|This document is available in two formats: this web page (for browsing content), and PDF (comparable to original document formatting). To view the PDF you will need Acrobat Reader, which may be downloaded from the Adobe site. For an official signed copy, please contact the Antitrust Documents Group.|
IN THE UNITED STATES DISTRICT COURT
MEMORANDUM OF THE UNITED STATES OF AMERICA IN SUPPORT OF
MOTION TO ENTER FINAL JUDGMENT AND IN OPPOSITION TO
THE POSITIONS OF I.D.E. CORPORATION AND AMICI
TABLE OF CONTENTS
TABLE OF AUTHORITIES CASES
Apache Survival Coalition v. United States, 21 F.3d 895 (9th Cir. 1994)
Baltimore Gas & Electric Co. v. NRDC, 462 U.S. 87 (1983)
Blitz v. Donovan, 740 F.2d 1241 (D.C. Cir. 1984)
Brunswick Corp. v. Pueblo Bowl-O-Mart, Inc., 429 U.S. 477 (1977)
Cargill, Inc. v. Monfort of Colorado, Inc., 479 U.S. 104 (1986)
Communication Workers of America v. Beck, 487 U.S. 735 (1988)
Confiscation Cases, 74 U.S. (7 Wall.) 454 (1869)
Edward J. DeBartolo Corp. v. Florida Gulf Coast Building & Construction Trades Council, 485 U.S. 568 (1988)
Hartford-Empire Co. v. United States, 323 U.S. 386 (1945)
*Heckler v. Chaney, 470 U.S. 821 (1985)
In re International Business Machines Corp., 687 F.2d 591 (2d Cir. 1982)
*National Society of Professional Engineers v. United States, 435 U.S. 679 (1978)
United States v. Airline Tariff Publishing Co., 1993-1 Trade Cas. (CCH) ¶ 70,191 (D.D.C. Mar. 8, 1993)
United States v. Airline Tariff Publishing Co., 836 F. Supp. 9 (D.D.C. 1993)
United States v. American Cyanmid Co., 719 F.2d 558 (2d Cir. 1983), cert. denied, 465 U.S. 1101 (1984)
United States v. American Telephone & Telegraph Co., 552 F. Supp. 131 (D.D.C. 1982), aff'd sub nom. Maryland v. United States, 460 U.S. 1001 (1983)
United States v. BNS Inc., 858 F.2d 456 (9th Cir. 1988)
United States v. Cox, 342 F.2d 167 (5th Cir.), cert. denied, 381 U.S. 935 (1965)
United States v. Gillette Co., 406 F. Supp. 713 (D. Mass. 1975)
United States v. Nixon, 418 U.S. 683 (1974)
*United States v. Western Electric Co. (Triennial Review Remand), 993 F.2d 1572 (D.C. Cir.), cert. denied, 114 S. Ct. 487 (1993)
United States v. Western Electric Co. (Triennial Review Opinion), 900 F.2d 283 (D.C. Cir. 1990)STATUTES
Sherman Act, 15 U.S.C. §§ 1-2
*Tunney Act, 15 U.S.C. § 16(b)-(h)
The Antitrust Procedures and Penalties Act: Hearings on S. 782 and S. 1088 Before the Subcomm. on Antitrust and Monopolies of the Senate Comm. on the Judiciary, 93d Cong., 1st Sess. (1973)
119 Cong. Rec. S13927 (daily ed. July 18, 1973)
119 Cong. Rec. S13930 (daily ed. July 18, 1973)
120 Cong. Rec. H10765 (daily ed. Nov. 19, 1974)
59 Fed. Reg. 42,849 (1994)
*H.R. Rep. No. 1463, 93d Cong., 2d Sess. (1974)
Michael Morris, Microsoft Deal: Too Little Too Late, S.F. Examiner, July 24, 1994, at C-5
*S. Rep. No. 280, 93d Cong., 1st Sess. (1973)
Transcript of Press Conference (July 16, 1994)
IN THE UNITED STATES DISTRICT COURT
MEMORANDUM OF THE UNITED STATES OF AMERICA IN SUPPORT OF
MOTION TO ENTER FINAL JUDGMENT AND IN OPPOSITION TO
THE POSITIONS OF I.D.E. CORPORATION AND AMICI
This memorandum, and the accompanying declaration of Nobel Laureate Kenneth J. Arrow (Exhibit 1) [hereinafter Arrow Dec'l], constitute the opposition of the United States to the position taken by the amici. This memorandum explains how amici's submission misstates the applicable Tunney Act standards. The Arrow declaration, as well as this memorandum, further explain how amici's submission is based on an erroneous view of the effect and scope of Microsoft's antitrust violations and the proposed decree. This memorandum also addresses the position of I.D.E. Corporation.
The case that the government brought against Microsoft challenged specific practices that limit the opportunities for Microsoft's competitors to enter the operating system market and compete successfully. The remedy that the parties agreed to would forbid all those practices, and more. By eliminating these unreasonable restraints, the remedy will provide important benefits to the public by eliminating artificial barriers to entry that Microsoft imposed to prevent competition in the operating system market. In large part, the full value of benefits has not yet been realized, simply because increased competition will result from successful entry into a monopolized market, and successful entry into this market takes time. That does not justify rejection of the proposed Consent Decree or further delay; it means, instead, that the proposed Consent Decree should be approved without further delay.
The amici point to the six-fold growth in Microsoft's installed base and argue that through its illegal licensing practices, Microsoft acquired market power that the proposed Consent Decree will not take away. They oppose the decree because they claim Microsoft will use this market power anticompetitively in applications software markets, concededly outside the operating system market. Their opposition is wrong for three reasons.
First, the premise of their argument is incorrect. As Professor Arrow shows, while Microsoft has enjoyed a dramatic expansion of its installed base, the growth of that base is largely the result of the success of the IBM-compatible PC platform and not of the licensing practices challenged in the government's Complaint. Accordingly, as Professor Arrow further explains, the proposed Consent Decree appropriately addresses and remedies the harmful effects of the challenged licensing practices by removing the artificial entry barriers they tended to erect, while imposing the expansive remedies suggested by the amici would be improper in this case.
Second, the decree forecloses nothing in the way of further cases against Microsoft. Microsoft will remain liable for any illegal acts, and anyone can challenge such conduct if and when appropriate. If any of the amici now have facts that provide a basis for suing Microsoft, they can do so. If the government obtains facts that it believes make out a cognizable legal claim, whether based on past or future conduct, it can do so as well. Conversely, if the decree is not entered, years of litigation will ensue, with no assurance that the public will obtain any relief, let alone relief as good as the relief provided in this decree.
Third, adopting the suggestions of the amici threatens to interfere with an on-going government investigation. Underscoring that the issues raised by the amici are well beyond the scope of the Complaint and proposed decree, the amici even include one matter, Microsoft's proposed acquisition of Intuit, which the government now is actively investigating. See Memorandum of Amici Curiae in Opposition to Proposed Consent Decree ("Memorandum") 32, 69-72. It would be improper to enter into a debate about this matter before our investigation is complete, and nothing in the Tunney Act either condones or requires such a debate. Granting the amici's request in this regard could defeat their own purposes -- it could prejudice our (or others') ability to challenge that acquisition. It also could prejudice the government's ability to prosecute other related cases against Microsoft, should it believe, in the future, that such a case or cases should be filed.
At bottom, amici miss a fundamental point: this is a proceeding under the Tunney Act to consider whether entry of this particular proposed Consent Decree, agreed to by the parties as a proposed Consent Decree of this case, would be in the public interest. It is not a proceeding to consider whether the government should have brought some other case. It is not a proceeding to consider whether other hypothetical settlements, to which the parties have not agreed, might also be in the public interest. It is not a proceeding to consider the optimal structure of the software industry without regard to the antitrust laws or where the government has not filed a case challenging that structure. And it is not a proceeding where approval of the decree will serve to insulate Microsoft from any liability for anything alleged in the amici's submission should such a case be brought by a private party or the government.
When the government settled this case last summer, the alternative to the proposed Consent Decree was massive litigation with an uncertain end at a distant point in time. And if the court rejects the proposed Consent Decree now, it is difficult to see that any but that result will be achieved.
The record in this case, including but not limited to Professor Arrow's declaration, establishes clearly that the proposed Consent Decree is in the public interest. All necessary Tunney Act procedures have been completed. We therefore ask the Court to approve the proposed Consent Decree forthwith.
I.D.E. Corporation ("IDEA") presents a much narrower concern. As IDEA raises it, that concern does not implicate the antitrust laws, and the Court should ignore it. And, although IDEA identifies behavior of Microsoft's that does implicate the antitrust laws, that behavior should not significantly bring into question whether entry of the decree would be in the public interest.
The United States and Microsoft have negotiated a proposed Consent Decree in this antitrust case. That negotiated settlement is now before the Court, and the Tunney Act, 15 U.S.C. 16(b)-(h), requires the Court to determine whether entry of the proposed decree is in the public interest, see id. § 16(e). The Act, however, does not give the Court the power to impose different terms on the parties. See, e.g., United States v. American Tel. & Tel. Co., 552 F. Supp. 131, 153 n.95 (D.D.C. 1982), aff'd sub nom. Maryland v. United States, 460 U.S. 1001 (1983) (Mem.); accord H.R. Rep. No. 1463, 93d Cong., 2d Sess. 8 (1974) [hereinafter House Report]. The Court, of course, can condition entry of a decree on the parties' agreement to a different bargain, see, e.g., American Tel. & Tel., 552 F. Supp. at 225, but if the parties do not agree to such terms, the Court's only choices are to enter the decree the parties proposed or to force the parties to litigate the very antitrust complaint already settled by the decree before the Court.
The realities and uncertainties of litigation thus constrain the Court in any particular Tunney Act case. See United States v. Gillette Co., 406 F. Supp. 713, 715-16 (D. Mass. 1975) (explaining that an antitrust defendant always could do something more but that "[a] point, however, comes, where an agreement ceases to be a compromise" and concluding that "[j]ust as the parties are compromising, so . . . must the court"). And the Court's action in a particular case has implications for antitrust enforcement generally. As Judge Greene observed:
If courts acting under the Tunney Act disapproved proposed consent decrees merely because they did not contain the exact relief which the court would have imposed after a finding of liability, defendants would have no incentive to consent to judgment and this element of compromise would be destroyed. The consent decree would thus as a practical matter be eliminated as an antitrust enforcement tool, despite Congress' directive that it be preserved.
American Tel. & Tel., 552 F. Supp. at 151.
The Court of Appeals has accordingly mandated that review of the government's proposed Consent Decree must be "deferential." United States v. Western Elec. Co. (Triennial Review Remand), 993 F.2d 1572, 1576 (D.C. Cir.), cert. denied, 114 S. Ct. 487 (1993). This Court is not "to make de novo determination of facts and issues." Id. at 1577 (internal quotation omitted). Rather, "[t]he balancing of competing social and political interests affected by a proposed antitrust decree must be left, in the first instance, to the discretion of the Attorney General." Id. (internal quotation omitted). As explained by the Court of Appeals, the district court's role in a Tunney Act proceeding is merely to ensure that the Department of Justice's explanation for the efficacy of the decree is "reasonable under the circumstances." Id. (internal quotations omitted).
This means that the Court's function is not to search for a remedy that would best serve society, "but only to confirm that the resulting proposed Consent Decree is within the reaches of the public interest." United States v. Western Elec. Co. (Triennial Review Opinion), 900 F.2d 283, 309 (D.C. Cir. 1990) (emphasis supplied by court) (internal quotations omitted); accord Triennial Review Remand, 993 F.2d at 1576-77; United States v. Airline Tariff Publishing Co., 836 F. Supp. 9, 12 (D.D.C. 1993).
Amici agree, but they insist that this standard of review is appropriate only when the Department has provided "economic affidavits that provide detailed support for the factual predicates underlying the Department's proposal." Memorandum 17-18. Amici are wrong. Neither this Court's standard of review nor its ability to find the decree in the public interest depends on the extent to which the Department develops a "detailed" factual record to support its competitive impact statement.(1) Indeed, Congress specifically contemplated that, in many cases, such a record would not be developed. See S. Rep. No. 280, 93d Cong., 1st Sess. 6 (1973) [hereinafter Senate Report] ("Where the public interest determination can be meaningfully evaluated simply on the basis of briefs and oral arguments, this is the approach that should be utilized." (emphasis added)); accord House Report, supra, at 8.
In any event, the Arrow Declaration explains why the relief sought in the proposed Consent Decree is appropriate. And this Court is entitled to find otherwise "only if it has exceptional confidence" that the government's predictive judgments are erroneous. Triennial Review Remand, 993 F.2d at 1577 (quoting Baltimore Gas & Elec. Co. v. NRDC, 462 U.S. 87, 103 (1983), and explaining that a "`court must generally be at its most deferential'" in reviewing the predictive judgments of the Antitrust Division).(2) This Court has no basis for such exceptional confidence on the record here.
A proper remedy for an antitrust violation is designed to "avoid a recurrence of th[at] violation and to eliminate its consequences." National Soc'y of Professional Eng'rs v. United States, 435 U.S. 679, 697 (1978). See generally American Tel. & Tel., 552 F. Supp. at 150-51 & nn.79-80. Accordingly, a proposed Consent Decree need not only prohibit the specific conduct constituting the violation alleged, but also can address other conduct if that is appropriate to achieve this objective. Professional Eng'rs, 435 U.S. at 697-98; American Tel. & Tel., 552 F. Supp. at 150. The proposed Consent Decree that is now before the Court does both. See 59 Fed. Reg. 42,845, 42,851-52 (1994).
However, not every remedy that extends beyond prohibiting the specific anticompetitive conduct is proper. Antitrust remedies must comprise "a reasonable method of eliminating the consequences of the illegal conduct." Professional Eng'rs, 435 U.S. at 698 (emphasis added). For the same reason, an antitrust remedy that proscribes more than the precise anticompetitive conduct alleged should impinge on other important public policies and interests no more than is reasonable to achieve antitrust goals. See id. at 697-98. And a remedy may not impose a penalty "in the guise of preventing future violations" or impose new duties, which is the task of Congress. Hartford-Empire Co. v. United States, 323 U.S. 386, 409 (1945).
Congress codified this traditional set of criteria for judging the propriety of antitrust remedies when it directed courts to assess whether the proposed Consent Decrees comport with the public interest. See House Report, supra, at 11-12; Triennial Review Remand, 900 F.2d at 308 ("[T]he `public interest' test must take its meaning from the nation's antitrust laws." (citing United States v. American Cyanmid Co., 719 F.2d 558, 565 (2d Cir. 1983), cert. denied, 465 U.S. 1101 (1984))); American Tel. & Tel., 552 F. Supp. at 150-51.(3)
Amici urge this Court to depart impermissibly from these settled principles. Relying upon a misreading of the Ninth Circuit's decision in United States v. BNS Inc., 858 F.2d 456 (9th Cir. 1988), they insist that this Court has carte blanche to consider the impact of the proposed Consent Decree in markets other than that alleged in the government's complaint. Memorandum 15-16. This is wrong. In BNS, the Ninth Circuit actually held that the Tunney Act "does not authorize a district court to base its public interest determination on antitrust concerns in markets other than those alleged in the government's Complaint." BNS, 858 F.2d at 462-63 (emphasis added). Although the court also held that "the statute clearly indicates that the court may consider the impact of the consent judgment on the public interest, even though that effect may be on an unrelated sphere of economic activity," id. at 463, this statement merely reflects the principle that an antitrust remedy should not unreasonably impair other public policies.
We do not assert that the antitrust consequences of Microsoft's illegal conduct in other markets necessarily are irrelevant to this Court's inquiry. But, under the law of antitrust remedies, such considerations are relevant only if addressing them is necessary in order to fashion adequate relief with respect to the specific restraints in the market for operating systems that were identified in the government's Complaint.(4) And we show below that there is no such necessity here.
Amici further suggest, see Memorandum 16, that this Court may second guess the government on whether it should have brought a different case -- one that, for instance, challenged a range of other alleged practices engaged in by Microsoft that, amici assert, have an anticompetitive impact in the operating systems market or some other market. But the Court may not do so; the decision to prosecute or not to prosecute any particular antitrust case is the government's and only the government's. The Court's public interest determination simply does not extend to this question.
It is well-established that "an agency's decision not to prosecute or enforce, whether through civil or criminal process, is generally committed to an agency's absolute discretion." Heckler v. Chaney, 470 U.S. 821, 831-32 (1985). As the Court explained in Heckler, "[t]his recognition of the existence of discretion is attributable in no small part to the general unsuitability for judicial review of agency decision to refuse enforcement." Id. at 831. It also is rooted in the Constitution. "[A]n agency's refusal to initiate proceedings shares to some extent the characteristics of the decisions of a prosecutor in the Executive Branch not to indict -- a decision which has long been regarded as the special province of the Executive Branch, inasmuch as it is the Executive who is charged by the Constitution to `take Care that the Laws be faithfully executed.' U.S. Const., Art. II, § 3." Id. at 832; cf. United States v. Nixon, 418 U.S. 683, 693 (1974) ("[T]he Executive Branch has exclusive authority and absolute discretion to decide whether to prosecute a case." (citing Confiscation Cases, 74 U.S. (7 Wall.) 454 (1869))).
An examination of the text, structure, and legislative history of the Tunney Act demonstrates that Congress did not intend to authorize judicial review of prosecutorial discretion through the public interest inquiry. The text of the act requires the government to describe in its competitive impact statement the practices giving rise to the "alleged violations." 15 U.S.C. § 16(b) (emphasis added). Similarly, in enumerating the facts that a court may consider in conducting its public interest inquiry, Congress made clear that the impact of the judgment properly is assessed in light of the "violations set forth in the complaint." Id. § 16(e)(2) (emphasis added); accord § 16(e)(1) (stating that a court may consider whether the judgment terminates the "alleged violations"). The Act's consistent focus on whether the proposed judgment adequately and reasonably remedies the violation alleged clearly implies that the United States is not obliged to discuss, and the court is not to review, whether the United States should have alleged different violations.
Construing the Tunney Act to encroach on the Justice Department's prosecutorial discretion would confound Congress' "unambiguous" intent to "preserv[e] antitrust precedent rather than innovat[e] in the usage of the phrase `public interest.'" House Report, supra, at 11. The pre-Tunney Act caselaw contains not one case in which courts, in undertaking their public interest assessment, examined whether the government should have brought a different case. And Congress clearly was aware of this fact when it characterized the adequacy of relief as the courts' proper concern. See Senate Report, supra, at 3 (explaining that the role of the court is "to make a judgment as to whether or not the proposed relief is sufficient with respect to the conduct alleged in the complaint" (emphasis added)). Reaching a contrary conclusion also would be inconsistent with the settled principles that the court has no power to force the Justice Department to bring a particular case, see In re Int'l Bus. Machs. Corp., 687 F.2d 591, 601-03 (2d Cir. 1982) (holding that the Tunney Act does not authorize review of the Division's decision to dismiss a case), or to compel the Department to agree to a particular settlement, see House Report, supra, at 8.
As the Ninth Circuit succinctly held, the Tunney Act does not authorize, in the context of assessing the antitrust implications of a decree, "`look[ing] beyond the strict relationship between complaint and remedy.'" BNS, 858 F.2d at 462-63. The reasons for Congress's choice are plain. To review the Department's exercise of prosecutorial discretion to bring a particular case at a particular time and not to bring another case would embroil courts in inquiries that are exceptionally difficult to answer. The Department's decision not to bring a particular case on the facts and law before it at a particular time, like any other decision not to prosecute, involves "a competing balance of a number of factors which are peculiarly within [the Department's] expertise" such as "whether [the Department's] resources are best spent on this violation or another, whether the [Department] is likely to succeed if it acts, whether the particular enforcement action requested best fits the [Department's] overall policies, and, indeed, whether the [Department] has enough resources to undertake the action at all." Heckler, 470 U.S. at 831. Congress properly did not charge the federal judiciary with making these difficult assessments.(5)
Nothing in the Tunney Act or its legislative history, then, provides any basis for stripping the Department of its discretion. The Department chose to sue Microsoft on the particular allegations contained in the Complaint because the Department believed that it could prove those allegations and obtain effective and reasonable relief. It chose not to include other allegations because the facts the Department then had available did not support reaching the same conclusion. If facts come to the Department's attention which the Department believes justifies filing an action under the applicable law, such an action will be filed. In the meantime, the decision to file the particular case which was filed, and not to file a different case, cannot be revisited by this Court in its public interest assessment.
It is important for this Court to bear in mind that the proposed Consent Decree does not arm Microsoft with a license to violate the antitrust laws. The government remains vigilant, and, indeed, will continue to investigate any possibly anticompetitive Microsoft conduct that comes to its attention.
For this reason, the Court must be especially careful to confine itself, as the Tunney Act requires, to an assessment of whether the relief sought is adequate to remedy the antitrust problem created by the specific conduct challenged in the government Complaint. If the government is improperly required to comment publicly on aspects of Microsoft's behavior that are not legitimately the subject of these proceedings, ongoing investigations or future investigations might be seriously compromised.
This is an immediate concern in the case of Microsoft's proposed acquisition of the applications maker Intuit, which, as amici well know, see Memorandum 70-71, the Division continues to examine. Amici plainly want the government to comment on whether appropriate relief in this case should include, inter alia, barring Microsoft from "acquiring stock in companies that make or sell application programs," such as Intuit. Id. at 95. But clearly, requiring the government to go on record with what amounts to our conclusions concerning an ongoing investigation is neither proper nor wise. Nor do we think that it is proper to force the government to risk compromising its position in any other investigations, pending or future, or to disclose materials germane to such investigations, disclosure which might raise serious issues of privilege, and compromise or harm the government's ability to prosecute other claims against Microsoft which may come to its attention in the future.
The basic point is this: the Court must carefully keep itself within the limits of the task with which it is seised. A "broad ranging inquiry" that Congress had no intent to permit, The Antitrust Procedures and Penalties Act: Hearings on S. 782 and S. 1088 Before the Subcomm. on Antitrust and Monopolies of the Senate Comm. on the Judiciary, 93d Cong., 1st Sess. 107 (1973) (statement of Sen. Tunney), can only reduce, not improve, the effectiveness of the antitrust laws to restrain Microsoft's potentially anticompetitive behavior.
Finally, this Court has a duty to expedite these proceedings. In view of the need to "preserve the consent decree as a viable settlement option," Congress directed that the public interest assessment should be made in "the least complicated and least time-consuming means possible." Senate Report, supra, at 6; accord House Report, supra, at 8. "Where the public interest can be meaningfully evaluated simply on the basis of briefs and oral arguments, this is the approach that should be utilized. Only where it is imperative that the court should resort to calling witnesses for the purpose of eliciting additional facts should it do so." Senate Report, supra, at 6. Indeed, Congress changed "shall" to "may" in what became 15 U.S.C. § 16(e) for the very purpose of underscoring this point. See id. at 8.
The Tunney Act thus plainly does not "mandate a hearing prior to the entry of every proposed Consent Decree." Senate Report, supra, at 3. To the contrary, Congress recognized that for courts to "engage in extended proceedings" might "have the effect of vitiating the benefits of prompt and less costly settlement through the consent decree process." 119 Cong. Rec. S13927 (daily ed. July 18, 1973) (statement of Sen. Tunney); accord id. at S13930 (prepared statement of Sen. Gurney); see also 120 Cong. Rec. H10765 (daily ed. Nov. 19, 1974) (statement of Rep. Jordan) (same). Still less does it mandate -- and the Court should not contemplate -- discovery into the government's files, with the attendant likelihood of protracted delay and prejudice to the government's ability to prosecute such future cases against Microsoft as it might decide to bring. See United States v. Airline Tariff Publishing Co., 1993-1 Trade Cas. (CCH) ¶ 70,191, at 69,684 (D.D.C. Mar. 8, 1993) (observing that "an order to compel production of documents . . . runs the risk of turning the flexible proceedings of the Tunney Act into a full-scale trial" and concluding that Congress did not "contemplate such a result in enacting the Tunney Act").(6)
If there is no very good reason for conducting further proceedings, none should be held. See Senate Report, supra, at 6; House Report, supra, at 8. As explained below, the Department of Justice believes that entry of the proposed Consent Decree is wholly within the reaches of the public interest, and that the Court now has fully sufficient information on which to base its determination. Consequently, this Court should require no proceedings beyond the scheduled January 20, 1995 hearing.
The Complaint filed in this case alleged that Microsoft had used exclusionary and anticompetitive contracts to market certain of its PC operating systems in violation of Sections 1 and 2 of the Sherman Act, 15 U.S.C. §§ 1, 2. Most significantly, Microsoft used its monopoly power to induce a significant percentage of PC manufacturers to enter into long-term "per processor" licenses under which they must pay Microsoft not only when they sell PCs containing Microsoft's operating systems, but also when they sell PCs containing non-Microsoft operating systems. These anticompetitive, long-term licenses created an artificial barrier to entry that helped Microsoft maintain its monopoly, because OEMs effectively would be required to pay a "tax" for using non-Microsoft operating systems. The Complaint also alleged that, in connection with pre-release testing of a new Microsoft operating system code-named "Chicago" (now, "Windows 95"), Microsoft sought to impose unreasonably restrictive and anticompetitive non-disclosure agreements on a number of leading developers of applications software products.
As explained fully in the Competitive Impact Statement, 59 Fed. Reg. 42,849, 42,849-54 (1994), the proposed Consent Decree completely cures the competitive problems caused by these practices and provides additional prophylactic relief as well. First, the proposed Consent Decree totally bans "per processor" licenses. Microsoft's revenue from a license may not be calculated on anything other than a per copy basis (i.e., a royalty for each unit of Microsoft operating system software licensed, sold or distributed) or a per system basis (i.e., a royalty for each computer system bearing a particular model name or number). Second, to prevent any abuse of per system licenses, the proposed Consent Decree requires, inter alia, Microsoft to provide its per system licensees with a statement advising the licensee of its rights under the license, and allows easy creation of new "systems" with different operating systems. Third, the proposed Consent Decree also severely limits the duration of license agreements for operating system software between Microsoft and personal computer manufacturers. Microsoft is prohibited from entering into any such license with a term exceeding one year, except that a license may include a term permitting the computer manufacturer to renew the agreement for up to one additional year on the same terms and conditions as those applicable in the original license period. Fourth, the proposed Consent Decree also bars unreasonably restrictive non-disclosure agreements of the type identified in the Complaint. Fifth, the proposed Consent Decree bans Microsoft from entering into license agreements that prohibit or restrict a personal computer manufacturer from licensing, selling, or distributing competing operating system products. Sixth, Microsoft may not condition the licensing of its operating systems on the licensing or use of other products. Seventh, Microsoft may not enter into any license containing a minimum commitment. Eighth, Microsoft is prohibited from using lump sum pricing.
Finally, the proposed Consent Decree has detailed transition rules that allow its provisions to have their intended effect immediately. Importantly, because the government secured a stipulation from Microsoft that it would comply with the proposed Consent Decree upon the filing of the Complaint, the procompetitive benefits took effect immediately. They will be lost entirely for several years (at least) if the Court rejects the proposed Consent Decree now before it, and forces the government to litigate allegations in the very Complaint whose claims are addressed so fully by the proposed Consent Decree.
The competitive benefits of the proposed Consent Decree are particularly important in the operating system software market today. As the Competitive Impact Statement notes, although Microsoft's per processor licenses did not begin until 1988, and gathered momentum relatively slowly, by 1993, when the Department took up this investigation, some 60% of OEM's were operating under the restrictive terms of per processor licenses. Those per processor licenses also contained large minimum commitments, and often were for lengthy durations as well. This meant that Microsoft's anticompetitive licensing practices, which had been relatively insignificant as late as 1991 were, on a forward-looking basis, likely to foreclose competitors from access to OEM's, the major distribution channel for competing operating systems. Meanwhile, as the amici point out, Microsoft today is developing and testing its new operating system, Windows 95, which now is expected to be released in August 1995. Entirely new licenses for this operating system must be negotiated by Microsoft. It is vitally important that these licenses be negotiated by Microsoft on terms which are free of the anticompetitive practices set forth in the government's Complaint. In the short run, the best hope for a competing operating system is now, when there is a "window of opportunity" to obtain the foothold for a new product so crucial to begin to generate the "positive feedback" described by Professor Arrow.
In short, the government's proposed Consent Decree is not only the right remedy: it comes at the right time. What will happen in the market is anyone's guess: it is not our job to pick winners or losers. It is our job, however, to level the playing field so that Microsoft and its competitors can fight it out in the market, in the best American tradition, with no artificial or unlawful restraints imposed by anyone. That is what the proposed Consent Decree does. For these reasons, it is vital that the Court enter it now.
The argument presented by the amici boils down to three contentions:
(1) the illegal practices challenged in this case led to a six-fold increase in the size of the installed base of Microsoft operating systems, and that increase accounts for much of Microsoft's market power in the market for PC operating systems;
(2) if the proposed Consent Decree is approved, Microsoft inevitably will leverage this market power into other markets unchecked by the antitrust laws; and
(3) sweeping remedies, perhaps even including a break-up of Microsoft, are appropriate in this case to deal with those problems.
Each of these contentions is wrong:
(1) While Microsoft has seen a dramatic expansion of its installed base, as Professor Arrow explains, "[T]he six-fold growth in the installed base is primarily the result of the extraordinary commercial success of the IBM-compatible PC platform . . . ." Arrow Dec'l at 11. Although Microsoft's anticompetitive licensing practices have had -- and certainly would have had in the future if left unchecked -- a significant effect on entry barriers in the PC operating system market, the contribution to the growth of Microsoft's installed base in the past has been relatively minor; it was not until 1992 when the licenses covered 50% of the distribution channel and Microsoft's installed base was already large. Id. at 12.
(2) Entry of the decree will not prevent anyone -- the Department or any private party -- from challenging any illegal conduct by Microsoft. Approval of this proposed Consent Decree will not prevent the government from challenging any prior illegal conduct if it believes that a case is warranted, or from challenging any future illegal conduct. The proposed Consent Decree will not prevent private parties from bringing any lawsuits, either.
(3) In this case, the law would not permit the sweeping remedies that the amici suggest, because those remedies go well beyond what is necessary to cure the effects of Microsoft's illegal licensing practices. Professor Arrow points out that it would be undesirable as a matter of economic policy to impose in this case the expansive remedies suggested by the amici and, moreover, the proposed Consent Decree appropriately addresses and remedies the anticompetitive effects of the practices challenged in the Complaint. Arrow Dec'l at 13.
Microsoft's rapid growth began in 1980, when it agreed to provide an operating system for IBM's personal computer. For reasons not relevant here, Microsoft's operating system, MS-DOS(7) was, for all practical purposes, the only available operating system for IBM-compatible personal computers, and IBM-compatible computers quickly became the dominant personal computers.
Neither the government nor the amici contend that Microsoft achieved monopoly power unlawfully.(8) Amici themselves offer an explanation for Microsoft's rapid success in their discussion of the economics of "Free Market Forces in Increasing Returns Industries." Memorandum 36-43. In essence, the first product in a new market may be able to achieve an enormous advantage over later products because the value of the product to particular users increases as more people use the product. Moreover, as amici observe, "compatibility" is an important factor giving rise to this advantage. These factors result in a fundamental characteristic of markets and technologies characterized by increasing returns: "the market tends quickly toward a single standard that dominates the market." Id. at 40. All this is plain from amici's submission.(9)
The operating system market evinces characteristics of increasing returns to scale. As Professor Arrow notes:
The analysis of the Department of Justice and the amici curiae brief agree that the software market is peculiarly characterized by increasing returns to scale and therefore natural barriers to entry. Large-scale operation is low-cost operation and also conveys advantages to the buyer. Virtually all the costs of production are in the design of the software and therefore independent of the amount sold, so that marginal costs are virtually zero. There are also fixed costs in the need to risk large amounts of capital and the costs associated with developing a reputation as a quality supplier. Further, there are network externalities, in particular, the importance of an established product with a large installed base and the related advantage of a product that is compatible with other complementary applications.
Arrow Dec'l at 5-6.
It is obvious that these market forces benefitted Microsoft. Professor Arrow refers to these market forces as creating "natural barriers to entry" and observes that "[t]he large installed base of IBM-compatible PCs that use Microsoft's operating system software reflects Microsoft's dominance of that market and undoubtedly contributes to its competitive advantage over competing operating system vendors."(10) Id. at 7. Indeed, as amici's submission makes clear, the central role of Microsoft's operating system in this industry is not a measure of its technological superiority to alternatives.(11) In short, Microsoft benefitted from the establishment of MS-DOS as the industry standard through the market success of the IBM PC and its clones.
This analysis, which can be drawn entirely from amici's submission and would be familiar to anyone who has studied this industry, explains Microsoft's rise to dominance. As Michael Morris, the General Counsel of Sun Microsystems, recently wrote:
The present source of Microsoft's domination in the PC world derives from its status as the standard-holder, not the practices the Justice Department condemned and which would now be prohibited under the settlement.
Michael Morris, Microsoft Deal: Too Little, Too Late, S.F. Examiner, July 24, 1994, at C-5. (quoted in App. to Mem. at Tab 33).
Thus, as the standard, the technological requirements of Microsoft's operating system are unquestionably important.(12) It is important to note, however, that an alternative to Microsoft's operating system might arise at some point, an operating system that either displaces Microsoft's or attracts sufficient users to gain the benefits of increasing returns to the point where the market is divided between the world of Microsoft and the world of this new operating system. The proposed Consent Decree insures that this new operating system, when developed, will have access to the market.
In 1988, Microsoft began to enter into "per processor" contracts with major OEMs. Complaint ¶ 26. These contracts, as the government explained in its Competitive Impact Statement, accounted for 20% of all units of MS-DOS that were sold to OEMs in Microsoft's fiscal year 1989, 22% of such units in FY 1990, 27% in FY 1991, 50% in FY 1992, and 60% in FY 1993. Competitive Impact Statement 59, Fed. Reg. 42,845, 42,850 & n.3 (1994). Microsoft's use of these licenses, which the government challenged as a device to "maintain" illegally Microsoft's monopoly, Complaint ¶ 36, cannot be explained as the product of market forces in an industry characterized by increasing returns. Instead, the licenses were artificial restraints on competition, precisely the sort of restraint properly condemned under the Sherman Act. As Professor Arrow explains, "Microsoft erected artificial barriers to the entry and growth of competing operating system vendors through its contractual relations with original equipment manufacturers of IBM-compatible PCs (OEMs)." Arrow Dec'l at 2.
The economic significance of the Microsoft licenses to other operating systems competitors, however, was closely tied to the increasing returns characteristic of the software industry. Most importantly, these licenses had the potential to prevent non-Microsoft operating systems from gaining enough of a foothold in the market to successfully generate their own positive feedback process. As the Complaint explains (¶ 18),
these practices reduce the likelihood that OEMs will license and promote non-Microsoft PC operating systems, make it more difficult for Microsoft's competitors to persuade [independent software vendors] to develop applications for their operating systems, and impede the ability of a non-Microsoft PC operating system to expand its installed base of users.
It was precisely this relationship between the restraint and the feedback process that created the most pernicious effect of the per processor licenses. The greatest potential threat to Microsoft's monopoly -- and the greatest potential boon to consumers -- would arise from a new operating system that is sufficiently superior technologically that it could displace Microsoft's operating system as the industry standard. Of course, any such new operating system could not be expected to displace Microsoft instantaneously. It first would have to establish a foothold in the market from which it could convince ISVs to write applications for it, which in turn would generate more consumer interest in the operating system. Per processor licenses and the other licensing practices ended by the proposed Consent Decree served to protect Microsoft's monopoly by preventing that from happening.
Professor Arrow describes with approval the proposed Consent Decree's benefits as the removal of these artificial entry barriers imposed by Microsoft's contracting practices, so as to allow for the possibility of entry by an innovative and technologically superior product:
Despite the importance of natural advantages (see section III below) in the market for IBM-compatible PCs, the complaint and proposed remedies addressed competitive issues that are critical to the success of new competition in this market. The most effective and economic point of entry for sales of IBM-compatible PC operating systems is the OEM distribution channel. New operating system software products should have unimpeded access to this channel. The Government's complaint and proposed settlement provide needed relief to facilitate the entry of new competitors, such as IBM's OS/2.
Arrow Dec'l at 5.
We cannot say when, or indeed whether, such an operating system will succeed in displacing Microsoft, but the rewards for that success are so large that we expect to see continued attempts. We also can say that those who mount such attempts should not have to face the additional artificial barriers created by Microsoft's unlawful license provisions.(13) The relief the government seeks will assure they will not have to.
We also can say with substantial confidence that DR-DOS, the principal competitor to Microsoft in 1988, was not likely to displace Microsoft as an industry standard. This is because DR-DOS marketed itself on the claim that it successfully mimicked most of MS-DOS's characteristics, e.g., it would run the same applications in the same manner on the same hardware. Such "clone" operating systems have no need to persuade independent software vendors to write applications: applications written for the Microsoft operating system work perfectly well on the clone. Such clone operating systems do not seriously challenge Microsoft's dominant position, because that position depends on Microsoft's role as standard setter. And an operating system that merely conforms to the standards Microsoft sets leaves Microsoft setting the standard.(14) As Professor Arrow explains, "[B]ecause DR-DOS supported the same application program interfaces as did MS-DOS, application program developers would have continued to write for MS-DOS (or Windows) even if DR-DOS sales had been much larger." Arrow Dec'l at 13.
A second competitor, IBM's OS/2 operating system, potentially threatens Microsoft's monopoly today, and the proposed Consent Decree also should help to level the playing field for it. OS/2 originally was developed and jointly marketed by Microsoft and IBM. In September 1993, that OS/2 "code sharing" arrangement ended, and OS/2 has belonged to IBM only. In one sense, OS/2 might be considered a clone of MS-DOS and Windows and part of the installed base of those products, because it runs applications written for both. But OS/2 also has capabilities lacking in MS-DOS and the current Windows, capabilities used by applications written for OS/2. It is possible that OS/2's acceptance in the marketplace was impeded by Microsoft's illegal licensing practices in the period between 1993 and the filing of this case. These adverse effects on OS/2, however, are unlikely to have significantly increased Microsoft's installed base in that relatively short period. The elimination of the restraints, however, is potentially significant, since IBM is now marketing OS/2 very aggressively in the hope of significantly increasing its market penetration.
Where the proposed Consent Decree addresses specific unlawful practices in the PC operating system market, and eliminates artificial barriers to competition in that market, amici reference a variety of remedies apparently aimed at reducing Microsoft's role in markets for applications programs. We believe that amici have missed the point of the government's case.
The crux of their argument is that even though Microsoft achieved its monopoly position lawfully, the growth of Microsoft's installed base since 1988 results from the practices the government challenged as anticompetitive, see Memorandum 8, 9, 44, 50-51, 84, and that the government's relief is inadequate, because it will not dissipate the market power that the installed base reflects. However, contrary to amici's assertions, the government has not contended that Microsoft "illegally acquired its massive installed base," Memorandum 6, or that the size of its installed base would be substantially smaller today but for Microsoft's challenged licensing practices.
Professor Arrow points out a number of failings in amici's assertions that the growth in the installed base resulted from the challenged licensing practices. In particular, he explains that the amici's contention is in large measure refuted by their concession that Microsoft's monopoly resulted from natural market forces. As Professor Arrow notes:
This conclusion appears flawed for a number of reasons. Clearly, the six-fold growth in the installed base is primarily the result of the extraordinary commercial success of the IBM-compatible PC platform, in which Microsoft's product development and marketing played a part. In such a situation of rapid growth, the previous installed base should have provided a relatively weak constraint of entry. For the most part, Microsoft appears to have achieved its dominant position in its market as a consequence of good fortune and possibly superior product and business acumen.
Arrow Dec'l at 11-12.
To be sure, the challenged practices had serious anticompetitive effects. But the government's concern about those practices focussed primarily on their implications for the future -- on their effects in preventing the emergence of a new, technologically superior operating system -- rather than their past effects on clone operating systems, which had only minor effects on the installed base. Unlike the amici, the government does not adopt the view that Microsoft's monopoly, once attained, must be permanent. There are indeed substantial natural barriers that stand in the way of any would-be challenger to Microsoft, and which the antitrust laws do not make the subject of legal challenge. But the existence of those substantial natural barriers to entry makes it all the more necessary to eliminate the artificial barriers that Microsoft erected.
The proposed Consent Decree eliminates the artificial barriers, and so should help to level the playing field and open the market in the future. The government brought the case it did and obtained the relief in the decree precisely to remove these artificial and anticompetitive restraints. For this reason, and because it is wholly within the public interest, the government believes that the decree should be entered forthwith.
Accepting amici's invitation to restructure the computer industry more to their liking through sweeping remedies such as dismembering Microsoft very well might advance the private interests of the anonymous amici; but such remedies would not necessarily benefit competition and would, in Professor Arrow's view, act against the public interest. As Professor Arrow states, "a rule of penalizing market successes that are not the result of anticompetitive practices will, among other consequences, have the effect of taxing technological improvements and is unlikely to improve welfare in the long run." Arrow Dec'l at 10. In any event, that is not the issue before the Court in this Tunney Act proceeding. Before the Court today is one issue: whether to find that the proposed Consent Decree is within the reaches of the public interest. For the reasons set forth herein, the government believes that it is, and should be entered forthwith.
Amici claim that this proceeding is the final act in the story of Microsoft and the antitrust laws. If this proposed Consent Decree is entered, they suggest, Microsoft inevitably will engage in a litany of practices that will lead to its total domination of all corners of the software industry. They also seem to suggest that the entry of this proposed Consent Decree is the equivalent of granting Microsoft a license to violate the antitrust laws.
The amici recognize that the government is reviewing Microsoft's proposed acquisition of Intuit, a proposal that arose after the government brought this case. The government also will vigorously and promptly investigate any and all other facts or claims brought to it which might make an antitrust case. But amici apparently consider the outcome of that and any other investigation to be a foregone conclusion. The government does not. We have not prejudged any investigation, and the outcome of any investigation of Microsoft in no way turns on whether the Court enters the proposed Consent Decree here.
IDEA's grievance and its demand for a refund of money it paid to Microsoft pursuant to a contract raise two questions that properly are addressed in these proceedings. First, does Microsoft's apparent unwillingness to return to IDEA money IDEA paid to Microsoft pursuant to contract raise a concern bearing on whether entry of the proposed Consent Decree is within the reaches of the public interest? Second, do the likely consequences for competition in the PC operating systems market of Microsoft's proposed amendment to its licensing agreement with IDEA lead to the conclusion that entry of that proposed Consent Decree would not be in the public interest? The answer to both questions is no.
IDEA says that in 1989, another company, whose relevant business IDEA later acquired, entered into a licensing agreement with Microsoft for PC operating systems.(16) The agreement contained minimum commitment provisions requiring the payment of royalties on at least a certain number of operating systems each quarter even if fewer than that number of operating systems were shipped on its computers. This turned out to be a bad bargain for IDEA. Its shipments fell far below the minimum commitments over time and IDEA had to pay approximately $2 million not attributable to royalties on operating systems it shipped.
In 1993, Microsoft and IDEA entered into a new licensing agreement for a three-year term. It provided for minimum commitments at a dollar level nearly 90% below that in the 1989 agreement. It also provided that in any quarter that IDEA incurred payment obligations above the minimum commitment, the overage would be covered by the $2 million IDEA had previously paid Microsoft in "unused" commitments. Under this agreement, IDEA's shipments have been greater than the minimum commitments in some quarters and significantly less in other quarters. The net result is that the amount of "unused" commitment payments remains today at approximately $2 million.
Microsoft, by its stipulation, is currently bound by the provisions of the proposed Consent Decree which prohibit its enforcement of the minimum commitment provision of the 1993 license agreement. Proposed Consent Decree § IV(J)(2). Therefore, IDEA is free to terminate the agreement or to negotiate with Microsoft to eliminate the inconsistent provision, id. § IV(I).
IDEA has not chosen to terminate the agreement. Microsoft has proposed amending the agreement to provide that IDEA is relieved, prospectively, of its obligation to pay minimum commitments, and that IDEA may "recoup" by having the "prepaid royalties" applied to any amounts in royalty obligation IDEA incurs in a quarter above the minimum commitment figure in the agreement. If that proposal were accepted, IDEA would pay for any quarter an amount less than the minimum commitment figure or, if its use of operating systems was sufficiently large, an amount equal to the minimum commitment figure. But as long as any of the $2 million in "prepaid royalties" remained, IDEA would not have to pay more than the minimum commitment figure, no matter how many operating systems it ships in the quarter. That, obviously, is a better deal for IDEA than the one it signed in 1993.
IDEA has not accepted Microsoft's proposed amendment. Instead, it urges the Court to reject the proposed Consent Decree unless the United States and Microsoft modify it to require that Microsoft refund the full amount of the prepaid royalties to IDEA and other similarly-situated Microsoft licensees, if any. It wants to reverse the allocation of risk to which it agreed in 1989 and 1993 while not revising the price.
IDEA's interest in getting a refund is not relevant in this proceeding, because its desire to recover its unused $2 million in minimum commitments has nothing to do with competition in the operating systems market.(17) Those agreements did not inflict a competitive injury on IDEA, or injure the markets in which IDEA competes (we discuss the PC operating systems market below). IDEA entered into these agreements because they looked better than the available alternatives. But what turns out to be a bad business deal does not become an antitrust violation merely because one party to the business deal is a monopolist.(18)
The Tunney Act permits this Court to consider the effect of a proposed decree on "individuals alleging specific injury from the violations set forth in the complaint." 15 U.S.C. 16(e)(2). However, because IDEA's injury was the result not of antitrust violations, but instead of a business deal gone sour, there is no reason to consider it here.
Under Microsoft's proposed amendment to the 1993 license agreement, Microsoft neither keeps all the prepaid royalties nor refunds them all to IDEA. Instead, it returns part of the money to IDEA for any quarter in which IDEA incurs royalty obligations greater than the minimum commitment amount. In effect, once IDEA reaches the minimum commitment level of incurred royalties, Microsoft reduces the royalty for further operating system shipments in that quarter to zero. As IDEA observes, such an arrangement could provide an OEM with a disincentive to ship its computers with competing operating systems.(19)
Nothing suggests that the disincentive involved here raises questions about whether the proposed Consent Decree is in the public interest. The proposed Consent Decree was not intended to eliminate all disincentives to purchase operating systems from suppliers other than Microsoft. It specifically does not prohibit volume discounts, proposed Consent Decree § IV(J), although volume discounts may provide an incentive to buy from the vendor who offers them and a disincentive to buy from others. This is not an oversight. "[T]he Department . . . considered whether to require limitations on the manner in which Microsoft could structure volume discount pricing arrangements," but decided not to do so, in part because of the potential for procompetitive benefits from volume discounts. 59 Fed. Reg. 42,845, 42,854 (1994). Microsoft's proposed amendment creates a volume discount structure;(20) the price up to the minimum commitment level is the specified royalty per copy, and the price for copies above that level is zero.(21) Whether the volume discount actually provides IDEA with a significant incentive to purchase licensed operating systems from Microsoft rather than someone else is not clear. In any quarter IDEA anticipates shipments below the triggering figure, the volume discount should be irrelevant to IDEA's choice of operating system, since IDEA would not expect to reach the volume discount. In any event, IDEA has not contended that it would seriously consider shipping its products with a non-Microsoft operating system whatever the relative prices.
Widespread use by Microsoft of volume discounts resulting in a marginal royalty rate of zero could present competitive problems in the PC operating system market and might serve to foreclose access to the marketplace by competing vendors. In some circumstances such discounts could be arranged so as to be lump sum pricing, which is barred by the proposed Consent Decree, in all but name. But, we have no reason to believe that situations like that of IDEA are common in the industry or that the resulting volume discount provisions present any significant foreclosure problem. In IDEA's case, the large overhang of prepaid royalties in relation to both current shipments and the trigger figure for the volume discounts reflects a very substantial decline in IDEA's volume of business, and not a short-term one. This is an unusual case that does not create any significant foreclosure.
In the circumstances, the Court should conclude that Microsoft's proposed amendment to its license agreement is a reasonable way to handle the current contractual commitment to set off against existing prepaid royalties. It should also conclude that the proposed amendment does not cast any doubt on whether entry of the proposed Consent Decree falls within the reaches of the public interest.
At the September 29th Hearing, the Court expressed concerns over the allegations of a journalist about a number of Microsoft practices. These practices include the use of false or misleading product preannouncements ("vaporware"), the misappropriation of intellectual property, and the failure to maintain a "Chinese Wall" between operating systems and applications developers.
After evaluating the known facts before it in light of the relevant case law, the government to date has chosen not to seek antitrust relief relating to such conduct. As discussed above, that decision does not preclude the government from challenging such conduct, or any other anticompetitive conduct, in the future should circumstances so warrant. Nor does it prevent private parties from bringing a suit if they believe it appropriate. We note that the Federal Trade Commission and the European antitrust authorities were aware of the same allegations. The Federal Trade Commission neither sought nor obtained any relief. The European authorities sought, and obtained, precisely the relief contained in the proposed Consent Decree.
Because these issues involve conduct unrelated to that charged in the Complaint, examination of whether the government's decision to challenge such conduct was correct is beyond the scope of this Tunney Act proceeding. See supra Section I. Moreover, because relief directed at such issues has no relationship to the specific violations alleged in this Complaint, the Court cannot properly investigate further into these issues.
The relief provided by the proposed Consent Decree properly addresses the violations alleged in the Complaint. It frees the market from the artificial barriers Microsoft created. If the government learns of facts based on conduct upon which it can file a case under antitrust case law, it will do so. Private parties, including the amici, also are free to sue Microsoft at any time on any claim they and their lawyers believe justified under the law. The proposed Consent Decree here forecloses nothing. It achieves relief not available only six months ago, and leaves the door wide open to further suits on other claims by the government or private parties. Not to enter the decree would, in effect, be a major step backward: it would restore Microsoft's ability to engage in the very anticompetitive practices the proposed Consent Decree prohibits, and Microsoft could engage in such practices until the conclusion of a long, difficult case.
The record currently before the Court is sufficient for the Court to conclude that this proposed Consent Decree is in the public interest. It should do so forthwith.
The investigation by the Department of Justice in this case is, so far as we know, unprecedented, in that it began as a result of two consecutive 2-2 votes by the Federal Trade Commission after its three and one-half year investigation of Microsoft on the same claims. As the Assistant Attorney General explained to the Court in the hearing on November 2, 1994, the Department asked to assume the investigation in order to act, in effect, as the "Fifth Commissioner," because of the importance of the industry to the American economy, and the need for a final decision other than by default. See Transcript of Status Call, Nov. 2, 1994, at 22.
The Department had full access to the voluminous files and records of the FTC, and in addition issued 21 its own Civil Investigative Demands upon Microsoft and numerous third parties. In all, the Department's investigation of every claim of which it was aware took some 14,000 attorney hours, 5,500 paralegal hours, and 3,650 economist hours. In the investigation, the Department reviewed a total of one million pages of documents, including those transferred from the FTC, took 22 depositions, including depositions of top Microsoft officials, and conducted well in excess of 100 interviews. This included interviews of former Microsoft employees, and individuals at approximately 80 companies, including competing software companies, original equipment manufacturers ("OEM's"), and important end users, among others.
As the Assistant Attorney General represented to the Court at the November 2, 1994 Status Call, she personally participated throughout the case from September, 1993 through July, 1994 and spent hundreds of hours reviewing the evidence, deciding what case was appropriate at that time, on the facts then known to the Department under applicable precedent, and engaging in and leading the settlement discussions with Microsoft and the EC which resulted in the proposed Consent Decree now before the Court in this Tunney Act proceeding.
It is now more than six months from the date that the proposed Consent Decree was filed with the Court. The government and the defendant have complied with all of the procedures required by the Tunney Act. The record demonstrates that the proposed Consent Decree is in the public interest. The Court should reject the procedural and substantive recommendations of I.D.E. Corporation and of the amici and should enter the proposed Consent Decree forthwith. A motion to enter judgment is appended to this Memorandum.
January 18, 1995
1.United States v. Western Elec. Co., 993 F.2d 1572 (D.C. Cir.), cert. denied, 114 S. Ct. 487 (1993), and United States v. American Tel. & Tel. Co., 552 F. Supp. 131 (D.D.C. 1972), aff'd sub nom. Maryland v. United States, 460 U.S. 1001 (1983) (Mem.), heavily relied upon by amici, involved proposed modifications to a decree and not, like this case, the entry of a decree. This distinction is significant. Requiring, as the D.C. Circuit appears to, "substantial factual support," Triennial Review Remand, 993 F.2d at 1581, for a proposed modification (as opposed to an initial decree) will not significantly deter parties from entering into proposed decrees initially.
2. See Triannial Review Remand, 993 F.2d at 1582 (holding that the presentation of Professor Arrow was "enough . . . to establish ample factual foundation for the judgment call made by the Department of Justice and to make its conclusions reasonable").
3. Congress also included within that criteria the special concerns "inherent in the process of settling cases through the Consent Decree procedure." House Report, supra, at 12.
4. A different question might be raised if the proposed relief in the market that is the government's concern has the effect of making other markets less competitive than before. Such concerns are not raised here.
5. Moreover, construing the Tunney Act to permit courts to consider the Executive's exercise of its discretion in undertaking the public interest assessment would raise difficult constitutional issues. Cf. Nixon, 418 U.S. at 693 (asserting that some Executive Branch decisions are committed to the Executive's "absolute discretion"); United States v. Cox, 342 F.2d 167, 171 (5th Cir.) ("It follows, as an incident of the constitutional separation of powers, that the courts are not free to interfere with the free exercise of the discretionary powers of the attorneys of the United States in their control over criminal prosecutions.") (cited approvingly in Nixon, 418 U.S. at 693), cert. denied, 381 U.S. 935 (1965).
"[W]here an otherwise acceptable construction of a statute would raise serious constitutional problems, the Court will construe the statute to avoid such problems unless such construction is plainly contrary to the intent of Congress." Edward J. DeBartolo Corp. v. Florida Gulf Coast Build. & Const. Trades Council, 485 U.S. 568, 575 (1988); accord Communication Workers of Am. v. Beck, 487 U.S. 735, 762 (1988); see also Blitz v. Donovan, 740 F.2d 1241, 1244 (D.C. Cir. 1984). Because, as demonstrated in the text, construing the Tunney Act to preclude an inquiry into the breadth of the allegations in the complaint is "permissible," Apache Survival Coalition v. United States, 21 F.3d 895, 904 (9th Cir. 1994), the Act should be so construed to avoid this constitutional question.
6. The Court should reject amici's request, see Memorandum 94, that the government should be compelled to turn unspecified "key" documents over to the Court. Whatever these documents might be, the Court already has sufficient information to make its public interest determination. See generally infra Section III. Moreover, we object to the request by amici to produce "key documents" on the grounds of privilege and lack of specificity.
7. MS-DOS was often distributed under other names, such as PC-DOS and Compaq-DOS, reflecting the computer manufacturer who licensed MS-DOS for its computers.
8. See Memorandum 44 (not challenging that "Microsoft's initial monopoly was lawfully obtained"); 49 (attributing Microsoft's "control of the personal computer market" to "riding IBM's coattails," not to illegality (quoting Complaint ¶ 19 ("Microsoft quickly dominated and gained a monopoly in the market for PC operating systems"))).
9. It is, of course, also plain from the Complaint in this case (¶ 17) that:
the value of an operating system to a consumer is directly related to two factors: the availability of a variety of high quality applications that run on that system, and the number of users who use that operating system and thus are able to share information and work with the system without additional training. [Independent software vendors], in turn, tend to develop applications for operating systems with a large installed base of users, and consumers gravitate towards operating systems with a large base of applications.
10. Although Professor Arrow recognizes that increasing returns may lead to natural barriers to entry, he explains that attempts to regulate or interfere with "purely natural barriers to entry can be dangerous to the economy's welfare." Arrow Dec'l at 10.
11. Amici believe that at various times there were technologically superior alternatives to Microsoft's operating system products. See Memorandum 51. The accuracy of that belief is irrelevant in this proceeding, and we express no view.
12. For simplicity in exposition, we have focused the text on MS-DOS, ignoring Microsoft Windows, as the market has not. In brief, some years ago Microsoft decided to offer a product that would in effect extend the capabilities of MS-DOS in various ways, including provision for what is called a Graphical User Interface (GUI), a feature widely thought to be highly desirable. Microsoft's first two versions of Windows were not commercial successes, and relatively few applications were written to work with them. Version 3 was considerably more successful, while Version 3.1 was wildly successful. As a result, commercial success for applications programs now generally depends on compatibility with the technological and other requirements of Windows, while vendors of competing operating system products must worry about whether their products are capable of running Windows and/or applications written to run under Windows. And again Microsoft controls and best knows the technological requirements of Windows.
13. As the Assistant Attorney General said in the press conference announcing the proposed Consent Decree, "if a competitor has a better product that can run computers faster, run them better, support better applications, build a base, cut into Microsoft's market share so that applications writers will write for it, that could have profound consequences for the American economy. What we are about is precisely that -- promoting competition, innovation, better products at cheaper prices, and letting the market take care of whatever happens." Transcript of Press Conference, July 16, 1994, at 15 (quoted in App. to Mem. Tab 12).
14. Microsoft's licensing practices did, however, affect such clone competitors in ways that harmed consumers, and the proposed Consent Decree should help to remedy that problem. Per processor licenses discouraged the use of MS-DOS clones, and the consumers who bought machines with clone operating systems may have paid a "tax" to Microsoft, because of the effects of the per processor license. The vendor of such an operating system, and PC users generally, should benefit from the proposed Consent Decree because it eliminates the penalty on OEMs that install such an operating system on their computers. Eliminating this penalty may allow clone products to increase their market share, but that effect does not alter the size of the installed base of PCs that run applications written for the Microsoft operating system. It should, however, lower prices to consumers, as the Attorney General noted on July 15, 1994 in announcing the settlement.
15. Amici must know that the sources of Microsoft's power lie elsewhere, for their own submission makes clear that the fundamental source of Microsoft's power is the economics of increasing returns. Memorandum 36-43.
16. For purposes of this discussion, we rely on IDEA's representations in its filings and in its communications with us for the relevant facts, and since it does not matter for present purposes, we will not distinguish IDEA from the other company.
17. Minimum commitments provisions are among the license provisions that the Division challenged. But, "minimum commitments are not in and of themselves illegal." 59 Fed. Reg. 42,845, 42,852 (1994). The matrix of Microsoft's licensing practices and its wide use were the reason the Department challenged a minimum commitments agreement such as those between Microsoft and IDEA; the effect on willing OEMs who entered into these agreements did not concern the government or the Sherman Act. The injured parties were competing operating system vendors, not OEMs.
18. A plaintiff seeking relief under the antitrust laws must show injury, or threatened injury "`of the type the antitrust laws were intended to prevent and that flows from that which makes the defendants' acts unlawful.'" Cargill, Inc. v. Monfort of Colorado, Inc., 479 U.S. 104, 113 (1986) (quoting Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977)).
19. This disincentive does not turn on the use of the minimum commitment figure in the arrangement. Any other triggering figure that IDEA had any hope of reaching in a quarter would provide a similar disincentive.
20. The proposed amendment does not create "lump sum pricing," which the proposed Consent Decree prohibits prospectively, see proposed Consent Decree § IV(H), because the royalty payment actually does vary with the number of copies licensed, sold, or distributed, see id. § II(F), in IDEA's case.
21. The government previously noted its concern about the possibility that Microsoft might in the future adopt anticompetitive volume discount structures that effectively coerced buyers into buying all or substantially all their operating system requirements from Microsoft with the result of foreclosing competing suppliers from the marketplace. The government also noted that its concern lay with coercion resulting from the volume discount structure, not from the level of the price. See 59 Fed. Reg. 42,845, 42,854 (1994). Since under Microsoft's proposed amendment IDEA would receive no discount on copies purchased up to the minimum commitment as a result of exceeding the minimum commitment, we view any coercive effect here as resulting essentially from the low price above the trigger level, not from the structure of the discount.