sherman Act litigation.
Indeed, the Court majority warned that its discussion of integration was not to be final even in the consent decree litigation, much less in the coming Sherman Act trial. The majority opinion makes that point repeatedly. The panel admonished, for example, that the consent decree "does not embody either the entirety of the Sherman Act or even all 'tying' law under the Act."
More important is its emphasis that, "The guidance this opinion seeks to provide is limited to setting out the legal framework for analysis. The ultimate sorting out of any factual disputes is a different question, and one we of course cannot resolve on the limited record before us"; and again, "The factual conclusion [about integration] is, of course, subject to reexamination on a more complete record." In a word, the discussion of integration is tentative and may be altered or abandoned after a trial of the facts.
Directly relevant to the proof the government will offer in the Sherman Act case are the Court majority's observations that, "Manufacturers can stick products together in ways that purchasers cannot without the link serving any purpose but an anticompetitive one. The concept of integration should exclude a case where the manufacturer has done nothing more than to metaphorically 'bolt' two products together." Again: "[T]he overwhelming of the separate market [for the product tied in] is precisely what is feared and may simply indicate anticompetitive practices."
The government will show that the compulsory combination of Microsoft's monopoly operating system and its browser in its licenses to manufacturers is not necessary to efficiency or intended to be. "Sticking" or "bolting" the browser onto the operating system serves no purpose but an anticompetitive one and is designed to overwhelm the market for browsers.
This concept of "sticking" or "bolting" products together highlights the weakness of the majority's discussion of "integration": [I]ntegration may be considered genuine if it is beneficial when compared to a purchaser combination. But we do not propose that in making this inquiry the court should embark on product design assessment. In antitrust law, from which this whole proceeding springs, the courts have recognized the limits of their institutional competence and have on that ground rejected theories of "technological tying."
A court's evaluation of a claim of integration must be narrow and deferential. (footnotes omitted) While this passage speaks in welcome tones of judicial restraint, that restraint is misplaced where, as here, it amounts to giving Microsoft a judicially created exemption from the antitrust laws. The majority thought its test was satisfied because the Internet Explorer adds features to the operating system that cannot be included without also including the browser functionally.
Judge Wald's dissent on this point seems a sufficient rebuttal: The majority's interpretation would not, for example, appear to prevent Microsoft from requiring OEMs to license the right to sell computer peripherals (e.g., mice) as a condition of licensing Windows 95 so long as Microsoft had the prescience to include code in Windows 95 that made the cursor more responsive to the end user's touch than it would be with other mice. . . . I fail to see why the analysis should be so different for software than for peripherals or why our "evaluation of a claim of integration must be [any more] narrow and deferential." (footnotes omitted)
Judge Wald summarized: Thus, antitrust law cannot avoid determining whether a particular technological development has occurred because it is efficient or merely because it permits a monopolist to extend its monopoly to a new market. Software code is a particularly stark example of why such analysis is essential if antitrust concepts are to survive at all. Here, the majority effectively exempts software products from antitrust analysis by stating that "[s]oftware code by its nature is susceptible to division and combination in a way that physical products are not." . . . But this to me is an argument for closer, rather than more relaxed, scrutiny of Microsoft's claims of integration. An operating-system designer who wished to turn two products into one could easily commingle the code of two formerly separate products, arranging it so that "Windows 95 without IE's code will not boot," . . . so that Windows 95 without Internet Explorer would "represent a disabled version of Windows 95 . . . ." This is not to say that commingling of code is per se pernicious or even suspicious. Rather, the point is that commingling alone is not sufficient evidence of true integration; the courts must consider whether the resulting product confers benefits on the consumer that justify a product's bridging two formerly separate markets. (page citations to majority opinion omitted)
Judge Wald supported her argument by citing Jefferson Parish Hospital District No. 2 v. Hyde, 466 U.S. 2 (1984), where the Supreme Court rejected the hospital's argument that it could require patients to use certain anesthesiologists because their services were part of a "functionally integrated package of services." The hospital cited benefits from joining the anesthesiologists to other hospital services: 24-hour coverage, flexible scheduling, and facilitation of work routine, professional standards, and equipment maintenance. But these benefits could be obtained by promoting the benefits of the hospital's anesthesiologists and setting standards of compatibility. The Court held that the question of whether one or two products were involved did not depend on the functional relation between them, but on the character of the demand for each product. There being a separate demand for the services of anesthesiologists, there were two products, not one, and the hospital's requirement was a tie-in rather than a true integration.
So it is in the Microsoft case. There is a demand and a separate market for browsers from the market for operating systems. If there are any benefits or efficiencies from bundling the browser and the operating system, such benefits can be obtained, if consumers want them, by promoting the combined product. Hence, there is a tie-in and not an integration. That is a particularly appropriate result because, in the case of software, two products may be put together merely by adding a few lines of code to one.
That fact means that accepting the "facial plausibility" argument results in a complete exemption from the antitrust laws of Microsoft's incorporation of any feature or product into its operating system by adding just a few lines of code. This is a license for unrestricted monopolization. It is not to be expected that the courts will ultimately grant such a license.
Even without access to the full evidence in the government's hands, it is clear that the incorporation of the Internet Explorer into the monopoly Windows operating system is predatory and anticompetitive as are the restrictive contracts Microsoft demands. The record before the Court of Appeals on the consent decree litigation did not contain most of this material. The government will present that evidence in the Sherman Act case.
Microsoft contends that no remedy can be devised that will not do more harm than good. That argument is both premature and disingenous. Appropriate remedies will become more apparent as the factual situation is more fully developed. It is unthinkable, however, that a firm in blatant violation of the Sherman Act's prohibitions of monopolizing and making agreements in restraint of trade should escape the consequences on the basis of a specious argument about remedies. Where there is a clear violation, as there is here, there is an appropriate remedy. To say otherwise is to say that the Sherman Act is a straw man.
The essence of the government's case against Microsoft is simply this:
Microsoft has a monopoly of personal computer operating systems, maintains that monopoly with a pattern of interlocking exclusionary practices and agreements, and has repeatedly admitted its intention to do just that in its own internal documents. The conclusion that Microsoft is violating ''1 and 2 of the Sherman Act is inescapable.
If Microsoft is successful in this predation, and in its attacks on Sun's Java, Microsoft will retain its operating system monopoly and be well on the way to achieving its goals of becoming the only gateway to the Internet and spreading its power to other markets. This is what is at stake in the government's antitrust case against Microsoft.
This is not a unique case. Well-settled and economically sensible antitrust jurisprudence has consistently been used to halt predatory conduct employed to maintain an existing monopoly. Devotion to the principles of a free market requires support of the Justice Department's case. The result for consumers will be lower prices and many more competitive sources of innovation.