Highlights and updates from Microsoft @ 25: Impact, Influence, and Legacy – Part 1
On February 21, New York University School of Law hosted “Microsoft @ 25: Impact, Influence, and Legacy”, a conference dedicated to assessing United States v. Microsoft’s impact on both antitrust law and digital markets in anticipation of its 25th anniversary in 2026. Beyond discussions of law and economics, the conference offered stories of spy-like meetings, dramatic witness confrontations, and reminders about the long-term impact of antitrust enforcement in shaping and re-shaping the economic landscape.
This article covers the keynote address and first two conference panels. In the next issue of the Competition Bulletin, the remaining panels will be covered.
WATCH VIDEO - Keynote Remarks from David Boies, trial counsel for the United States in United States v. Microsoft
New York University professor emeritus Harry First welcomed attendees and stressed the ongoing importance of the Microsoft case, noting the existence of five ongoing “big tech” cases. In these cases, First emphasized that “The Microsoft case has been critical to framing [and] shaping all of these cases, really, in an explicit and implicit way.”[1]
The conference moved into keynote remarks from Chairman Emeritus of Boies Schiller Flexner LLP David Boies, who served as counsel for the United States at the District Court in United States v. Microsoft, elicited in conversation with Hausfeld LLP Chair Emeritus Michael Hausfeld.
In considering the significance of the Microsoft case to the current antitrust environment, Boies emphasized that one disappointment he had with the case was that it did not assert a Section 7 Clayton Act claim. While the mergers and acquisitions that helped build Microsoft were not as extensive as they have been for current tech giants like Meta or Google, Boies highlighted that it was “very, very important to Microsoft’s dominance.”
Boies underscored that, at the time Microsoft was brought, there was “considerable debate as to whether the antitrust laws even ought to apply to really high-tech companies . . . a number of people thought that high tech industries were different, they were too dynamic. The markets were changing too rapidly for the antitrust laws really to play a role.” Instead, he noted his belief that, if Microsoft had come out differently, “Google might not have had the opportunity to develop. Apple might not have survived.”
In assessing the outcome of the trial, Boies highlighted the courtroom drama that, in his mind, might have brought the court to the conclusion that it could not trust Microsoft or its witnesses. Boies highlighted two scenarios that contributed to this potential distrust. In the first, Microsoft showed a video to demonstrate that Microsoft’s internet explorer worked much better than Netscape Navigator. In the video, “everything went very fast, and they got very quick responses and then they supposedly put on the Netscape browser on the same computer, and it slowed down. It really slowed down.”
Though the government was unable to replicate this video in its own efforts, during the cross-examination, Boies admitted “I didn’t confront it directly because I didn’t know how to do it.” Had Microsoft simply dismissed the witness, the video would have remained as untouched evidence. Instead, the court dismissed the parties early that day and with a night of review, the government identified a dramatic inconsistency in the opposing expert’s presentation demonstrating how Netscape caused Windows to operate more slowly. Following a break in testimony, Boies asked his team what they had found and, through screenshots of the video, revealed that the computer running Netscape had additional programs in operation, as reflected by icons on the screen. This enabled him to confront the witness, who could offer no explanation for the inconsistency.
Boies also recounted an expert examination in the trial where he asked an expert witness whether he had considered Microsoft’s extremely high profits in assessing whether Microsoft had market power. The expert’s response? “No, monopoly profits have no relevance to the market power analysis”.
The next question, “have you ever thought that profits should be looked at in a monopoly case” elicited the response “Absolutely not. No sensible economist ever would look at this.”
At this point, Boies recounted presenting the witness with an article from the Harvard Law Review written by the same expert. From that point on, Boies said, that expert was known as the “what could I have been thinking of?” expert.
Boies noted that Microsoft would present these occurrences to the media as matters of little significance. He even agreed, to some extent, stating “they were basically right that it didn’t, it didn’t matter, except for the loss of credibility. And that’s a little bit like ‘other than that, Mrs. Lincoln, how was the play?’ Because in a trial where the issues are very complicated and uncertain, credibility makes a huge amount of difference.”
WATCH VIDEO - Panel: The Development of Monopolization Law
The first panel, moderated by Professor Andrew Gavil of Howard University School of Law, assessed Microsoft’s role in shaping monopolization law.
Professor Rebecca Haw Allensworth of Vanderbilt University Law School highlighted that there are two fundamental approaches to monopoly and how to induce innovation and that these approaches are traded off in platform cases.
In the first of these approaches, “there is a monopoly prize for which companies will strive, and they will innovate. And once they have it, we ought not to take it away, or not take it away fully, lest they don’t strive to get it in the first place.” This view could be contrasted with the idea that innovation is driven by ease of entry, that providing avenues for new entrants to quickly reach numerous customers can itself spur innovation.
In the lower court, she noted, the court offered some “pushback” to the former prize theory and rejected Microsoft’s assertion that its “desire to limit free riding on the firm’s investment in consumer-oriented features” was pro-competitive.[2] The district court rejected this justification as insufficient and imposed a remedy that would have opened up the platform to more companies.
Professor Haw Allensworth noted that instead, on appeal, the appellate court said, “Microsoft made no pro-competitive justifications.” This, she added, “makes it sound like anything will do. Just get up there and say something . . . about how this design worked better.”
Professor Emeritus Jonathan Baker of American University Washington College of Law highlighted that, at the time Microsoft was decided, there were precedents “that could be read to exempt application of antitrust rules to the development of new products” and there was also a “policy concern to avoid chilling innovation.” If Microsoft had offered “that kind of counterfactual opinion,” Professor Baker noted, “it could have transformed antitrust in the courts.” But later writings from two judges on the case “placed particular emphasis . . . on the absence of a non-pretextual justification for at least some of the challenged practices” making it “hard . . . to write a counterfactual opinion saying monopolization claims in dynamic markets often make no economic sense when the defendants didn’t proffer a plausible justification for the practices.”
Professor Daniel Francis of New York University School of Law emphasized that the Microsoft case had, in some ways, “become a bit of a millstone around Section 2’s neck and . . . a little bit more critical distance from . . . Microsoft . . . would be great for monopolization litigation going forward.” The unease he noted with the Microsoft opinion included a failure to provide a basic foundational rule as to what the law of monopolization prohibits. The opinion, he noted, famously states that “harm to one or more competitors will not suffice” to show anticompetitive effect, but the bases for concluding an anticompetitive effect was the reduction of “rival browser share. Pure, explicit harm to rivals.” The opinion, in his view, fails to inform practitioners “when we can make that inference from bare harm to competitors”.
Francis later highlighted that one harm of Microsoft is that a defendant gains “an irrebuttable shield from the antitrust laws if you can persuade a fact finder that what you’re doing is competition on the merits without us having, in the fabric of Section 2, a sense of what that means. For some courts, it’s conduct motivated by legitimate purpose. For some courts, it’s conduct that can be shown to be beneficial overall. For some courts, there are specific kinds of behavior like above-cost unconditional pricing, that, even when they harm consumers and competition, can never be the source of Section 2 liability.”
Professor John Newman of the University of Miami School of Law described Microsoft as an example of what he calls “lawless antitrust”. As an example, he noted that, before the Microsoft opinion cited any binding legal authority, it cited nearly a dozen secondary sources. Similarly, he noted that the Court’s statement that “Plaintiffs must prove harm to competition and consumers” is not supported by the legal citations that the Court provided. The Court then adopts a test with framing that is “blatantly defense-friendly.” As an example, Professor Newman cited disparate language in the Court of Appeals’ opinion.
Plaintiffs apparently must “demonstrate that the monopolist's conduct indeed has the requisite anticompetitive effect”[3] and “demonstrate that the monopolist's conduct harmed competition, not just a competitor.”.[4] In contrast, the defendant in a section 2 case is required only to “assert a procompetitive justification” with the only requirement being that the justification be non-pretextual;[5] the D.C. Circuit’s language provides that this non-pretextual shifts the burden “back to the plaintiff to rebut that claim.”[6]
This framework, Professor Newman noted, is not required by the citation the Court provides, nor can it be grounded in the text of the Sherman Act itself, which provides that it is illegal to monopolize “any part of the U.S. economy”. While this would give a textual grounding to simply reject out of market justifications, it did not do so. That “missed opportunity . . . is one that . . . is going to hang [and] already has hung really heavily over a lot of the big tech cases,” Professor Newman said.
WATCH VIDEO - Panel: The Economics of Platforms and Innovation
The second panel, moderated by Professor Christopher Jon Sprigman of the New York University School of Law, assessed whether the approach taken in Microsoft has endured given that it predates much of the – then still nascent – work on platform economics and digital markets.
Steven Salop, Professor of Economics and Law Emeritus at the Georgetown University Law Center, offered a brief history of the study of network effects that began in the 1970s and gained significant in the 1980s and 1990s.
Professor Salop noted that in the context of Microsoft, Microsoft was increasing the barriers to entry of middleware like Java and Netscape – products that would encourage developers to write for non-Windows-specific platforms. Microsoft addressed this “with threats and exclusives.” He noted that while Chicago economists would say that exclusive arrangements were simply Microsoft competing, the playing field on which Microsoft was competing was not a level one, because “the stakes are asymmetric.” If a monopolist is competing with a rival and the two are bidding for distribution, the value will always be greater to a monopolist unless a duopoly would increase the overall revenues. “As long as monopoly profits exceed combined duopoly profits, the monopolist will win the bidding . . . If, on the other hand, duopoly profits were higher, then the entrant would tend to win.”
Dr. David L. Rubinfeld, Professor of Law and Professor of Economics Emeritus at the University of California, Berkeley and Professor of Law Emeritus at New York University School of Law, first noted that even in light of the Supreme Court’s opinion in Ohio v. American Express,[7] Microsoft would still be adjudicated in the same way, because unlike American Express, there are not simultaneous transactions in the two different markets. Rubinfeld reflected upon his time as Chief Economist at the Department of Justice, Antitrust Division in 1998-99, noting that it was only because Microsoft agreed to waive a conflict that he was able to work on the Microsoft case. Once that occurred, Rubinfeld began meeting with market participants, which he characterized as “working like in a spy network, because none of the folks I met with wanted to be known publicly.” Ultimately, those meetings helped him to determine that the government “had to pursue a Section Two case” that would focus on “whether there was monopoly power. And the key to that was how to deal with barriers to entry.” This resulted in a focus on the “applications barriers to entry”. Soon, “Microsoft was talking about it in their defense, the judge started to use the term, and it became a popular term during the trial.”
Professor Erik Hovenkamp of Cornell Law School argued that the Microsoft opinion “is really a model of how courts should account for the economics of platforms” because it “avoids a number of pitfalls” that have been seen in subsequent cases. This included “changing things up in an extreme and unnecessary way” and “evaluating platform conduct using . . . old formalistic doctrine that is really divorced from economics.”
In the former category, Professor Hovenkamp discussed American Express. He noted that the district court in Microsoft “was actually very explicit that you need to consider both sides” of the market. But it did not reach the conclusion that “if one side is harmed and the other side is benefited . . . you have to show a net harm”. Hovenkamp contrasted American Express’s two-sided market analysis with the approach taken toward resale price maintenance, where the practice is justified in an expected promotion of sales of the good through increased incentives for retailers to carry and market the product. In that context, he noted, “nobody says ‘Well, we need to define the market to include consumers and retailers’.”
Hovenkamp also stressed that Microsoft avoided the formalism of recent platform cases. He contrasted it with the recent decision in FTC v. Meta which considered Meta’s conduct with reference to Aspen Skiing Company v. Aspen Highlands Skiing Corporation.[8] Hovenkamp cited the court’s conclusion that “[i]n considering whether the monopolist’s conduct harms competition, our focus is upon the effect of that conduct and not upon the intent behind it.”[9] Other courts’ use of Aspen Skiing, in contrast, adopts a “sort of old fashioned test that’s just … completely mapped onto the facts of Aspen and has very little to do with economics.”
Michal Gal, Professor at the Faculty of Law, University of Haifa and Raz Agranat, JSD Candidate at the University of Chicago presented a paper focusing on Microsoft’s treatment of network effects. Professor Gal noted that Microsoft’s assessment of network effects concluded that in industries characterized by network effects competition is “‘for the field’ rather than ‘within the field’”.[10] She highlighted that this assumption was simplistic, because it presumes “that all platform participants are comparably valuable to the platform’s network effects and its integrity … [which is] not true for most networks.” Agranat underscored that this is because “empirical observation reveals that many networks . . . have hubs,” nodes with orders of magnitude more connections within a network. For example, Cristiano Ronaldo’s Facebook account has more than 170 million followers. Removal of such hubs can have an outsized impact on a network and such hubs have the ability to push back on abuses by a platform. To that end, Gal and Agranat proposed to “revise the Microsoft formula to account for hubs” and highlighted that various legal mechanisms could do so.
* Timothy Kearns is a Partner in Washington, D.C. and the Managing Editor of the Competition Bulletin.
Footnotes
[1] Except where otherwise cited, all quotations are from the statements made by panelists at the event.
[2] United States v. Microsoft Corp., 87 F.Supp.2d 30, 42 (D.D.C. 2000).
[3] United States v. Microsoft Corp., 254 F.3d 34, 58-59 (D.C. Cir. 2001) (emphasis added).
[4] Id. at 59 (emphasis added).
[5] Id. (emphasis added).
[6] Id.
[7] Ohio v. Am. Express Co., 138 S. Ct. 2274 (2018).
[8] See Fed. Trade Comm'n v. Facebook, Inc., 560 F. Supp. 3d 1, 22 (D.D.C. 2021).
[9] United States v. Microsoft Corp., 254 F.3d 34, 59 (D.C. Cir. 2001).
[10] Id. at 49.
