January 10, 2001
CERTIFICATE AS TO PARTIES, RULINGS,
To the best of amicus' knowledge, all information on these topics is the same as that provided by Microsoft in its Brief for Defendant-Appellant. Any corrections thereto will presumably be made by the Plaintiffs-Appellees.
CERTIFICATE OF COMPLIANCE
This brief complies with all style and length requirements, being set in 12-point type, with all margins at least one inch, and with those sections of its content requiring page numbers not exceeding 25 pages total, and is being filed in both ".pdf" and paper format.
The amicus, Carl Lundgren, is an individual who is filing pro se without the use of an expensive attorney. The amicus would have preferred to file as the business, Valmarpro Antitrust, but was barred by the Court's rules from doing so without an attorney. Valmarpro Antitrust is a sole proprietorship wholly owned by Carl Lundgren.
TABLE OF CONTENTS
TABLE OF AUTHORITIES
United States v. Airline Tariff Pub. Co., 836 F.Supp. 9 (D.D.C. 1993)
United States v. American Telephone and Telegraph Co., 552 F.Supp. 131 (D.C.D.C. 1982), aff'd, 103 S.Ct. 1240, 460 U.S. 1001, 75 L.Ed.2d 472
United States v. Associated Milk Producers, Inc., 394 F.Supp. 29, aff'd, 534 F.2d 113 (8th Cir.), cert. denied sub nom. National Farmers' Organization, Inc. v. United States, 429 U.S. 940 (1976)
United States v. Bechtel Corp., 648 F.2d 660 (9th Cir.), cert. denied, 454 U.S. 1083 (1981)
United States v. E. I. du Pont de Nemours & Co., 366 U.S. 316 (1961)
United States v. Gillette Co., 406 F.Supp. 713 (D.C.Mass. 1975)
United States v. Glaxo Group Ltd., 410 U.S. 52 (1973)
United States v. Grinnell Corp., 384 U.S. 563 (1966)
United States v. GTE Corp., 603 F.Supp. 730 (D.C.D.C. 1984)
United States v. Microsoft Corp., 56 F.3d 1448 (D.C.Cir. 1995)
United States v. Microsoft Corp., 97 F.Supp. 2d 59 (D.D.C. 2000)
United States v. United Shoe Machinery Corp., 110 F.Supp. 295 (D.Mass. 1953), aff'd, 347 U.S. 521 (1954)
United States v. United Shoe Machinery Corp., 391 U.S. 244 (1968)
United States v. United States Gypsum Co., 340 U.S. 76 (1950)
United States v. Visa U.S.A., Inc. (S.D.N.Y., Order dated August 17, 2000), published citation not yet known to amicus, downloaded from www.usdoj.gov/atr/cases/f6200/6214.pdf
United States v. Western Electric Co., Inc., 767 F.Supp. 308 (D.D.C. 1991), aff'd 993 F.2d 1572, 301 U.S.App.D.C. 268, cert. denied, 114 S.Ct. 487, 510 U.S. 984, 126 L.Ed.2d 438
15 U.S.C. § 16 [a.k.a. "Tunney Act"]
28 U.S.C. § 24 [a.k.a. Federal Rules of Civil Procedure, Rule 24]
Association for Competitive Technology, "Proposal of the Association for Competitive Technology for Establishment of an Orderly Procedure for Public Participation on the Issue of Remedies," United States v. Microsoft Corp., filed with District Court, May 16, 2000 [cited as "ACT Proposal"]
Mike France, et al, "Does a Breakup Make Sense?" Business Week, November 22, 1999, pp. 38-41
James V. Grimaldi, "Reluctant Ruling for Judge," Washington Post, June 8, 2000, p. A01
Robert E. Litan, Roger G. Noll, William D. Nordhaus, Frederic Scherer, "Remedies Brief of Amici Curiae," United States v. Microsoft Corp., filed with District Court, April 27, 2000 [cited as "Litan Brief"]
Carl Lundgren, "Using Relative Profit Incentives to Prevent Collusion," Review of Industrial Organization, Vol. 11, No. 4, August 1996, pp. 533-550
Carl Lundgren, "An Open Letter to All Parties in the Microsoft Antitrust Mediation Process," December 2, 1999
Carl Lundgren, "Request to File a Friend of the Court Brief on Behalf of Valmarpro Antitrust," United States v. Microsoft Corp., filed with District Court, April 25, 2000 [filing refused]
Carl Lundgren, "Brief Amicus Curiae of Valmarpro Antitrust on Behalf of an Alternative Remedy for Microsoft Based on Using Relative Profit Incentives," United States Microsoft Corp., filed with District Court, April 25, 2000 [filing refused]
Carl Lundgren, "Motion by Valmarpro Antitrust to Intervene as an Independent Third Party," United States v. Microsoft Corp., filed with District Court, May 17, 2000
Office of Management and Budget, Economic Analysis of Federal Regulations Under Executive Order 12866, January 11, 1996, downloaded from http://www.whitehouse.gov/omb/inforeg/riaguide.html
Richard Posner, Antitrust Law: An Economic Perspective, University of Chicago Press (1976)
The President, William J. Clinton, "Executive Order 12866 of September 30, 1993-- Regulatory Planning and Review," Federal Register, Vol. 58, No. 190, Monday, October 4, 1993, pp. 51735-44
Lawrence Anthony Sullivan, Handbook of the Law of Antitrust, West Publishing Co. (1977)
U.S. Department of Justice and Federal Trade Commission, Antitrust Guidelines for the Licensing of Intellectual Property, April 6, 1995, downloaded from http://www.usdoj.gov/atr/public/guidelines/ipguide.htm
U.S. Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines, issued April 2, 1992, revised April 8, 1997, downloaded from http://www.ftc.gov/bc/docs/horizmer.htm
U.S. Department of Justice and Plaintiff States, "Plaintiffs' Memorandum in Support of Proposed Final Judgement", public redacted version, United States v. Microsoft Corp., filed with District Court, [April 28, 2000]
INTRODUCTION AND INTEREST OF THE AMICUS CURIAE
Carl Lundgren is the author and inventor of a new economic method for preventing collusion. This economic method is described in an article published in a refereed academic journal.(1) It is also the subject of a pending patent application. If Microsoft is split into two or more parts, this economic invention has possible application to Microsoft.
It is not the purpose of this amicus brief to prove to the Appeals Court the merits of this economic invention. Instead, this brief is dedicated to arguing that the District Court needed to test the government's proposed remedy, both for sensibleness and for adequacy, and needed to consider alternative remedies. For this purpose, the District Court needed to establish remedy proceedings and needed to admit third parties to such remedy proceedings.
This brief will not argue for or against the District Court's findings of fact and conclusions of law. Rather, where needed, this brief will simply assume the truth of such facts and conclusions for purposes of fashioning a remedy. Naturally, if Microsoft's conviction is overturned, in whole or in part, the remedy may need to be scaled back or eliminated entirely.
Microsoft has been convicted of illegally maintaining a monopoly in computer operating systems and illegally acquiring a monopoly in Internet browsers. These are serious antitrust violations. The government plaintiffs proposed what appeared to be an equally serious remedy: Divide Microsoft into two companies.
However, as serious as this remedy may appear, it is not obviously a good remedy, nor is it obviously an adequate remedy. The remedy selected by the government plaintiffs would simply split one monopoly into two monopolies, and then hope that the two monopolies decide to compete with each other. As a treatment for monopoly power, it is suspiciously inadequate. The government has not shown, either by way of public hearings and published regulatory analysis, or by way of evidence presented to the District Court, that this two-monopolies remedy is in the public interest.
The need and expectation for remedy proceedings, to ascertain the best remedy, seems to have been held by most of the parties and amici. One may speculate on why the District Court did not hold remedy hearings of any substance. One hypothesis is that the District Court was uncertain whether its findings and conclusions (which Microsoft vociferously objected to) would be upheld on appeal. Rather than spend time immediately on remedy hearings, the District Court may have reasoned that it was better to find out what findings and conclusions would survive appeal, and then spend time later to fashion the most appropriate remedy. As plausible as this hypothesis may be, it does not fully square with the District Court's words, which proclaimed itself to be incompetent in economic matters, and which claimed that testimony concerning future events is virtually useless, thereby implying that no remedy hearings of any substance would ever be provided.
Possible remedies for Microsoft have been much discussed in the press, and in various formal submissions to the District Court. Press accounts include an article in Business Week, which briefly discusses pros and cons of numerous options.(2) Filings with the District Court include the amicus brief filed by Robert Litan, on behalf of himself and three other economists.(3) The Litan Brief discusses in some detail a multitude of options, before setting forth their own preferred remedy, which includes dividing Microsoft into two companies along the lines suggested by Plaintiffs, and then further sub-dividing the operating system company into three competing parts. These two references alone suggest a multiplicity of possible remedies for Microsoft, which will require more than two parties to elucidate fully.
The public interest is like a patient who cannot speak for himself. Imagine a patient who is a child, comatose, senile, or insane. Suppose further that the patient clearly needs medical treatment, but the patient's Guardian (defendant) is adamantly opposed to medical treatment of any kind. Doctor A (the plaintiff) petitions the Court to order treatment. However, the patient's condition is complex, and different treatments are possible. Doctors B, C, and D petition the Court to intervene as additional parties, in order to explain alternative treatments. If the Court is truly concerned with the welfare of the patient, the Court will admit Doctors B, C, and D as additional parties. Such additional parties would have full authority to present their own evidence and testimony, not simply amicus briefs. To do otherwise would mean limiting the evidence solely to the medical views of Doctor A and the nonmedical views of the Guardian. This would effectively exclude the medical views of Doctors B, C, and D, even if their medical treatments would be better.
In this analogy, Doctor A is the Department of Justice ("DOJ"), and Microsoft is the Guardian. The Guardian was supposed to further the public interest, by pursuing a profit within the confines of the law. However, the Guardian now stands convicted of ignoring and violating the antitrust laws, and thereby harming the public interest. The Plaintiff offers economic treatment (Split the Guardian in two!), which the Court adopts. However, the Court does not seriously consider the opinions of additional Doctors of Economics, nor does it permit the advocates of alternative economic treatments to intervene as additional parties to present their own evidence and testimony. Clearly, under these circumstances the public interest must suffer, just as the patient's interests would suffer in the analogous example.
Unlike the patient, the public is composed of many persons, each of whom has different goals, opportunities, and interests. For purposes of analytical convenience, it is useful to divide and group the public into several categories, depending on the similarity or distinctiveness of their respective interests. Ascertaining the public interest requires an assessment of the costs and benefits of each remedy proposal to each distinctive group, as well as a consideration of the equity of the distribution of those costs and benefits.
This undertaking is sufficiently complex, that Congress does not entrust the DOJ to complete this task successfully without public input. Such public input is necessary for two reasons. First, the DOJ may not have adequate information concerning the impacts of a proposed remedy upon particular types of consumers or producers. Second, the DOJ in its economic analysis may be operating from limited perspectives, or other biases. Hence, with respect to every proposed consent decree in antitrust cases, the DOJ is required to publish a "competitive impact statement", and to provide a sixty-day window in which the public can present comments. 15 U.S.C. § 16(b,d). The DOJ is then obligated to respond, in writing, to all significant comments. 15 U.S.C. § 16(b,d). This competitive impact statement, the public comments, and the DOJ's responses are presented to the District Judge for review, before a consent decree can be approved. 15 U.S.C. § 16(b,d,e,f).
If the Plaintiffs had undertaken a consent decree with Microsoft, all third parties would have the right to present comments (including evidence and affidavits), and would have the right to expect a written response to their comments from the DOJ, as well as a review of their comments by the District Court. This would include members of the public who may have a direct or indirect financial stake, as well as interested individuals with expertise who have no significant financial stake. In a trial on the merits, we would expect all third parties to have the same or superior rights.(4) For the District Court arbitrarily to disregard the rights of any third party to comment upon or influence the course of remedy is a serious abuse of discretion, which contracts the public's rights and impedes a determination of the public interest.
According to the District Court's reasoning, remedy hearings are both useless and unnecessary. Useless, because the Court has no economic expertise. Unnecessary, because the DOJ (supposedly) has all those top-notch economists at its disposal. The District Court sends up a white flag: DOJ has all those economists, the Court has none, so let the DOJ be the de facto regulator of Microsoft. Since the District Court is still the de jure regulator, the Court's only further contribution is to rubber-stamp the DOJ remedy proposal.
This places the DOJ into a regulatory dilemma. If the DOJ is the de facto regulator of Microsoft, the DOJ might as well be the de jure regulator of Microsoft. If so, then the DOJ is potentially subject to Executive Order 12866. Among other things, this Order identifies twelve principles of regulation, including:
(3) Each agency shall identify and assess available alternatives to direct regulation, including providing economic incentives to encourage the desired behavior,...
If the DOJ were subject to this Order, the regulation of Microsoft would clearly be a "significant regulatory action", since it has an annual effect on the economy of more than $100 million. In addition, the DOJ would be obligated to "provide the public with meaningful participation in the regulatory process." The DOJ did none of this. There is no published regulatory analysis of the costs and benefits of the proposed regulation of Microsoft, nor is there any published economic analysis of the various regulatory alternatives. Moreover, the whole process of selecting a remedy for Microsoft was performed in secret; the public was provided no meaningful participation.
The District Court remains the de jure regulator who must approve any remedy imposed on Microsoft. Even when faced with a consent decree, the antitrust Court has no authority to act like a rubber stamp.(6) In the absence of any public record of remedy hearings, whether by the DOJ or by the Federal Trade Commission ("FTC"), which the Court can rely upon and possibly supplement, the Court cannot credibly evade its own responsibility to select a best remedy. This means that the Court must set up sensible procedures which test the government's proposed remedy, and which consider and weigh various alternative remedies. The Court must develop a public record on its own initiative. As Dr. Litan argued:
We begin by underscoring the need for a careful and independent judicial inquiry into the nature of the appropriate remedy. Indeed, as the Supreme Court has highlighted, the development of an appropriate remedy is "the most significant phase of the [antitrust] case." United States v. Glaxo Group Ltd., 410 U.S. 52, 64 (1973) (emphasis added). (Litan Brief, p. 9)
* * * * * * * * * * *
Our review of the relevant literature indicates that the relief phase of antitrust cases is often treated as an afterthought, even in cases as important as monopolization findings under Section 2 of the Sherman Act. For example, Professor Lawrence Sullivan has observed: "Perhaps the best hope is that, hereafter, Courts facing structural remedy issues will get more help than they have customarily received from the Department of Justice. As Judge Wyzanski implied in United Shoe Machinery, the government is sometimes extremely casual about remedy."(7) A similar point of view has been voiced by Chief Judge Richard Posner (who attempted to mediate a settlement in this case): "Another reason for the poor record of divestiture as an antitrust remedy is that the government's lawyers tend to lose interest in a case at the relief stage. They derive both personal satisfaction and career advancement from the trial of an antitrust case, but gain neither from the post-trial relief negotiations and proceedings, which they frequently tend to pay scant attention."(8) (Litan Brief, p. 11)
* * * * * * * * * * *
The Court should recognize that it has broad discretion to fashion relief. It is the Court's job to accomplish the objectives set out by the Supreme Court for addressing the ill effects of monopolization. This simple statement leads to a critical conclusion, however: the remedy should not be limited to the relief that the government plaintiffs may propose. The Court's role in fashioning a remedy for proven antitrust violations is analogous to but much more extensive than its much more limited role in reviewing proposed antitrust consent decrees.(9)
The Court has ample authority under 28 U.S.C. § 24(a,b) to admit new parties and amicus briefs as needed to develop the evidence, cross-examination, and briefing needed to find the best remedies. Motions for full intervention should be granted, where the parties intend to propose distinctive remedies which require additional evidence, testimony, or cross examination to support. Motions for limited intervention (amicus status) should be granted where distinctive perspectives do not require additional evidence, perhaps because the supported remedy is similar to the remedy proposals of other parties. While the Court cannot directly force the DOJ to pay adequate attention to its remedy, the Court can indirectly provide incentive by embarrassing the DOJ if it fails to pay the required attention.
The Court's rules of procedure should balance both speed and quality of decision making in the public interest. Microsoft does not have a "due process" right to impede the remedy hearings; its testimony is relevant only insofar as ascertaining the public interest requires assessing the impacts upon Microsoft and others. Six months is probably adequate to determine whether and what type of divestiture of Microsoft is most appropriate.
In rejecting both my amicus brief and my motion for intervention, the District Court gave no reasons.(10) Perhaps a sense of possible reasons can be gleaned from a ruling in a different District Court, in a different antitrust case. In a recent antitrust case involving Visa Card and MasterCard, Discover Card sought intervention in the remedies phase of trial to protect both its own interests and the public interest. The District Court reuffed Discover's motion, saying among other things:
While Discover undoubtedly has an interest relating to the subject matter of the action, in government antitrust actions, courts have uniformly recognized that the government represents the public interest in competition, unless a private party makes an extraordinary showing to the contrary. See United States v. Bechtel Corp., 648 F.2d 660, 666 (9th Cir.), cert. denied, 454 U.S. 1083 (1981); United States v. Associated Milk Producers, Inc., 394 F.Supp. 29, aff'd, 534 F.2d 113, 117-18 (8th Cir.), cert. denied sub nom. National Farmers' Organization, Inc. v. United States, 429 U.S. 940 (1976).(11)
The trouble with this dictum is that it implies too much. Saying that the government represents every member of the public is similar to saying that a single creditor represents the interests of all creditors in a bankruptcy proceeding. Though the government may be obligated to pursue the public interest (suitably defined), it does not define the public interest, nor is it the only entity which can represent the public interest (however fully or partially), nor does an uncorrupt government always know and pursue the public interest. Both cases cited in support of this overly broad proposition involve consent decrees, not trials on the merits. In consent decrees, the government's decisions are entitled to more deference (in particular, the agreed-to remedy need not be optimal), otherwise neither party would have much incentive to settle out of court. Such considerations do not apply to a trial on the merits, where the very purpose of the remedy hearing is to ascertain the public interest, based on evidence and testimony.
Moreover, none of the cited cases stands for the proposition that third parties should be arbitrarily excluded from remedy hearings. The first case cited involves an attempt by Bechtel Corporation to withdraw consent from its own consent decree; no issue of third-party intervention is present. 648 F.2d at 663. The second case cited involves a denial of third- party intervention, but only based on good causes shown by the Court. 534 F.2d at 118. Finally, even the Court quoted above did not entirely exclude Discover from the proceedings. Discover had supplied witnesses to the proceeding, had presumably been consulted by the government, and was permitted to file an amicus brief on its preferred remedy.
The District Court argues against the desirability of extended remedy hearings in the following terms:
Finally, the Court believes that extended proceedings on the form a remedy should take are unlikely to give any significantly greater assurance that it will be able to identify what might be generally regarded as an optimum remedy.... In its experience the Court has found testimonial predictions of future events generally less reliable even than testimony as to historical fact, and cross-examination to be of little use in enhancing or detracting from their accuracy.
The District Court is unduly pessimistic about the ability of further proceedings to identify optimal remedies. It is not relevant that predictions of future events are less accurate than testimony concerning past events. What the Court needs in order to determine best remedies is not actual predictions of future events, but predictions of the form, "Action A will be better or worse than Action B, for the following reasons...." For example, specific predictions of the actual price of software under monopoly or under competition are not needed in order to determine that the price of software will be higher under monopoly than under competition. That prices are higher under monopoly than under competition is a basic proposition or theorem of economics, which admits of relatively few exceptions. It is also the basic principle which underlies the antitrust laws. Numerous more complicated predictions can be reasoned out using both theoretical and empirical economic techniques
The Washington Post quotes Judge Jackson as saying: "It's important you understand what my function is here. I am not an economist. I do not have the resources of economic research or any significant ability to be able to craft a remedy of my own devising." ("Reluctant Ruling for Judge," James V. Grimaldi, Washington Post, June 8, 2000, p. A01) The Post concluded, "As a result, Jackson relied heavily on the work of the government." It is not essential that the Court retain this self-imposed limitation on its expertise.
One way the District Court can expand its economic expertise would be to hold remedy hearings before an expert jury of economists, who would act as special masters to the Court. Such an expert jury can be selected at random from lists of the American Economic Association and other sources. The jurors can be paid a reasonable compensation, and would be strictly volunteers, who would read any affidavits and transcripts in private, then meet for discussions. Each expert juror would provide, individually, best recommendations to the Court, and would not be required to arrive at a unanimous conclusion. The Court could then decide, based on the various recommendations, what type of remedy to adopt. Several experts selected at random may be better than a single Court Master, particularly if there is lacking any agreement on which economists may be impartial.
The antitrust court has remedial power to order many things. Antitrust courts have ordered the divestiture of assets, and even the divestiture of whole companies. Antitrust courts have ordered the licensing, for free or for a fee set by the Court, of intellectual property in the form of patents. Antitrust courts have imposed various conduct remedies, both to forbid unwanted conduct and to require wanted conduct. The Court has power to threaten imprisonment of any corporate officer who refuses its edicts. Presumably, the Court has power to replace management personnel, much as a Court might take over or put into receivership a recalcitrant or corrupt labor union, school, or prison. However, the Court must not act in an arbitrary manner; standards of remedy are needed.
An important goal of the antitrust remedy should be to prevent or deter future antitrust violations, and to deprive violators of the fruits of their illegal actions. Dr. Litan explains:
Fortunately, the Supreme Court also has provided guidance on several occasions for evaluating relief in monopolization cases, announcing that the remedy should terminate the illegal monopoly; prevent practices likely to result in monopolization in the future; and deny to the defendant the fruits of its statutory violation. See, e.g., United States v. United Shoe Machinery Corporation, 391 U.S. 244, 255 (1968). Or, as the Court spoke in United States v. Grinnell Corp., 384 U.S. 563, 577 (1966), relief in a Sherman Act case "should put an end to the combination and deprive the defendants of any of the benefits of the illegal conduct, and break up or render impotent the monopoly power found to be in violation of the Act." (Emphasis added).(12) (Litan Brief, p. 9; see also pp. 15, 24, 27-28)
Likewise, as both the Plaintiffs and the District Court recognized:
The proposed final judgment is represented to the Court as incorporating provisions employed successfully in the past, and it appears to the Court to address all the principal objectives of relief in such cases, namely, to terminate the unlawful conduct, to prevent its repetition in the future, and to revive competition in the relevant markets. (United States v. Microsoft Corp., 97 F.Supp. 2d 59, 63, D.D.C. 2000)
One constraint upon a forward-looking antitrust remedy (as opposed to a fine or payment of damages for past bad behavior) is that no current property of Microsoft may be taken without "just compensation." Naturally, this includes physical property, but it also includes arbitrary confiscation of intellectual property, such as patents, copyrights, or trade secrets.(13) Nevertheless, such patent, copyright, or trade secret monopolies (otherwise legal) can be non-arbitrarily denied further enforcement or permitted only partial enforcement, if they are determined to be part of a monopolization scheme in violation of the antitrust laws. Hence, prior courts have ordered the non-enforcement of patent monopolies, or required their licensing on terms set by the Court. (Litan Brief, p. 38)
Patents cover an inventor's "idea" for a practical application having utility, which may be embodied in material form or as a method (including computer programs). Copyrights, on the other hand, do not protect either ideas or functional utilities. Copyrights only protect against imitators who would simply copy works of art or literature (including computer programs). Copyrights do not protect against imitation of ideas or functional utilities (which may be embodied in computer programs), so long as the imitation is not done through copying. Patents and copyrights enjoy Constitutional and statutory protection.
Trade secrets do not enjoy Constitutional protection. Trade secrecy does not prevent imitation by any lawful means, for example, by literal copying, through reverse engineering, or otherwise. Trade secrecy simply prohibits theft of trade secrets by unlawful means, such as breaking and entering Microsoft headquarters, hacking into Microsoft computers, bribing Microsoft employees, and so forth. The public's interest in maintaining an orderly society argues against such disorderly conduct by private parties, and argues in favor of preventing disorderly persons from achieving the fruits (e.g., trade secrets) of their disorderly conduct. The same considerations do not prevent the government from ordering, in an orderly fashion, the divulgence or revelation of trade secrets.
Where trade secrets, or a general pattern of trade secrecy, is used to protect monopolies, there is no Constitutional impediment to obligating Microsoft to reveal such trade secrets, in whole or in part. Microsoft, in part, uses trade secrecy to acquire value from its intellectual innovations. Microsoft, in part, uses trade secrecy to maintain as proprietary information a Windows standard, upon which much of the computer world must rely. A mere "standard" is not an intellectual innovation of much value in itself. Standards exist both inside and outside the computer world, and most standards are public information, not proprietary information. The Court may properly order Microsoft to divulge some or all of its software "standards", without thereby confiscating Microsoft's "intellectual property."
Cost-benefit analysis is a set of techniques developed by economists over a period of decades for analyzing a variety of practical problems in a variety of contexts.(14) Benefits and costs can be calculated from a variety of perspectives, including that of a particular individual, a particular group, the nation as a whole, or even the whole world. Typically, the calculation is summarized by a Net Benefit figure (total benefits minus total costs). Where feasible, it is appropriate to express all benefits and costs in monetary units.
For purposes of selecting an antitrust remedy, it is important to ascertain which benefits and costs will be used to calculate the Net Benefit of any proposed antitrust remedy. The scope of the proceeding would seem to limit the accounting to only antitrust-related benefits and costs, rather than other types of benefits and costs (e.g., health, safety, or environment), unless such other benefits and costs are interactively related to the antitrust benefits and costs. Since it is not the purpose of the antitrust laws to redistribute income, gains and losses to rich people are weighted in the same manner as gains and losses to poor people. Likewise, gains and losses to producers are weighted in the same manner as gains and losses to consumers.
However, it is a premise of the antitrust laws that certain gains to monopoly sellers (or monopsony buyers) are illegitimate and unproductive. For purposes of calculating the legitimate costs and benefits to society, such illegitimate gains should be excluded from the net benefit calculation. Thus, when evaluating whether to approve a merger, the DOJ estimates whether the merger will result in a higher price or a lower price.(15) The price might rise due to monopoly power; the price might fall due to increased efficiency in production. Even if production efficiency would increase, if prices rise, the merger is not approved, even if producers would gain more than consumers lose. This is because the gain in monopoly profits for producers is viewed as illegitimate, and therefore is not counted as a legitimate benefit for society under the antitrust laws.
Many issues affecting remedy can be placed in a cost-benefit framework. For example, it has been much discussed whether Microsoft is "trustworthy." Indicators of trustworthiness can be such factors as whether Microsoft previously obeyed Court orders, whether Microsoft witnesses gave trustworthy testimony, whether Microsoft accepts and repents of its law violations, whether Microsoft's violations were a single act or a series of orchestrated acts, whether Microsoft could reasonably believe that its actions were lawful, and so forth. The issue of trustworthiness necessarily affects the cost-benefit calculation of conduct remedies vis-a-vis structural remedies. If Microsoft is trustworthy, then conduct remedies may require minimal monitoring and enforcement action by the DOJ, and the benefits fairly certain. On the other hand, if Microsoft is untrustworthy, monitoring and enforcement actions could prove to be quite costly, while the benefits would be uncertain and spotty. Indeed, if the cost of enforcement exceeded the normal budget of the DOJ, a conduct remedy would need to include a special set-up (at Microsoft's expense) of an independent entity to enforce the remedy provisions against Microsoft.
However, Microsoft's trustworthiness is not the sole factor affecting a choice between conduct remedies and structural remedies. Another major factor is the extent to which Microsoft is earning illegitimate monopoly profits as a result of prior violations. If the antitrust violation is a too-aggressive approach to the competitive threat posed by Netscape/Java, the illegitimate profits result from two factors: 1) Illegitimate profits resulting directly from the improper elimination of the Netscape/Java threat. 2) Illegitimate profits resulting indirectly, by developing a reputation for overly aggressive toughness, so as to deter future competitive threats which may arise from future innovation.
Neither set of illegitimate profits can be measured precisely, because neither the future nor the past can be re-run and measured under the counterfactual assumption that Microsoft committed no violations. However, estimates of the likelihood and extent of such profits can certainly be made. If these illegitimate profits are estimated to be large, then the Net Benefit of a conduct remedy is likely to be low relative to the Net Benefit of a structural remedy. Alternatively, if these illegitimate profits are estimated to be low or nonexistent, then the Net Benefit of a conduct remedy may be relatively higher, provided it is enforceable.
The government has the burden to prove, by a preponderance of evidence, that Microsoft violated the antitrust laws. After meeting that burden and obtaining a conviction, there is some presumption that the government's proposed remedy is correct. That presumption can be overcome, by a preponderance of evidence, that another remedy (or no remedy) would better serve the public interest. The government's remedy may be too much, too little, not the right type, or otherwise not the best. If the government's proposal goes too far, Microsoft can be counted on to argue against it. If the government's remedy goes not far enough, or is otherwise not the best, only a third party, or an actively involved, highly competent, and assertive Court, can supply the needed correction. The remedy Court, after reviewing the evidence and opinions offered by the Plaintiff, Defendant, and third parties, should rank all remedy proposals (including any of the Court's own devising) from best to worst. Only the best proposal(s) should be considered further. If any two or more proposals are tied for first place, and no further investigation can tip the scales, the government may choose from the first place winners.
The remedy plan devised by the government appears poorly designed to meet the important objective of eliminating Microsoft's monopoly power in the markets for operating systems, Office software, and Internet browsers. The government proposes to divide the tri-part Microsoft monopoly into two monopolies, and then hope the two smaller monopolies will eventually compete with each other. Indeed, one government witness limply suggests that the two monopolists might simply "foster competition" with each other.
The operative word here is "hope", not realistic expectation, that robust software competition will re-emerge in these markets. As the government acknowledges, the truly threatening competition to Microsoft, Netscape and Java, has essentially died away, and there is no definite prognosis that another truly threatening competitor will soon emerge. Even under an optimistic scenario, actual and serious competition will take years to develop, after the government's proposed break up occurs. Under a pessimistic scenario, competition may not develop at all. The pessimistic scenario is just as plausible as the optimistic scenario.
The government's proposal therefore permits Microsoft to enjoy for several years, perhaps even indefinitely, the fruits of Microsoft's illegal monopolization campaign. The benefits of this remedy are limited. Indeed, the benefits appear inadequate.
The potential costs of the remedy are two-fold. One is the problem of "double-marginalization". If two monopolists over complementary products separately charge their.20 respective profit-maximizing prices, the price will be higher and consumers worse off, than if the two monopolists colluded or a single monopolist set both prices. Adequate competition in one or both markets may be able to contain this problem.
The second problem is that arbitrarily preventing two different software products from linking together, except at "arms length", may forego the possibility of useful kinds of integration. This problem may be tolerable, if all important integrated software is easily developed at "arms length", or if there exists no commercially important potential development of integrated software. While the District Court has found, strictly as a factual matter, that past "integration" of Windows software with browser software was a sham, this says little about future possibilities of genuine integration having significant consumer value. In any case, the likelihood of future possibilities is a factual matter that should be explored in remedy hearings, if the Plaintiffs continue to support this remedy.
Arguably, under certain assumptions, both potential harms may be limited in nature, but they are a potential cost of a remedy proposal which offers limited benefits.
Plaintiffs' remedy rests too much on a narrow reading of the historical record. In the past, Microsoft used its monopoly operating system to obtain a monopoly in Internet browser software. Although not presented as evidence in this case, Plaintiffs believe that in the past Microsoft used its monopoly operating system to obtain a monopoly in Office software. Ergo, the operating system monopoly is being abused to form monopolies in various kinds of applications. Hence, the operating system monopoly is especially pernicious, and if cordoned off from abusive integration with applications, then all will be well.
However, the testimony offered by the Plaintiffs in the remedy stage of trial paints a somewhat different picture. That testimony suggests that both the Office software monopoly, and the Internet browser monopoly, can just as easily act as potential platforms for the extension of monopoly power into other applications. Moreover, Plaintiffs argue that the operating system monopoly for personal computers (most often used as "client computers" in a computer network) is being integrated (usefully and/or abusively) to help Microsoft obtain a monopoly over "server computer" operating systems. In other words, dividing Microsoft in two along the lines suggested by Plaintiffs, merely slows down, but does not really defeat Microsoft's ability to extend its monopolies into additional software sectors.
There exists a different, broader, and more general way of looking at Plaintiffs' evidence. One of the underlying causes of Microsoft's successful predatory behavior is the extensive use of trade secrecy as an anti-competitive weapon. At the retail level, Microsoft code is essentially compiled code, whose meaning is virtually incomprehensible to humans. Human programmers normally program in higher-level, "source-code" computer languages, which are then translated by machine into compiled code. Microsoft has amassed a very large structure of interlocking machine code, and used trade secrecy with respect to source code as an anti-competitive weapon both to maintain and to expand its monopolies.
Looked at more generally, a monopoly in any "must-have" software program, be it Office software, browsers, or operating systems, can be abusively integrated into any other software category with respect to which a monopoly is also desired. The extension of monopoly power can just as easily move from application to operating system, as from operating system to application. It is merely historical happenstance that Microsoft moved from operating systems to applications; the movement could just as easily be the reverse, as Plaintiffs hypothesize was the competitive threat posed to Microsoft by Netscape/Java.
A different diagnosis leads to different suggestions for remedy. Two suggestions are immediately apparent: 1) Break up Microsoft into two or three companies that compete in each of Microsoft's lines of business, thereby breaking up all of Microsoft's monopolies simultaneously. 2) Require Microsoft to give up its trade secrecy in source code, which is the basic weapon in Microsoft's anticompetitive strategy.
The two suggestions just mentioned may be called "conventional" because they do not require any invention of new knowledge or techniques, and were already publicly suggested months before Plaintiffs finally agreed upon and proposed a remedy.
With respect to the first suggestion, none of Plaintiffs' three economist witnesses (Drs. Henderson, Romer, Shapiro) addresses the issue of whether or not it would be better simply to break up Microsoft within all its lines of business, so as to introduce competition into Microsoft's businesses directly and immediately. Plaintiffs' "reasons" for rejecting obvious solutions consist of a mere two sentences in their remedies brief, unsupported by any other evidence or testimony.(16) Plaintiffs' three economists basically state only that the government plan is a good plan; they do not testify that it is the best plan.
The second "conventional" remedy suggestion is somewhat peculiar to the software industry itself, but nonetheless is applicable to Microsoft. The "open-source" movement, most notably among Linux developers, has two premises: 1) The source code for all open- source compiled code should be publicly available. 2) The open-source code may be taken and used without pay, and any programs using that source code will likewise be subject to the open-source provisions. The first provision makes a great deal of sense, and is a potential solution for preventing Microsoft from using trade secrecy as an anticompetitive weapon. The second provision is a requirement of charity, and would likely not be imposed by an antitrust court upon the for-profit Microsoft. Instead, Microsoft should retain its copyrights and patent rights to both source code and compiled code, while being obligated to license such code on reasonable terms. Such licensing should allow fractional portions of code to be purchased at reduced rates from the purchase of whole code.
Both of these conventional remedies would bring about competition directly, and with more certainty, than the Plaintiffs' two-monopolies proposal. In addition, both of these conventional remedies would be consistent with allowing the development of integrated products, which might include aspects of what we currently call "operating systems" and "applications." Hence, both of these alternatives would be associated with greater benefits, and fewer harms, than the Plaintiffs' current remedy proposal.
Finally, there is a third pro-competitive action which the Plaintiffs could have advocated, but did not advocate. This might be called an "unconventional" remedy, since it rests upon new economic knowledge or invention. This unconventional remedy would use relative profit maximizing ("RPM") incentives. RPM incentives can be implemented by rearranging the compensation of business managers so that the business firms they manage are placed into a "zero-sum game." This alternative incentive scheme for business managers can reduce or eliminate incentives for collusion and predatory behavior, even when there are as few as two business firms in the same market.
This incentive scheme can be implemented in more that one manner. In December 1999, I wrote a letter to all parties in the unsuccessful Microsoft mediation process before Judge Posner.(17) In that letter I suggested a proposal to divide Microsoft into two equal-sized companies in each of its lines of business, and then placing these two firms into a zero-sum game with each other using RPM incentives.
That initial proposal is not the only possible application of RPM incentives to Microsoft. One can use RPM incentives with respect to any two or more firms. In particular, even if we assume that the government's two-monopolies proposal is the best of the conventional remedies, RPM incentives can make the government's remedy even better. The government hopes that eventually these two monopolies will start competing with each other. RPM incentives can help assure this competition, and avoid tacit collusion.
With respect to amicus' original proposal, or other possible uses of RPM incentives, neither the District Court, nor the DOJ, has shown the slightest interest or curiosity. Neither the DOJ nor any of DOJ's economists or attorneys has called or written to ask questions, nor invited the amicus to any interviews, seminars, or discussions concerning the invention, or its possible application to Microsoft or any other antitrust case, past, present, or future, whether actual, pending, or hypothetical. For its part, the District Court denied the timely filing of an amicus brief, and also denied amicus' timely motion to intervene.
The District Court, perhaps unwittingly, abused its discretion by failing to hold
substantive remedy hearings that would admit third parties. This deprived the public of
substantive procedural rights, which prevents proper evaluation of the public interest. Moreover,
the District Court adopted too strong a presumption in favor of the government's proposed
remedy, and rashly adopted a government remedy proposal that is both inadequate and
inadequately explained. Institution of remedy proceedings that admit third parties will provide
competition within the Court room to advocate best remedies. Such competition will promote
innovation, quality, and efficiency in antitrust remedies, laudable goals which the DOJ advocates
for Microsoft, the computer industry, and the private sector generally. The Appeals Court should
provide the District Court with further instruction on these matters.
January 10, 2001.
CERTIFICATE OF SERVICE
Service of the foregoing Brief Amicus Curiae of Carl Lundgren in Support of Remedy Proceedings in District Court That Admit Third Parties was made by sending two copies each, in a sealed envelope, postage fully prepaid, addressed to:
Bradley P. Smith (by hand only)
John L. Warden
William H. Neukom, Sr. Vice President
A. Douglas Melamed
Catherine G. O'Sullivan (by hand only)
Phillip R. Malone
Richard L. Schwartz
Kevin J. O'Connor
Lars H. Liebeler
Donald Manwell Falk
Paul T. Cappuccio
Lee A. Hollaar
Robert H. Bork
Edward J. Black
Laura Bennett Peterson
Robert S. Getman
David R. Burton, CMDC
Louis R. Cohen
and deposited in the United States Mail by the undersigned this 10th day of January, 2001.
1. Carl Lundgren, "Using Relative Profit Incentives to Prevent Collusion," Review of Industrial Organization, Vol. 11, No. 4, August 1996, pp. 533-550.
2. Mike France, et al, "Does a Breakup Make Sense?" Business Week, November 22, 1999, pp. 38-41
3. The four economists on the Litan Brief were: Robert E. Litan, Roger G. Noll, William D. Nordhaus, and Frederic Scherer.
4. A similar point was made by Microsoft's ally, ACT: "Accordingly, ACT respectfully suggests that the Court establish an orderly mechanism by which it and other affected members of the public can submit further comments and evidence on the likely impacts of the plaintiffs's proposals." (ACT Proposal, end of § I)
5. The President, "Executive Order 12866 of September 30, 1993--Regulatory Planning and Review," Federal Register, Vol. 58, No. 190, October 4, 1993, 51735, 51735-36.
6. 15 U.S.C. § 16(e,f); U.S. v. Microsoft Corp., 56 F.3d 1448, 1458, 1462 (D.C.Cir. 1995); U.S.v. Airline Tariff Pub. Co., 836 F.Supp. 9, 11 (D.D.C. 1993); U.S. v. Gillette Co., 406 F.Supp. 713, 715 (D.C.Mass. 1975); U.S. v. Western Elec. Co., Inc., 767 F.Supp. 308, 328-329 (D.D.C. 1991), aff'd 993 F.2d 1572, 301 U.S.App.D.C. 268, cert. denied, 114 S.Ct. 487, 510 U.S. 984, 126 L.Ed.2d 438; U.S. v. GTE Corp., 603 F.Supp. 730, 742[n. 42] (D.C.D.C. 1984); U.S. v. American Tel. and Tel. Co., 552 F.Supp. 131, 147-153 (D.C.D.C. 1982), aff'd 103 S.Ct. 1240, 460 U.S. 1001, 75 L.Ed.2d 472.
7. [Litan Brief, note 3] Lawrence Anthony Sullivan, Handbook of the Law of Antitrust, West, 1977, at 146.
8. [Litan Brief, note 4] Richard Posner, Antitrust: An Economic Perspective, University of Chicago Press, 1978 [sic, 1976], at [88-]89.
9. [Litan Brief, note 5] This proceeding differs from that arising from the 1994 consent decree that Microsoft entered into with the Justice Department. In that matter, the D.C. Circuit held that Judge Sporkin exceeded his authority under the Tunney Act, which applies only to consent decrees and not to final judgments, because Judge Sporkin wanted the government to pursue a broader case than the one it actually filed.
10. On April 25, 2000 Carl Lundgren filed a "Request to File a Friend of the Court Brief on Behalf of Valmarpro Antitrust" along with a "Brief Amicus Curiae of Valmarpro Antitrust on Behalf of an Alternative Remedy for Microsoft Based on Using Relative Profit Incentives". On May 17, 2000 Carl Lundgren filed a "Motion by Valmarpro Antitrust to Intervene as an Independent Third Party".
11. United States v. Visa U.S.A. (S.D.N.Y., August 17, 2000).
12. [Litan Brief, note 2] See also United States v. United States Gypsum Co., 340 U.S. 76, 88(1950) [a decree must "pry open to competition a market that has been closed by defendants' illegal restraints."]; and United States v. E. I. du Pont de Nemours & Co., 366 U.S. 316, 326 (1961).
13. For a more detailed overview, see Antitrust Guidelines for the Licensing of Intellectual Property, US DOJ & FTC, April 6, 1995.
14. A document which describes the current state of benefit-cost analysis is Economic Analysis of Federal Regulations Under Executive Order 12866, OMB, January 11, 1996.
15. Benefits and costs other than price are also considered. See Horizontal Merger Guidelines, US DOJ and FTC, issued April 2, 1992, revised April 8, 1997.
16. Plaintiffs' purported reasons are: " There could be efficiency losses because the division of all operating system personnel among multiple companies might disrupt engineering units and could result in unnecessary duplication;  in addition, particularly in an industry characterized by network effects, rivalry among identical products tends to offer less long-run competition and consumer choice than among differentiated products."  [The government proposal]... "does not raise the efficiency concerns that a division of Microsoft's operating business might entail." ("Plaintiffs' Memorandum in Support of Proposed Final Judgement", April 28, 2000, pp. 8, 9, public redacted version, bracketed numbers added.)
17. Carl Lundgren, "An Open Letter to All Parties in the Microsoft Antitrust Mediation Process," December 2, 1999.