FOR PUBLICATION

UNITED STATES COURT OF APPEALS

FOR THE NINTH CIRCUIT

FEDERAL TRADE COMMISSION, No. 23-15992

Plaintiff-Appellant, D.C. No. 3:23-cv- 02880-JSC v.

MICROSOFT CORPORATION; OPINION

ACTIVISION BLIZZARD, INC.,

Defendants-Appellees.

Appeal from the United States District Court for the Northern District of California Jacqueline Scott Corley, District Judge, Presiding

Argued and Submitted December 6, 2023 San Francisco, California

Filed May 7, 2025

Before: Daniel P. Collins, Danielle J. Forrest, and Jennifer Sung, Circuit Judges.

Opinion by Judge Collins SUMMARY *

Clayton Act

The panel affirmed the district court’s denial of a motion by the Federal Trade Commission (“FTC”) for preliminary injunctive relief against Microsoft’s acquisition of the video game developer Activision Blizzard, Inc. The merger is the subject of an administrative proceeding that remains pending before the FTC. In its administrative complaint and in seeking a preliminary injunction in the district court, the FTC asserted that the merger would likely violate § 7 of the Clayton Act because, viewing the merger as a vertical integration between a content-platform operator and a content producer, competition would be substantially lessened in the relevant U.S.-based content-platform markets for gaming console devices, gaming subscription services, and gaming cloud- streaming services. The panel held that the district court applied the correct legal standards and did not abuse its discretion, or rely on clearly erroneous findings, in holding that the FTC failed to make a sufficient evidentiary showing to establish the requisite likelihood of success on the merits of its § 7 claim. Thus, the FTC had not raised serious questions regarding whether the proposed merger was likely to substantially lessen competition in the relevant markets. First, pertaining to the console market, the panel agreed with the district court that the FTC failed to sufficiently show * This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader.

FTC V. MICROSOFT CORP. 3

that Microsoft would foreclose or partially foreclose rivals after the merger either by making the popular game Call of Duty exclusive to its Xbox console or by releasing only an inferior version of the game for Sony’s rival PlayStation. The panel next found that as to the library subscription services market, the district court did not abuse its discretion by holding that the FTC had not made an adequate showing that the merger would substantially lessen competition. Because Activision Blizzard had long opposed putting its content on library subscription services, the merger’s effect of making such content available for the first time in the subscription market, even if exclusive to Microsoft, would not substantially lessen competition. Finally, the district court did not abuse its discretion in similarly finding an insufficient likelihood of success on the FTC’s claim that the merger would substantially lessen competition in the cloud-streaming market, given that the FTC failed to show that Activision Blizzard content would be available to this market in the absence of the merger. Imad D. Abyad (argued) and Mark S. Hegedus, Attorneys; Mariel Goetz, Acting Director of Litigation; Anisha S. Dasgupta, General Counsel; Office of the General Counsel; James H. Weingarten, Peggy B. Femenella, James Abell, Cem Akleman, Meredith R. Levert, and Jennifer Fleury, Attorneys; Shaoul Sussman, Associate Director for Litigation; John Newman, Deputy Director; Henry Liu and Holly Vedova, Directors; Bureau of Competition; Federal Trade Commission, Washington, D.C., for Plaintiff- Appellant. Rakesh N. Kilaru (argued), Jennifer H. Pavelec, Anastasia M. Pastan, and Beth A. Wilkinson, Wilkinson Stekloff LLP, Washington, D.C.; Adam B. Banks, Weil Gotshal & Manges, New York, New York; C. Frederick Beckner III, Daniel J. Hay, William R. Levi, Lucas Croslow, and Jonathan E. Nuechterlein, Sidley Austin LLP, Washington, D.C.; Evan Kreiner, Michael Sheerin, Steven C. Sunshine, and Julia K. York, Skadden Arps Slate Meagher & Flom LLP, Washington, D.C.; Grant Dixton, Activision Blizzard Inc., Santa Monica, California; for Defendants-Appellees. Jennifer E. Sturiale, Delaware Law School, Widener University, Wilmingtion, Delaware, for Amicus Curiae Law Professors. Jeane A. Thomas and Ann L. Rives, Crowell & Moring LLP, Washington, D.C., for Amici Curiae Former State Attorneys General. Maureen K. Ohlhausen and Stacy L. Turner, Baker Botts LLP, Washington, D.C., for Amicus Curiae Business Roundtable.

FTC V. MICROSOFT CORP. 5

Jeffrey M. Harris and Frank H. Chang, Consovoy McCarthy PLLC, Arlington, Virginia; Tyler S. Badgley and Maria C. Monaghan, United States Chamber Litigation Center, Washington, D.C.; for Amicus Curiae The Chamber of Commerce of the United States of the America. Allen P. Grunes, David B. Meschke, and Rosa L. Baum, Brownstein Hyatt Farber Schreck LLP, Washington, D.C.; Matthew J. Ginsburg, AFL-CIO, Washington, D.C.; for Amici Curiae Communications Workers of America and the American Federation of Labor and Congress of Industrial Organizations. Melissa A. Sherry, Jonathan C. Su, and Brent T. Murphy, Latham & Watkins LLP, Washington, D.C., for Amici Curiae Independent Game Publishers and Developers. Edward D. Hassi, Debevoise & Plimpton LLP, Washington, D.C.; J. Robert Abraham, Isabel Feldman, and Jaime M. Freilich-Fried, Debevoise & Plimpton LLP, New York, New York; for Amici Curiae Venture Capital Firms. Aaron M. Panner and Derek C. Reinbold, Kellogg Hansen Todd Figel & Frederick PLLC, Washington, D.C., for Amici Curiae Former Antitrust Enforcers. Dana Foster and Soraya Todd, White & Case LLP, Washington, D.C.; Jack E. Pace III, White & Case LLP, New York, New York; for Amici Curiae Economists and Antitrust Scholars. Bilal Sayyed and Andy Jung, TechFreedom, Washington, D.C., for Amicus Curiae TechFreedom. COLLINS, Circuit Judge:

The Federal Trade Commission (“FTC”) appeals the district court’s denial of its motion for preliminary injunctive relief against Microsoft Corporation’s acquisition of the video game developer Activision Blizzard, Inc. The merger is the subject of an administrative proceeding that remains pending before the FTC. In its administrative complaint, and in seeking a preliminary injunction in the district court, the FTC asserted that the merger would likely violate § 7 of the Clayton Act, 15 U.S.C. § 18, by substantially lessening competition in various relevant markets. Specifically, viewing the merger as a vertical integration between a content-platform operator and a content producer, the FTC asserted below that competition in what it contended were the relevant U.S.-based content-platform markets (i.e., the markets for gaming console devices, gaming subscription services, and gaming cloud-streaming services) would be substantially lessened. The FTC argued that, under the more lenient standards this court applies to preliminary injunctions sought under § 13(b) of the Federal Trade Commission Act (“FTC Act”), 15 U.S.C. § 53(b), the FTC made an adequate showing of likelihood of success and that the balance of equities favored enjoining the merger. After a lengthy evidentiary hearing, the district court disagreed and denied the preliminary injunction in a detailed opinion. The FTC immediately filed this appeal, and a panel of this court denied the FTC’s emergency request for an injunction pending appeal. The merger was subsequently completed shortly after the FTC’s reply brief was filed in this court.

FTC V. MICROSOFT CORP. 7

We conclude that the district court applied the correct legal standards and that it did not abuse its discretion, or rely on clearly erroneous findings, in holding that the FTC had failed to make a sufficient evidentiary showing to establish the requisite likelihood of success on the merits of its § 7 claim. We therefore affirm. I A Playing video games has become extraordinarily popular, with an estimated three billion or more persons throughout the world regularly playing single-player and multiplayer games. The companies satisfying this demand for gaming include the developers who produce such games and the manufacturers who provide the platforms on which they are played. Many companies perform more than one of these tasks—for example, Microsoft Corporation (“Microsoft”) manufactures physical video game consoles (e.g., the “Xbox” console) that can play a variety of games that are loaded into them, and Microsoft also develops and publishes some of its own video games (e.g., Halo and Forza). Likewise, Nintendo Co. Ltd. (“Nintendo”) makes the “Nintendo Switch” game console, and Nintendo is also the first-party developer and publisher of the Mario and Pokémon game franchises. And Sony Interactive Entertainment (“Sony”) manufactures the “PlayStation” gaming console and also publishes games such as God of War and Spider-Man. Other companies, such as Activision Blizzard Inc. (“Activision Blizzard”), develop and publish games (such as Call of Duty) but do not manufacture the devices on which those games would be played. Games developed by device manufacturers such as Microsoft, Sony, and Nintendo are sometimes referred to as “first-party” games, while games produced by independent developers such as Activision Blizzard are called “third-party” games. As noted, one of the ways in which video games can be played is by using a physical console that is “designed for, and whose primary use is, to play video games.” At present, there are three main manufacturers of gaming consoles, namely, Microsoft (with its Xbox), Nintendo (with its Switch), and Sony (with its PlayStation). Video games can also be played on personal computers (“PCs”) or on mobile devices (such as tablets or smart phones), but more sophisticated games may require either consoles or “gaming PCs.” Microsoft introduced its Xbox in 2001, thereby competing with then-established market participants Sony and Nintendo. Over the years, the three major console manufacturers released successive generations of their consoles, with different manufacturers coming out on top across the competing generations. For example, in the United States market, Microsoft’s Xbox 360 outsold Sony’s PlayStation 3, but Sony won the next generation, with the PlayStation 4 outselling the Xbox One. Currently, the Xbox Series X and the PlayStation 5 have competed since they were both released by their respective manufacturers in November 2020. For the current generation, Xbox ranks third behind PlayStation and Nintendo Switch. In recent years, however, consoles have receded in overall popularity among gamers. Today, more than half of gamers play on mobile devices, with PCs being the next most popular option, ahead of consoles. To varying degrees, the major console manufacturers have used exclusive content as a means to differentiate themselves in the console market. Some of this exclusivity

FTC V. MICROSOFT CORP. 9

is achieved by limiting the availability of a manufacturer’s first-party games to its own console. All major manufacturers have engaged in this practice. Microsoft has in recent years released its first-party games exclusively on Xbox and PCs, most of which use Microsoft’s Windows operating system. As the district court found, however, Nintendo and Sony “both have significantly higher number of exclusive games on their platform than [Microsoft] does.” In particular, the court found that there are approximately “eight exclusive games on [Sony’s] PlayStation for every one on Xbox.” Sony has also made deals with independent third-party game publishers to get “timed exclusivity,” whereby a game would launch first on PlayStation before being released on other platforms. Sony has also paid third- party game developers to skip releasing particular games on Xbox altogether. For example, after Sony had paid for platform exclusivity, a third-party developer released Final Fantasy XVI exclusively on PlayStation 5, leaving Xbox with only older versions of Final Fantasy. Over time, the means by which gamers obtain games to be played on their devices has changed. While it was once common for gamers to purchase or rent a physical cartridge, DVD, or disc to play games, most games today are distributed digitally onto the device. Although some games can be played for free, a physical copy or downloaded digital copy of a single standard title normally costs about $70. However, many gamers today rely on digital subscription services rather than the prior “‘buy-to-play’ model of purchasing the games.” For example, Microsoft launched its subscription service, Xbox Game Pass, in 2017. For a flat monthly fee, Game Pass gives subscribers access on their Xbox console to a large rotating catalog of video games, including Microsoft’s first-party content. Microsoft’s CEO has described Game Pass as “Netflix for Games.” In 2019, Microsoft made Game Pass available on PCs, thereby allowing gamers to access Game Pass without purchasing an Xbox. Microsoft also offers a higher-tier service called Game Pass Ultimate. Microsoft has generally made all of its first-party content immediately available to Game Pass subscribers on the same day it is released for individual purchase. Although Microsoft thereby loses out on some sales of individual copies that might otherwise have been purchased by subscribers—a phenomenon known as “cannibalization”—any such losses are offset by the fact that, overall, Game Pass subscribers spend more time and consequently more money on games compared to non- subscribers. Sony offers a competing subscription service with two main tiers—namely, PlayStation Plus Extra and PlayStation Plus Premium. However, unlike Microsoft, Sony does not release its games on PlayStation Plus on the same day that they are released for individual purchase. Other participants in the market for subscription services include Amazon (which offers Luna+), Electronic Arts (which offers EA Play), and Ubisoft. In late 2020, Microsoft reached an agreement with Electronic Arts to include access to EA Play in Game Pass Ultimate. Game Pass Ultimate also includes access to several Ubisoft games. Another way in which gamers obtain access to games is through “cloud gaming.” In cloud gaming, the game is run on remote servers and streamed to the gamer on his or her device. One of the primary advantages of cloud gaming is that it allows players “to play games on less highly-powered and more affordable devices.” While some cloud gaming

FTC V. MICROSOFT CORP. 11

services, such as Microsoft’s xCloud, offer the ability to play games from a content library, others, such as Nvidia’s GeForce Now, use a so-called “bring-your-own-game” model (“BYOG”). In the BYOG model, “users stream individual games that they already own.” The major competitors in cloud gaming are Microsoft’s xCloud, PlayStation Plus Premium, Nvidia’s GeForce Now, and Amazon’s Luna+ and Prime Gaming. At present, Microsoft bundles Game Pass and xCloud, meaning that Game Pass Ultimate subscribers receive access to xCloud as part of their subscription and that it is not possible to use xCloud without subscribing to Game Pass Ultimate. But even as they have become paired with cloud gaming, both Microsoft’s Game Pass Ultimate and Sony’s PlayStation Plus Premium (Sony’s analogous subscription tier) remain available on console and PC. B As for independent game developers, they earn revenue in primarily two ways. First, they can sell copies of their games. When a developer sells a game suitable for use on a particular platform, the developer and the platform owner will generally split the revenues from the sale (sometimes referred to as a “royalty split”). Ordinarily, the publisher receives 70% of the revenue, and the platform operator receives 30%. Second, developers can sell content within the games (i.e., in-game microtransactions), which is most popular with mobile gaming and free-to-play titles, such as Overwatch 2 or Fortnite. As noted earlier, video games can be multiplayer or single-player. In single-player games, the gamer plays through the game’s built-in narrative, interacting with “non- player characters” as the gamer progresses. In multiplayer games, by contrast, gamers play with others simultaneously, usually through an online connection. Because multiplayer games involve humans playing against one another, they have an important social component, thereby deepening gamers’ connections with each other and the game. Video games can also have both single-player and multiplayer modes; for instance, Call of Duty offers a popular online multiplayer component, as well as a single-player option. A limited number of multiplayer games also have so-called “cross play,” in which gamers on different platforms can play online with gamers on other platforms. Of particular importance in the game-development industry are the high-quality games known as “AAA” games. Although the industry has no precise definition of the term, the “AAA” moniker generally refers to games developed at considerable expense to provide a technically sophisticated experience with “cinematic storytelling, immersive environments, and detailed graphics.” Because of their technical and narrative complexity, AAA games take a long time to develop, and only a limited supply of approximately 10 to 20 AAA games are released each year. And only a handful of game publishers have the resources to produce multiple AAA games, namely, the so-called “Big 4”—Activision Blizzard, Electronic Arts, Take-Two, and Ubisoft. While these are not the only companies capable of producing AAA games, the Big 4 each offer a suite of AAA games, and they have accounted for a substantial volume of the game sales on Xbox and PlayStation consoles for many years. As a video game executive put the point, “[a]ccess to AAA titles . . . is critical to the success of any gaming platform.” As one of the Big 4, Activision Blizzard is one of the largest game developers in the world. Activision has three

FTC V. MICROSOFT CORP. 13

divisions (Activision, Blizzard, and King), each with respective marquee franchises—respectively, Call of Duty, World of Warcraft, and Candy Crush. These three game franchises generated 80% of Activision Blizzard’s 2022 revenue. Other Activision Blizzard game series include Diablo, Hearthstone, Overwatch, and StarCraft, each with over $1 billion in lifetime revenue. As of December 2022, Activision Blizzard had more than 380 million monthly active users across all of its games. Activision Blizzard’s success is driven in large part by Call of Duty, a AAA game and one of the most popular video game franchises of all time. The Call of Duty franchise has approximately 100 million monthly active users, of which roughly half play on mobile devices. On any given day, between 7 and 10 million people play Call of Duty, according to Activision Blizzard’s CEO. Because of its widespread popularity, Call of Duty has generated a sizable proportion of Activision Blizzard’s total net revenue of $7.5 billion in 2022. In the United States, a Call of Duty game has been the top selling console game every year but one since 2014. The Call of Duty series is so popular that, in 2020, different versions of Call of Duty were both the first and second best- selling console games in the United States, and in 2021, they were first and third. The district court found that, “with the exception of sports games,” Call of Duty is “unique among AAA games” in that a new Call of Duty title is typically released every year. In addition to its annual releases, the Call of Duty franchise also includes the free-to-play Call of Duty Warzone, a multiplayer online game that has over 100 million downloads and that generates revenue through in- game microtransactions. As one of the largest gaming franchises, Call of Duty has been important to Sony. Since 2019, tens of millions of unique PlayStation users have played Call of Duty, representing a significant percentage of all PlayStation users and accounting for a substantial portion of Sony’s overall revenue. This dedicated fan base also spends a substantial amount of time on PlayStation playing Call of Duty. As of the time of the district court’s ruling in this case, Call of Duty was not available on the Nintendo Switch and was not available on any gaming subscription service or on any cloud gaming service. Activision Blizzard has other popular AAA franchises in addition to Call of Duty. For example, its Blizzard division is known for the World of Warcraft franchise, which consists of a multiplayer online roleplaying game. The World of Warcraft franchise also includes the popular free-to-play game Hearthstone. Among Blizzard’s other AAA games are the Diablo franchise and Overwatch 2, both of which have generated substantial revenue. Activision Blizzard also owns a number of other popular yet dormant franchises, including Crash Bandicoot and Tony Hawk’s skating games. Activision Blizzard also has a presence in mobile gaming, as it owns Call of Duty: Mobile, and King, the creator of Candy Crush. C On January 18, 2022, Microsoft announced that it would acquire Activision Blizzard for $68.7 billion. Pursuant to the Hart-Scott-Rodino Antitrust Improvements Act, see 15 U.S.C. § 18a, Microsoft reported the planned merger to the FTC on February 1. The FTC then began a lengthy and thorough investigation involving the production of nearly three million documents and 15 investigational hearings. On

FTC V. MICROSOFT CORP. 15

December 8, 2022, the FTC filed an administrative complaint against the merger. Shortly after the FTC filed its complaint, Microsoft entered into binding agreements with console and cloud gaming competitors to ameliorate the concerns of antitrust regulators. In February 2023, Microsoft signed a ten-year agreement with Nintendo to bring future Call of Duty titles to Nintendo consoles simultaneously with their release on Microsoft platforms. Thereafter, Microsoft also entered into ten-year agreements with five cloud gaming companies, bringing Activision Blizzard content to platforms where it had previously been absent. Microsoft also made repeated offers to Sony to keep Call of Duty on PlayStation for at least ten years, alongside public commitments to the same effect. After this appeal was filed, Sony accepted Microsoft’s offer. While the administrative proceeding was ongoing, the FTC on June 12, 2023 sought a preliminary injunction in the district court under § 13(b) of the FTC Act, 15 U.S.C. § 53(b). The district court held a five-day evidentiary hearing on an expedited basis, given that the merger was set to close on July 18, 2023. On July 10, the district court denied the preliminary injunction, finding that the FTC had “not raised serious questions regarding whether the proposed merger is likely to substantially lessen competition” in the relevant markets. The FTC filed an emergency motion for an injunction pending appeal, and a panel of this court denied that motion on July 14, 2023. Simultaneously with the U.S. antitrust action, the United Kingdom’s Competition and Markets Authority (“CMA”) was also reviewing the merger. The CMA’s Final Report concluded that, following the merger, “Microsoft would have the ability and incentive to use Activision [Blizzard]’s content to foreclose current and future rival cloud gaming service platforms and that, as a result, [the merger] may be expected to result in [a substantial lessening of competition] in cloud gaming services in the UK.” On August 22, 2023, the CMA issued its final order prohibiting the merger. However, in response to concessions by Microsoft with respect to streaming rights for Activision Blizzard content, the CMA reversed course and granted final approval to the merger on October 13, 2023. In connection with that approval, Activision Blizzard agreed to divest, to Ubisoft, its cloud-streaming rights outside of the European Economic Area (“EEA”) to all current Activision Blizzard games and to all future games released within the next 15 years. 1 As a result, Ubisoft, rather than Microsoft or Activision Blizzard, will control which cloud service or services in the U.S. will have Activision Blizzard games. Moreover, Ubisoft “will not be authorised to license Cloud Streaming Rights to Microsoft or its affiliates on an exclusive basis.” Additionally, any non-exclusive license to Microsoft cannot give Microsoft preferential pricing or provide it with “material preferential treatment.” As part of the arrangement with Ubisoft, Microsoft is required “to provide Ubisoft with versions of Activision [Blizzard] games that are, with respect to ‘quality, content, features and performance[,] . . . the same in all material respects to the non-streaming version[s] of such games.’” The merger closed on the same day the CMA approved it, i.e., October 13, 2023.

1 Within the EEA, Microsoft will retain cloud streaming rights to Activision Blizzard games “to comply with its regulatory commitments to the European Commission.”

FTC V. MICROSOFT CORP. 17

II “The denial of a motion for preliminary injunction will be reversed only if the district court abused its discretion or based its decision on an erroneous legal premise.” FTC v. Warner Commc’ns Inc., 742 F.2d 1156, 1160 (9th Cir. 1984). While the district court’s ultimate decision to deny a preliminary injunction is thus reviewed for abuse of discretion, we review the district court’s legal conclusions de novo and its factual findings for clear error. K.W. ex rel. D.W. v. Armstrong, 789 F.3d 962, 969 (9th Cir. 2015). The FTC’s underlying claim in the administrative proceedings is that the merger of Microsoft and Activision Blizzard violates § 7 of the Clayton Act, 15 U.S.C. § 18. See Clayton Act § 11, 15 U.S.C. § 21 (granting authority to the FTC, subject to certain exceptions, to directly enforce § 7 of the Clayton Act in administrative proceedings). Section 7 prohibits mergers and acquisitions “where in any line of commerce or in any activity affecting commerce in any section of the country, the effect of such acquisition may be substantially to lessen competition, or to tend to create a monopoly.” Id. § 18. The statute is prospective, “requir[ing] not merely an appraisal of the immediate impact of the merger upon competition, but a prediction of its impact upon competitive conditions in the future; this is what is meant when it is said that . . . § 7 was intended to arrest anticompetitive tendencies in their incipiency.” St. Alphonsus Med. Ctr.-Nampa Inc. v. St. Luke’s Health Sys., Ltd., 778 F.3d 775, 783 (9th Cir. 2015) (citation omitted). Because a merger’s effects cannot be predicted with certainty, the FTC need only show a “reasonable probability that the merger will substantially lessen competition” in any relevant market to prevail on the merits of an underlying § 7 claim. Brown Shoe Co. v. United States, 370 U.S. 294, 325 (1962); see also Warner, 742 F.2d at 1160 (“It is well established that a section 7 violation is proven upon a showing of reasonable probability of anticompetitive effect.”). In addition to its administrative authorities, the FTC is also authorized, under § 13(b) of the FTC Act, to file suit in a federal district court seeking to preliminarily enjoin any actual or imminent violation of “any provision of law enforced by the Federal Trade Commission.” 15 U.S.C. § 53(b). “Upon a proper showing that, weighing the equities and considering the Commission’s likelihood of ultimate success, such action would be in the public interest,” a district court may grant a preliminary injunction. Id. We have held that § 13(b) “places a lighter burden on the Commission than that imposed on private litigants by the traditional equity standard” inasmuch as “the Commission need not show irreparable harm to obtain a preliminary injunction.” Warner, 742 F.2d at 1159. The inquiry under § 13(b) thus focuses on (1) “the likelihood that the Commission will ultimately succeed on the merits”; and (2) the “balance [of] the equities.” Id. at 1160. The district court concluded that both of these factors weighed against issuing a preliminary injunction. As we explain in the ensuing sections, we affirm based solely upon the likelihood-of-success factor.

III

In addressing the likelihood-of-success factor under § 13(b) of the FTC Act, we have stated that the FTC “meets its burden if it ‘raise[s] questions going to the merits so serious, substantial, difficult and doubtful as to make them fair ground for thorough investigation, study, deliberation and determination by the FTC in the first instance and

FTC V. MICROSOFT CORP. 19

ultimately by the Court of Appeals.’” Warner, 742 F.2d at 1162 (quoting FTC v. National Tea Co., 603 F.2d 694, 698 (8th Cir. 1979)) (alteration in original). The question, then, is whether the district court abused its discretion in concluding that the FTC had failed to raise sufficiently serious and substantial questions on the merits of its Clayton Act § 7 claim to support preliminary injunctive relief. Viewed through the lens of the FTC’s burden under § 7, the question under § 13(b) is whether the FTC’s evidentiary showing raised sufficiently serious and substantial questions as to a “reasonable probability that the merger will substantially lessen competition” in any relevant market. Brown Shoe, 370 U.S. at 325; see also United States v. Anthem, Inc., 855 F.3d 345, 368 (D.C. Cir. 2017) (noting that a finding of substantial anticompetitive effects in any one market “provides an independent basis for the injunction”). Here, the district court either agreed with, or assumed arguendo the correctness of, the FTC’s contentions as to the relevant product and geographic markets. Specifically, the district court agreed with the FTC’s definition of the “primary market” as the “high-performance console market,” and the court also accepted, for purposes of the preliminary injunction inquiry, the FTC’s assertion that Nintendo’s Switch was too different from the Xbox and PlayStation to be included in this market. The district court further assumed, “without deciding,” that “the FTC’s additional markets of the multigame content library subscription services [market] and [the] cloud gaming [market]” were “each their own product market when considered singly or in combination.” As to the geographic scope of the relevant product markets, the district court agreed with the FTC that the relevant geographic market for high-performance consoles is the United States, and the court “assume[d] without deciding” that the United States is also the relevant geographic market for “multigame content library subscription services and cloud gaming.” The district court ultimately held, however, that the FTC had “not raised serious questions regarding whether the proposed merger is likely to substantially lessen competition in the console, library subscription services, or cloud gaming markets.” At the outset, the FTC points to various phrases used in the district court’s opinion, and it argues that those phrases confirm that the district court fundamentally misunderstood the scope of the inquiry in a § 13(b) action seeking a preliminary injunction against an asserted § 7 violation. According to the FTC, rather than focus only on whether the FTC had raised “serious questions” about whether there was a “‘reasonable likelihood’ of a substantial lessening of competition in a relevant market,” the district court instead required the FTC to prove the underlying merits of its § 7 claim—i.e., that competition “would probably be substantially lessened” (emphasis altered). We reject this contention. As the FTC concedes, the district court preceded its substantive discussion of the FTC’s likelihood of success with a recitation of the “proper Section 13(b) standard,” which is that the FTC’s burden is to “raise[] questions going to the merits so serious, substantial, difficult and doubtful as to make them fair ground for thorough investigation, study, deliberation and determination by the FTC in the first instance and ultimately by the Court of Appeals.” Warner, 742 F.2d at 1162. At the end of its analysis of the likelihood- of-success factor, the district court framed its conclusion by again using Warner’s language in stating that “the FTC has not raised serious questions regarding whether the proposed merger is likely to substantially lessen competition” in one

FTC V. MICROSOFT CORP. 21

of the relevant markets. Despite the fact that the district court thus expressly framed its analysis, both at the beginning and the end, in terms of the correct § 13(b) “serious questions” preliminary-injunction standard, the FTC points to other sentences in the district court’s opinion that discuss the likelihood of success on the merits without repeating the “serious questions” phrase. But the fact that the district court did not repeat this phrase, or some equivalent, every time it made an observation about the FTC’s showing on the substantive merits of its § 7 claim does not mean that the district court thereby ignored the overlay that § 13(b) provides in the context of a preliminary injunction motion. 2 Our task is not to flyspeck, out-of- context, isolated phrases in a comprehensive opinion that was issued only four weeks after the FTC filed its time- sensitive emergency motion and that resolves highly complex issues against the backdrop of a voluminous factual record. Rather, in assessing whether the district court applied the wrong legal standards, we review that order as a whole, and in context. Viewing the order that way, we are confident that the district court adhered to, and applied, the Warner standard. The FTC nonetheless argues that the district court departed from the Warner standard because the court ruled against the FTC even though the court acknowledged that “at best ‘the record contains conflicting evidence on the anticompetitive effects of the merger.’” According to the

2 In this opinion, we too will not repeatedly use, in every merits-related statement, the cumbersome phrasing that would more precisely capture the relevant application of § 13(b)’s “serious questions” standard under Warner and the “reasonable probability” standard applicable to the underlying merits under § 7. Our analysis, however, must be understood as staying within the applicable legal framework that we have set forth. FTC, that was error because, once the district court identified “conflicting evidence,” it was bound to find serious questions going to the merits and was therefore required to hold that the FTC met its burden of showing the requisite likelihood of success. The district court thus committed legal error, the FTC argues, by “resolv[ing] evidentiary conflicts” based “on a preliminary record.” This argument profoundly misconceives the applicable § 13(b) standards under Warner. The FTC relies on Warner’s statement that, when this court reviews the denial of a preliminary injunction, “we do not resolve the conflicts in the evidence,” but merely assess whether the government has presented “evidence sufficient to raise ‘serious, substantial, difficult’ questions regarding the anticompetitive effects of the proposed joint venture.” Warner, 742 F.2d at 1164 (citation omitted). But we did not thereby suggest that no factual findings may be made in the course of deciding a preliminary injunction motion under § 13(b). Rather, as an earlier comment in the opinion in Warner made clear, we should not purport to “make a final determination on whether the proposed merger violates Section 7, but rather to make only a preliminary assessment of the merger’s impact on competition.” Id. at 1162 (emphasis added). That “preliminary assessment”—i.e., whether the FTC has raised “serious questions” concerning the merits of its § 7 claim—may properly rest upon pertinent factual findings bearing upon whether that showing has been made. We acknowledged as much in Warner, because we recognized that we ordinarily must “accord the usual deference to the district court’s findings regarding relevant market, market concentration and barriers to entry,” and that we were relieved of that obligation in Warner only because the district court’s findings in that case “were improperly

FTC V. MICROSOFT CORP. 23

based” on materials that the court should not have considered. Id. (emphasis added); see also FTC v. Affordable Media, 179 F.3d 1228, 1233 (9th Cir. 1999) (holding that the “clearly erroneous” standard applies to review of a district court’s “factual findings” in a decision granting a § 13(b) preliminary injunction). Just because we concluded that the district court’s findings in Warner were flawed does not mean that it is categorically inappropriate for district courts to make factual findings in all other cases. Indeed, the FTC’s position—viz., that every factual dispute should be resolved in its favor when it requests a preliminary injunction under § 13(b)—ignores the settled principle that a preliminary injunction remains “an extraordinary and drastic remedy” that must be affirmatively justified by the FTC. FTC v. Exxon Corp., 636 F.2d 1336, 1343 (D.C. Cir. 1980) (citation omitted). The FTC’s proposed construe-everything-my-way standard is more suited for defending against a summary-judgment dismissal of claims than it is for obtaining provisional affirmative injunctive relief. IV We turn, then, to whether the district court abused its discretion, or relied on clearly erroneous factual findings, in concluding that the FTC had “not raised serious questions” going to the merits of its § 7 claim. Although Microsoft contends that the district court’s market definitions were flawed in certain respects, it also argues that, even accepting these definitions arguendo, the district court correctly concluded that the FTC’s showing as to a likelihood of success on the merits was deficient as to each such market. Because we ultimately agree with the latter argument, we have no occasion to consider whether the district court’s definitions of the relevant markets were correct, and we instead assume arguendo that they were. We therefore address, with respect to each relevant market, whether the district court abused its discretion in concluding that the FTC made an insufficient showing on the merits of its § 7 claim. A At the evidentiary hearing in the district court, the FTC’s primary focus was on the high-performance console market. More specifically, the FTC’s main theory was that, in light of the enormous popularity of Call of Duty, Microsoft would be expected to make it exclusive to Xbox after the merger, thereby causing gamers to defect from PlayStation to Xbox and substantially lessening competition in the console market. According to the FTC, Microsoft’s incentive lies in the fact that such an exclusivity arrangement would lead to increased sales of Xbox consoles and associated derivative revenue that would well make up for any loss on Call of Duty sales to PlayStation users. Because “the diminution of the vigor of competition which may stem from a vertical arrangement results primarily from a foreclosure of a share of the market otherwise open to competitors,” Brown Shoe, 370 U.S. at 328, the FTC argues that excluding Call of Duty from PlayStation would substantially lessen competition by “leav[ing] consumers with fewer or worse options in the console market.” After an extensive evidentiary record was developed, including the testimony of several witnesses at a five-day evidentiary hearing, the district court concluded that the FTC had failed to make a sufficient showing to support a preliminary injunction on the agency’s theory that Microsoft would “foreclose” rivals by making Call of Duty exclusive

FTC V. MICROSOFT CORP. 25

to Xbox. The district court acknowledged that Microsoft would obviously have the ability to foreclose rivals in that, after the merger, it would own and control the rights to Call of Duty. But the court was not persuaded that, taking into account the likelihood-of-success standard under § 13(b), the FTC had sufficiently shown that Microsoft had the incentive to foreclose with respect to Call of Duty and that there was a reasonable possibility that Microsoft might do so. We discern no abuse of discretion in that conclusion and no clear error in the findings that underlie it. In particular, the district court found that Microsoft would be highly unlikely to withdraw Call of Duty from PlayStation, given that “Call of Duty’s cross-platform play is critical to its financial success.” As explained earlier, Call of Duty has a very popular multiplayer component, which allows gamers to play with others across devices. Removing Call of Duty from PlayStation would destroy the communities of players that have built up around the multiplayer aspect, particularly given the undisputed evidence that there are significantly more Call of Duty players on PlayStation than on Xbox. Indeed, at the hearing, the CEO of Activision Blizzard testified that the company’s Call of Duty revenues from PlayStation “are probably twice the Xbox revenues.” The district court also noted that, in addition to losing very substantial revenue from such PlayStation gamers, Microsoft would be expected to experience serious “reputational harm” if it pulled Call of Duty from PlayStation and thereby blocked millions of PlayStation gamers’ access to the game. Moreover, the district court emphasized that the FTC had “not identified any instance in which an established multiplayer, multi-platform game with cross-play . . . has been withdrawn from millions of gamers and made exclusive.” In this respect, the district court considered the evidence concerning Microsoft’s prior acquisitions of two game publishers, Mojang and ZeniMax. With respect to Microsoft’s acquisition of ZeniMax, the FTC pointed out that, notwithstanding Microsoft’s reassurances to regulators that it would have strong incentives to keep ZeniMax content on other platforms, after the merger, “Microsoft made future ZeniMax content—including AAA titles like Starfield, Redfall, and Elder Scrolls VI—exclusive to its platforms.” However, the district court permissibly concluded that the FTC’s reliance on the ZeniMax acquisition was inapt, because Microsoft’s exclusionary behavior regarding the post-merger ZeniMax games did not involve withdrawing existing multiplayer, cross-platform games from PlayStation. The much more pertinent example, the district court held, was Microsoft’s treatment of Minecraft after acquiring its publisher, Mojang. Minecraft “includes a popular multiplayer mode and has produced a large community across platforms.” Unsurprisingly, then, Microsoft “continued to ship Minecraft on all those same platforms post-acquisition.” Microsoft’s actions vis-à-vis Minecraft, the court concluded, better “exemplifie[d] how a console seller (and Microsoft in particular)” could be expected to behave “when acquiring a hugely popular multiplayer cross- platform game.” Against this backdrop, the district court also noted that, despite exhaustive discovery involving “nearly 1 million documents and 30 depositions, the FTC ha[d] not identified a single document which contradicts Microsoft’s publicly- stated commitment to make Call of Duty available on PlayStation (and Nintendo Switch).” Reviewing the assembled record, the district court concluded that the

FTC V. MICROSOFT CORP. 27

evidence of Microsoft’s actions and internal discussions was all consistent with its stated intention to continue to make Call of Duty available on PlayStation. In particular, the district court noted that Microsoft’s internal model evaluating the value of the Activision Blizzard purchase affirmatively “relie[d] on PlayStation sales and other non- Microsoft platforms post-acquisition” and did “not rely on increased sales of Xbox consoles for any reason, let alone caused by foreclosing Call of Duty from PlayStation.” In response, the FTC points to one set of internal documents in which Microsoft modeled whether, post-merger, it could recoup lost revenue from Call of Duty sales on PlayStation. But this model was based not on a plan to remove Call of Duty from PlayStation but rather on a hypothetical where Sony demanded a higher platform fee (i.e., royalty split) from having Call of Duty on PlayStation. Because even this internal model affirmatively assumed that Call of Duty would remain on PlayStation, it does not support an inference that Microsoft intended to make Call of Duty exclusive to Xbox. While noting that Microsoft’s internal documents were consistent with its public statements that Microsoft did not plan to pull Call of Duty from PlayStation consoles, the district court also appropriately recognized that such internal deal valuation analyses are “not dispositive of the incentive question,” particularly given Microsoft’s statements and behavior before and after the ZeniMax acquisition. But we cannot say that the district court abused its discretion in concluding that, when considered together with the other record evidence, these internal documents and external statements provided further support to what the overwhelming weight of the evidence already showed— namely, that Microsoft lacked any incentive to remove Call of Duty from PlayStation. 3 For the foregoing reasons, we conclude that the district court did not abuse its discretion in holding that the FTC had not made the requisite showing of a likelihood of success on its claim that Microsoft might make Call of Duty exclusive to Xbox after the merger. We therefore do not rely on an additional point that was cited by the district court—namely, “Microsoft’s written offer to Sony to offer PlayStation Call of Duty on parity with Microsoft for 10 years, including on future PlayStation consoles.” The district court expressly stated that this additional point was “not necessary” to its ruling on the likelihood-of-success issue, and we likewise find it unnecessary to address that point in reviewing that ruling. We therefore have no occasion to consider whether the FTC is correct in contending that contemplated post- merger arrangements constitute “proposed remedies” that 3 The district court also exhaustively analyzed the evidence and testimony presented by the FTC’s expert, Dr. Robin Lee, who sought to establish Microsoft’s incentive to make Call of Duty exclusive by using a model that he claimed showed that Microsoft would more than make up for lost PlayStation Call of Duty revenue by substantially increasing its position in the console market. The district court noted that Dr. Lee’s model depended critically on the assumed “Xbox conversion rate,” i.e., the rate at which PlayStation users “would purchase an Xbox console to play Call of Duty 2025 if it was not available on PlayStation.” In particular, if the conversion rate was only slightly lower than Dr. Lee’s assumed 20% rate, Dr. Lee’s own model would show net losses from making Call of Duty exclusive. In addition, over several pages, the district court carefully explained why Dr. Lee’s assumed 20% conversion rate was unsupported and speculative. The FTC’s opening brief makes no effort to address this detailed analysis or to explain why it is wrong, and it instead presents such an analysis for the first time in its reply brief. We therefore deem any argument challenging the district court’s discounting of Dr. Lee’s report to be forfeited. See Warfield v. Alaniz, 569 F.3d 1015, 1028 n.9 (9th Cir. 2009).

FTC V. MICROSOFT CORP. 29

should not be considered when courts assess the FTC’s likelihood of success on an underlying Clayton Act § 7 claim for purposes of a preliminary injunction under FTC Act § 13(b). The district court also rejected the FTC’s alternative argument that it had adequately shown that Microsoft would have the incentive to engage in what the FTC characterized as “partial foreclosure” with respect to Call of Duty—that is, to disfavor PlayStation by, for example, releasing only an inferior version of the game on PlayStation or by releasing new versions of the game later on PlayStation than on Xbox. The FTC argues that the district court erred in concluding that, “[i]f the FTC has not shown a financial incentive to engage in full foreclosure, then it has not shown a financial incentive to engage in partial foreclosure.” We agree that the mere fact that a company does not have a financial incentive to engage in full foreclosure does not, without more, establish that it similarly lacks an incentive to engage in partial foreclosure. But the district court also separately held, in addition, that the FTC presented insufficient evidence to support its partial foreclosure theory, and we discern no abuse of discretion in that holding. In particular, the district court noted that there was record evidence that no game developer had ever “intentionally develop[ed] a ‘subpar game for one platform versus another,’” because it would lead to a significant loss of goodwill among gamers. The court also stated that “the record does not include any evidence Microsoft has engaged in such conduct in the past—even with Sony.” Indeed, the court observed that even Sony’s CEO had testified that “publishers have every incentive to provide an equal gaming experience or as good a gaming experience as possible on all platforms.” On appeal, the FTC points to testimony concerning Microsoft’s favoring of Xbox vis-à-vis Starfield and Redfall after the ZeniMax merger, which assertedly shows that Microsoft may well engage in partial foreclosure by delaying introduction of games on other platforms. The FTC also suggests that Sony may itself cause a form of partial foreclosure to occur by delaying sharing with Microsoft, post-merger, the competitively sensitive development kits necessary to introduce Activision Blizzard games on future versions of Sony’s consoles. But the district court permissibly concluded that, absent “expert testimony” addressing the competitive impact of such feared partial disclosure practices, the FTC simply failed to raise serious questions as to whether there was a reasonable possibility that Microsoft would actually have an incentive to engage in such conduct with respect to a well-established multiplayer, multi-platform game such as Call of Duty. To the extent the FTC argues that Microsoft would have an incentive, after the merger, to make “other Activision titles exclusive to Xbox”—i.e., titles other than Call of Duty—the district court did not abuse its discretion in concluding that the FTC had failed to show that such exclusivity might substantially lessen competition in the console market. The mere fact that, after a vertical merger, a company might make some of its newly acquired intellectual property exclusive to its platforms does not, without more, show a substantial lessening of competition. Cf. Fruehauf Corp. v. FTC, 603 F.2d 345, 352 n.9 (2d Cir. 1979) (rejecting assumption that “any vertical foreclosure lessens competition”). It is in the nature of intellectual property rights that the holder ultimately has exclusive control over them, see Image Tech. Servs., Inc. v. Eastman Kodak Co., 125 F.3d 1195, 1215 (9th Cir. 1997), and the question under § 7 is whether there is a reasonable

FTC V. MICROSOFT CORP. 31

probability that, if Microsoft acquires such exclusivity rights with respect to the relevant intellectual property, Microsoft will exercise such rights in a manner that substantially lessens competition in the pertinent market, i.e., the console market. In the context of a vertical merger, that requires something more than merely showing that some of the rights acquired will be made exclusive. Cf. United States v. AT&T, Inc., 916 F.3d 1029, 1032 (D.C. Cir. 2019) (holding that, to carry its burden under § 7 for a vertical merger, the Government “must make a ‘fact-specific’ showing that the proposed merger is ‘likely to be anticompetitive’” (citation omitted)). The FTC itself seemed to recognize as much, because it tried to affirmatively establish a substantial lessening of competition from an exercise of exclusivity with respect to Call of Duty, but the district court permissibly concluded that the FTC had failed in that endeavor. On this record, the district court did not abuse its discretion in further holding that the FTC had not made a sufficient affirmative showing of a substantial lessening of competition with respect to the exclusivity of other titles in the console market. Finally, the FTC contends that the district court failed to adequately consider whether the FTC had made a sufficient “alternative” showing of a substantial lessening of competition in the console market under the framework set forth in Brown Shoe, apart from the “ability and incentive to foreclose” analysis that the district court employed. According to the FTC, Brown Shoe sets forth a multi-factor analysis for assessing the competitive impact of a proposed vertical merger, and the district court did not give adequate consideration to all of the relevant factors. But as the FTC’s own opening brief makes clear, the competitive significance of the various factors invoked by the FTC—such as the extent of any foreclosure, the purpose and nature of the merger, the effect of the merger on barriers to entry, and the effect on industry concentration tendencies—ultimately turns, in the context of the record evidence in this case, on the FTC’s central premise that Microsoft will engage in foreclosure. Accordingly, even if Brown Shoe leaves open alternative ways to establish a lessening of competition that do not rely on foreclosure, the FTC did not meaningfully rely on evidentiary proof of any such “alternative” theory of a substantial lessening of competition in the proceedings below. The district court therefore properly held that the FTC’s supposedly alternative Brown Shoe theory did not in fact “make any new arguments not considered” by the court in its analysis of the likelihood and competitive impact of potential foreclosure on the console market. B The FTC also challenges the district court’s holding that the FTC had not made an adequate showing that the merger would substantially lessen competition in the library subscription services market. We discern no abuse of discretion. We first address the FTC’s foreclosure-based theory in this market. In contrast to its conclusions with respect to the console market, the district court accepted, “for preliminary injunction purposes,” that “it is likely Call of Duty will be offered exclusively on Game Pass, and not offered on rival subscription services” (emphasis added). The court thus started from the premise that Microsoft would have both the ability and the incentive to exercise exclusivity rights with respect to Call of Duty and other Activision Blizzard content in the subscription market. The district court nonetheless concluded that, because Activision Blizzard had long

FTC V. MICROSOFT CORP. 33

opposed putting its content on subscription services, the merger’s effect of making such content available for the first time in the subscription market, even if exclusive to Microsoft, would not substantially lessen competition. The FTC derides the district court’s reasoning as an improper “efficiencies defense” to an otherwise-established “prima facie case.” Cf. St. Alphonsus, 778 F.3d at 788–90 (expressing skepticism about the validity of such a defense). But this argument rests on the premise that, merely by showing that Activision Blizzard content would be exclusive to Microsoft’s subscription services after the merger, the FTC sufficiently established a prima facie case that competition would be substantially lessened in that market. We disagree. As we have explained, and as the district court noted, merely showing that some content will be exclusive after a vertical merger does not, without more, establish as a factual matter that competition will be substantially lessened. See supra at 30–31. The paradigmatic premise of harm to competition from a vertical merger is that it will lead to “foreclosure of a share of the market otherwise open to competitors.” Brown Shoe, 370 U.S. at 328 (emphasis added). In the unusual circumstances presented here, in which Activision Blizzard as an independent company had persistently resisted allowing its content to be included in subscription services, making Activision Blizzard content exclusive to Microsoft’s subscription services would not foreclose a share of the subscription market “otherwise open to competitors.” Because this vertical merger would not be expected to result in “foreclosure” in the traditional sense of that term, the district court properly required the FTC to provide more evidence that this vertical merger would harm competition. The FTC argues that it did sufficiently show that Activision Blizzard’s content would eventually be available to subscription services in the absence of the merger, but we conclude that, in holding otherwise, the district court did not abuse its discretion or rely on clearly erroneous findings. The FTC points to testimony from Activision Blizzard’s CEO acknowledging that no “formal decision” had been made “never” to offer the company’s games on a subscription service; that Activision had engaged in discussions with Microsoft about putting its games on Game Pass but was unable to come to satisfactory terms; and that it was “possible” that its concerns about such arrangements could be addressed. The FTC also points to evidence showing that some Activision Blizzard titles had been included in subscription services in the past. But the district court permissibly concluded that, when considered in light of the record as a whole, such evidence did not sufficiently show that, in the absence of the merger, Activision Blizzard’s content would be available to Microsoft’s competitors in the subscription market. Specifically, the record evidence strongly supports the district court’s finding that Activision Blizzard had persistently concluded that, so long as it was an independent company, its financial interests would not be served by allowing its content to be included in a multi-game subscription service. Activision Blizzard’s CEO testified that concerns about “cannibalization”—i.e., a net loss of revenue from replacing sales to individual gamers with subscription access to those games—played a role in this long-held view and that, based on his experience, he did not “think that there is a circumstance where a company could ever offer us a commercial arrangement [concerning subscription access] . . . that would make sense” for Activision as a stand-alone

FTC V. MICROSOFT CORP. 35

company. Indeed, Sony’s CEO testified that he did not even try to ask Activision Blizzard to put Call of Duty on Sony’s subscription service, because its CEO “had been very public and very vocal that he did not see that as a route he wanted to take Activision.” The mere fact that Activision Blizzard’s CEO could not say that a satisfactory arrangement would never occur did not require the district court to conclude that the FTC had sufficiently shown that Activision Blizzard’s content might actually be available, absent the merger, in the subscription market. Nor was a contrary conclusion required by the limited evidence showing that some Activision Blizzard games had been included in subscription services in the past. The company’s CEO had explained that such occasional arrangements had been done on an “experimental or promotional[] basis” or by using a “very old catalog title for a short period of time.” As with its earlier Brown Shoe argument in the console market, the FTC alternatively contends that, even if the merger would not result in foreclosure in the traditional sense of that term in the subscription market, the merger still “would allow Microsoft to seriously disadvantage its rivals” in that market, where it already is a market leader, thereby resulting in a substantial lessening of competition. In support, the FTC relies on Dr. Lee’s report, but the district court properly concluded that Dr. Lee had failed to substantiate his largely conclusory assertions on this score. As the district court explained, Dr. Lee “did not perform any quantitative analysis” to determine how Microsoft’s exclusive access to Activision Blizzard content in the subscription services market would “affect competition with Game Pass competitors such as Amazon, Electronic Arts, Ubisoft and Sony.” To be sure, academics have posited, for example, that vertical mergers, particularly between content platforms and content creators, may lead to scenarios in which costs to competing platforms rise, leading to higher prices on those platforms and hampering competition overall. See HERBERT HOVENKAMP, PRINCIPLES OF ANTITRUST § 9.4 at pp. 383–86 (2d ed. 2021). But in the context of a vertical merger, the FTC cannot rely on intuition, theory, or other “short cut[s]” to carry its ultimate burden under § 7; rather, it “must make a ‘fact-specific’ showing that the proposed merger is ‘likely to be anticompetitive.’” AT&T, 916 F.3d at 1032 (citation omitted). And even when that ultimate burden is viewed through the lens of § 13(b)’s more lenient standard for assessing likelihood of success on the merits, the FTC still must come forward with evidence to make a sufficient showing as to the anticipated effect of this particular merger and how it would substantially lessen competition. The district court permissibly concluded that, as to the subscription market, the FTC simply failed to do so. C For reasons similar to those just discussed with respect to the subscription market, we also conclude that the district court did not abuse its discretion in finding an insufficient likelihood of success as to the cloud-streaming market. 4 As with the subscription market, the district court held that the FTC had failed to make a sufficient showing that Activision Blizzard content would be available to the cloud- streaming market in the absence of the merger. Although 4 The parties vigorously dispute whether we may consider the conditions imposed on Microsoft with respect to the cloud streaming market by British authorities in their approval of the merger in October 2023—i.e., after the FTC had already filed this appeal. See supra at 15–16. We find it unnecessary to resolve this issue because, even without considering those conditions, we conclude that the district court did not err.

FTC V. MICROSOFT CORP. 37

Activision Blizzard had allowed some titles, including some versions of Call of Duty, to be available on Nvidia’s GeForce Now streaming platform during Nvidia’s “beta test,” Activision Blizzard “instructed Nvidia to remove Activision Blizzard games from GeForce Now” in February 2020 when Nvidia “transitioned from the beta stage to a commercial version of GeForce Now” (capitalization altered). Since then, as the district court noted, Activision Blizzard content “has not been on a cloud-streaming service.” Moreover, in the limited streaming that Activision Blizzard had allowed during Nvidia’s beta testing, gamers “had to own the game.” Because the gamers had to already own the Activision Blizzard game in order to stream it on this beta-testing system, that limited use of streaming did not present the sort of “cannibalization” concerns that stand-alone streaming access would. The FTC points to no other evidence that Activision Blizzard had ever allowed its games to be included in streaming services, and the company’s CEO testified that Activision Blizzard did not view streaming, economically, as a “big opportunity for the company.” Although the FTC again notes that Activision Blizzard had not concluded that, as an independent company, it would “never” allow its games onto a streaming service and that Activision Blizzard was in conversation with Nvidia on that subject at the time the merger was announced, we cannot say that the district court abused its discretion in concluding that the FTC’s evidentiary showing on this point was simply too weak. Because the FTC failed to make an adequate showing that, absent the merger, Activision Blizzard’s content would be “otherwise open to competitors” in the streaming market, Brown Shoe, 370 U.S. at 328, it failed to show a sufficient likelihood of success as to its foreclosure-based theory of a substantial lessening of competition. The FTC again argues that, even apart from its theory of alleged foreclosure of otherwise available content, the FTC has also shown that Microsoft’s potential exclusive access to Activision Blizzard’s content in the streaming market could be so advantageous that it would substantially harm competition and lead to “higher prices, lower quality, less product variety, and reduced innovation.” But on this point, the FTC once again relies on the same sort of conclusory assertions by Dr. Lee that were discussed earlier with respect to the subscription market, and the district court permissibly found these assertions to be inadequate to carry the FTC’s burden. For these reasons, we hold that the district court did not abuse its discretion in concluding that the FTC had not shown a sufficient likelihood of success on its § 7 claim with respect to the cloud-streaming market. 5 V Given the FTC’s failure to make an adequate showing as to its likelihood of success on the merits as to any of its theories, the district court properly denied the FTC’s motion for a preliminary injunction on that basis. We therefore do not address the district court’s alternative holding that, even

5 We therefore find it unnecessary to address whether the FTC’s foreclosure-based theory fails for the additional reason that Microsoft entered into post-merger contracts allowing certain Activision Blizzard content to be available on competing cloud-streaming services. The FTC again argues that such post-merger agreements are relevant only to “remedies” and cannot be considered at this stage, but, as before, see supra at 28–29, we need not decide this point.

FTC V. MICROSOFT CORP. 39

if the FTC had made a sufficient showing, the balance of equities did not favor a preliminary injunction.

AFFIRMED.