The memorandum and order in FTC v. Meta Platforms Inc. that U.S. District Court Judge James E. Boasberg filed Nov. 18, ruling in favor of Meta, has now been followed by a Dec. 2 revised order that contained fewer redactions.
The memorandum doesn’t exactly provide the law & economics analysis I would have produced, had anyone asked for it. But it is, nonetheless, a good decision by a thoughtful, generalist trial-court judge wrestling with both the evidence before him and relevant precedent. He got key things right, not least of which was the decision for the defendant.
A concise summary of the reasoning can be found on its second page:
In this fifth year of litigation, the Court held a lengthy bench trial, hearing from myriad witnesses throughout the industry, as well as from dueling sets of experts. As it has forecast in prior Opinions over the years, the FTC has an uphill battle to establish the contours of any separate PSN market and Defendant’s monopoly therein. The Court ultimately concludes that the agency has not carried its burden: Meta holds no monopoly in the relevant market. Judgment must therefore be entered in its favor.
Boasberg rejected the Federal Trade Commission’s (FTC) gerrymandered market definition, which was critical to the agency’s indirect evidence of monopoly power. He also rejected the FTC’s very limited–and frankly, not credible–proffer of direct evidence for monopoly power. Hence, he found that the FTC had failed to prove unlawful monopolization under Section 2 of the Sherman Act.
A Predictable Outcome (for Some of Us)
Those who read my April post on the matter—subtly titled “The FTC’s Zombie Antitrust Action Continues to Lurch Forward”—will be unsurprised to read that I think Boasberg was right. For some quick highlights of my pre-trial post, I touched on:
- The commission’s narrow and eccentric (one might say “gerrymandered”) product-market definition;
- The FTC’s reliance on the fuzzy Brown Shoe factors, and especially the subjective (and, in this case, byzantine) cherry picking of “peculiar characteristics and uses” of the alleged product;
- The commission’s indirect evidence for monopoly power, which turned on that gerrymandered market definition and, on top of that, an unconvincing tally of Meta’s (Facebook’s) market share;
- The weakness of the market-share argument, which amounted to a distortion riding the back of another distortion;
- The FTC’s strained attempt to provide direct evidence of monopoly power (and consumer harm) via an overly simple nonprice (product quality) argument, which is always tough sledding in antitrust cases (and for good reason, and conspicuously off in this case);
- Various market developments in the years since Facebook’s Instagram and WhatsApp acquisitions (many still ongoing), including considerable innovation in available features and, across social media, both the growing bundle of features or functions provided by platforms and a significant degree of convergence in the bundles of features offered.
None of these issues escaped Boasberg’s attention. Of special interest is his recognition of some basic attributes of competition in what should have been the relevant market:
- This was always about differentiated competition; and
- This is a dynamic space, with ever-developing and shifting bundles of functionality and other nonprice attributes, changing consumer preferences (including expectations) driving demand, and competition among firms for the capacity (or capacities) to compete for present and future consumers with present and future technology.
No psychic powers were involved in my prediction. On the one hand, there’s always more to an antitrust case than one can glean from the pleadings; and district court judges may render surprising judgments in any case. On the other, the case was widely considered a weak one. See, e.g., Brian Albrecht and yours truly in the Yale Journal of Regulation’s Notice and Comment blog back in April 2024 (discussing an informal poll of antitrust economists) and again in September 2024 (discussing an informal poll of both expert economists and legal scholars).
For other thoughtful commentary, see Brian Albrecht on network effects and the FTC’s case; the American Enterprise Institute’s Mark Jamison arguing that “The FTC’s Case Against Meta Looks Like Politics, Not Antitrust”; and Joe Coniglio at Information Technology and Innovation Foundation in June 2024, December 2024, and May 2025. For some additional background, see Dan Crane and Herb Hovenkamp’s critique of the “monopoly broth” theory applied to what was originally the Facebook case, albeit in an amicus brief filed in another matter.
And, indeed, Boasberg himself had suggested not a little skepticism about the FTC’s case in preliminary rulings, including his June 2021 decision dismissing the case as legally insufficient (questioning the FTC’s market definition and market-share allegations) and his April 2025 decision, which permitted the FTC’s second amended complaint to go forward, but just barely:
Under the forgiving summary-judgment standard, the FTC has put forward evidence for a reasonable factfinder to rule in its favor. Prevailing here, however, does not obscure the fact that the Commission faces hard questions about whether its claims can hold up in the crucible of trial. Indeed, its positions at time strain this country’s creaking antitrust precedents to their limits.
Bear in mind that that was the biggest judicial thumbs up the FTC received in the roughly five years between when it filed its initial (dismissed) complaint in the matter and the time it arrived at trial. The decision may not have been a foregone conclusion, but the alarm and dismay it has prompted from some observers has seemed a little much. Perhaps not quite Capitaine Renault being shocked, shocked, to find gambling at his favorite casino in Casablanca, but getting there.
The FTC had screened Facebook’s acquisitions of Instagram and WhatsApp—in 2012 and 2014, respectively—when they were first proposed (with attendant Hart-Scott-Rodino filings). Neither deal raised serious concerns at the time. The vote to issue closing letters to the Facebook/Instagram transaction was 5-0. The WhatsApp screening passed without comment on antitrust grounds, although then-Bureau of Consumer Protection Director Jessica Rich wrote to the merging parties under her own name and title stating that WhatsApp’s privacy assurances to its consumers would still be subject to the FTC’s Section 5 consumer-protection jurisdiction.
I didn’t work on either merger review, but I was around, and people talk. For what it’s worth, there were numerous alternative photo-sharing apps at the time of the Instagram acquisition (just as there are numerous such apps still) and numerous messaging apps at the time of the WhatsApp acquisition, with WhatsApp being barely in the U.S. market at all. There was no case to bring against either deal at the time.
The FTC was hardly alone in its view of the mergers. The UK Office of Fair Trading also reviewed the Instagram deal in 2012, concluding that it “does not believe that it is or may be the case that the merger may be expected to result in a substantial lessening of competition within a market or markets in the United Kingdom.” And the European Commission reviewed the WhatsApp deal in 2014, concluding that, based on “the considerable differences between the functionalities and focus of each of the Parties,” Facebook and WhatsApp “were not close competitors in that potential market” and the deal did not pose a threat to competition in “the market for the provision of online advertising services, including any potential sub-segments of it” either.
I wasn’t part of the team that brought the Meta case in 2020 either. But I’m revealing no secrets when I recall that quite a few staff expressed surprise that Joe Simons—a solid antitrust and agency veteran—pushed it, over the dissents of Commissioners Christine Wilson and Noah Phillips. It’s hard for me to imagine that the director of the Bureau of Economics wrote in support of the case. Read the initial complaint—the one Boasberg dismissed in 2021. It’s no great shakes.
Is the Court ‘Out of Touch with Economic Realities’?
Still, I gather that some in the Bureau of Competition supported the complaint in 2020, 2021, and again in 2025, at least with whatever encouragement they might have received from upstairs. And not everyone on the outside thought the case a loser. Here’s Tim Wu with a slightly different take than my own:
The Facebook decision shows a court way too swamped in technical detail losing touch with the anti-monopoly goals of the Sherman Act.
Wu would go on to opine in a New York Times guest essay that Boasberg’s decision ran “in the face of strong evidence to the contrary, not to mention common sense.” Not only that, but:
Congress has done little to restore economic balance and courts are showing themselves to be out of touch with economic realities that any American could describe. It is no wonder the population is growing increasingly cynical — and more and more ready to look to ideologies that promise to upend the entire system.
I don’t mean to gainsay concerns about public confidence in the courts or our market institutions–not generally, at any rate, not that I’ve assessed such things. Populists to the right of us, populists to the left of us, and some of their volleys and thunders seem surprisingly popular. Still, it all seems rather a lot to hang on antitrust, never mind a single decision by a U.S. district court about mergers that were screened by U.S. and other enforcers—and consummated without a fuss—more than a decade ago.
To be fair, there’s only so much analysis one can do in the space of an op-ed column. Even so, Wu’s column seemed to rest heavily on rhetoric and appeals to intuition, such as asking whether “anyone seriously doubt[s] that Meta is the kind of company that antitrust laws were designed to restrain?” (Well, yeah, I doubt it.) He does hint at other arguments, albeit weak ones. For example, he notes:
The government presented records of the company’s extraordinary and durable profits ($87.1 billion operating profit on $164.5 billion in revenue in 2024, for example), which is a textbook signal of monopoly power.
Perhaps, but profits and related measures like markups over cost can be poor proxies for competition, or a lack of competition. As Brian Albrecht has explained, rising profits can also be explained by increased efficiency. If a grocery store finds a way to save money and passes on even the lion’s share of that savings to its customers, its profits will still rise. That increase in profits doesn’t signal consumer harm (consumers are better off), much less monopoly power or anything that the antitrust laws mean (according to the courts) by “harm to competition.”
As it happens, this is not a point that escaped Boasberg. A relevant discussion can be found on page 25 of the revised memorandum opinion, which begins: “The record here . . .. reveals several other factors that could be driving Meta’s handsome profits, none of which the FTC has foreclosed.”
Brian had more to say in an op-ed in RealClearMarkets. I also recommend this measured review of the decision from former FTC Acting Chair Maureen Ohlhausen and some of her colleagues at Wilson Sonsini, and this barn burner of a podcast with Geoffrey Manne and Corbin Barthold.
More on Gerrymandered Market Definitions
The FTC sought to establish that the relevant market in the Meta case was (and is) a small subset of social-media apps that the commission called “personal-social-networking” (PSN) services. That is, not just any social-media apps used for personal and social (if that’s different) networking, but those that depend centrally on “a particular type of social graph built around friends and family connections.”
According to the FTC, that meant a market comprising four, and only four, competitors: Facebook, Instagram, Snapchat, and MeWe. As I noted in my April post:
According to the FTC, X.com/Twitter, TikTok, YouTube, and most of the myriad messaging apps that people may use to communicate with, well, their friends and family members—e.g., iMessage (Apple), Discord, Signal, Google Message, Line, Kik, Skype, and WeChat—are not in the market.
Heck, if we’re mentioning messaging apps, we might include WhatsApp, which is much more a part of U.S. social media now than it was when Facebook bought it in 2014.
The FTC’s subjective slicing and dicing (and weaving and warping) of qualitative aspects of various social-media platforms was not their only evidence for their alleged market. But it was front and center, and it was dubious from the start. This was always about differentiated and dynamic competition, and complaint counsel’s gymnastics simply didn’t stand up in the face of considerable evidence about how people actually use social-media apps, including both apps that the FTC tried to include in its PSN market and those it sought to exclude from it.
That should have been obvious to a casual observer by the time the FTC went to trial, and it was clear enough to the court at trial. Among the factors Boasberg considered in the revised opinion was diverse evidence of consumer substitution, along with evidence that consumers’ time sharing posts with friends and family via “a particular type of social graph built around friends and family connections” on Facebook was swamped by the time they spent viewing short videos or “reels” from strangers on Facebook, much as one might do on TikTok or YouTube, among others. Moreover, the time spent sharing such things with friends and family members ranges across a broad array of apps, and by not a little bit.
The government’s subjective juggling or cherry picking of “peculiar characteristics and uses” of the alleged product come from the hoary and discredited (although not overruled) Brown Shoe factors. As noted above, Herbert Hovenkamp—widely considered the leading scholar of U.S. antitrust law—has pointed out that such supposedly “practical indicia” are “so generic that they do not provide much guidance in an attempt to prop up that market definition”
Hovenkamp has further argued that “[t]he so-called Brown Shoe factors are wrong in most cases” and that, in any case, even as described by the Supreme Court in its poorly aging 1962 opinion, “there is no reason to think that it was doing anything more than summarizing fact findings for that particular case.”
A case for a market definition (and for monopoly power based on that definition) built on Brown Shoe factors is admissible, and may be a helpful part of a plaintiff’s (or defendant’s) telling of an overarching story in court. It is not, however, very reliable as a diagnostic or analytic tool. There really are better and established ways to do it, and everybody knows it.
A focus on subjective “peculiar characteristics and uses” can easily be misleading, even in describing a market that roughly fits widespread pre-empirical intuitions about the space of competition. A more rigorous case for a market, and for monopoly power (or even for a lesser degree of market power) is all the more crucial when the alleged market already appears from the start to be dated, counterintuitive, and, frankly, an artifact of a litigation strategy.
What about the FTC’s “direct evidence” for monopoly power? A monopolist can exploit the power to raise prices and suppress output. A reduction in product quality can be tantamount to an increase in price; we can think of it as an increase in quality-adjusted price. And so, the FTC and expert economic witness C. Scott Hemphill attempted to argue that we see just such an increase in quality-adjusted price in the increase in “ad load” exposure to advertising that consumers experience on Facebook’s main or default feed.
At first blush, there’s some intuitive appeal to that approach. All else equal, many consumers might prefer fewer ads to more of them. But it doesn’t take much of an inquiry to see that there’s more to it than that. All else is not equal. For one thing, as Meta argued and Boasberg noticed, both Facebook and its rivals do not provide (and have not provided) a fixed and stable bundle of features but, in fact, an increasing number of them. As Boasberg reminds us, “[t]he Facebook and Instagram that exist today bear little resemblance to the versions that readers might remember from the 2010s.” Reviewing a few of the changes, he found that:
Americans now spend only 17% of their time on Facebook viewing content from their friends. On Instagram, that number is 7%. What has replaced content from friends? For the most part, short videos posted by strangers and recommended by AI. A majority of Americans’ time on Facebook is now spent watching videos. Same for their time on Instagram. In particular, both apps have shifted to primarily showing Reels. . . . How people use Facebook and Instagram socially has changed, too. Users have become far less likely to post publicly and instead primarily share content using private messages, either in the app or over text.
The court reviewed a number of technical, market, and social changes pertinent to the shift. Not least among them was the entry of TikTok, which launched in the United States in 2018. “It soon put enormous competitive pressure on Meta.” Meta was not a monopolist price setter or, for that matter, a monopolist feature setter or quality setter either.
The long and the short of it was that Meta and the other platforms have competed in ongoing refinement of their feature bundles and means of delivery. Facebook is not just an app providing set functions a, b, and c, and that also has seen a steep increase in ad load (however measured). Indeed, even the ads themselves have changed in quality. The FTC’s focus on a single dimension of product quality flew in the face of many qualitative changes that have been driven by competitive forces, including numerous conspicuous product improvements.
Those changes undercut the FTC’s alleged market definition. They also undercut its indirect evidence of monopoly power, argued via misleading market-share calculations. (Indeed, Boasberg noted that “Meta’s true share of [the] market is almost certainly below 54%.”) And they undercut the commission’s direct evidence of monopoly power, argued via an artificially narrow focus on putative ad load.
The alleged increase in ad load appears to be off, even on its own terms, and not just because a user can scroll right through the ads. Remember the special features that supposedly distinguish Facebook and just three other platforms as constituting the FTC’s alleged PSN market? By and large, they are still there. They are less popular—getting less viewing time and generating less content to be viewed—but they are still there.
Facebook has not eliminated the “Friends” tab or the option of a friends’ feed. There’s still a “Groups” tab, too, and a corresponding setting for a groups’ feed. Anyone can still construct (and moderate) a private (or public) Facebook group on the fly. A private group can be an interest group, such as, e.g., birding, photography, or bird photography (as might appeal to a person of a certain age approximating my own). But it can also be a set comprising one’s actual friends and family members—or some subset thereof that’s perhaps considerably smaller than one’s set of “Facebook friends.”
If I log in to Facebook and go to such a group, I can scroll through a feed of posts by group members. Or I can, in the alternative, look for posts by certain friends, family, or others of special interest. In either case, I don’t see the same stream of Reels from strangers—or the same ads or the same number of ads—that I’d otherwise see. That is, my ad load when I choose to focus on friends and family (or birders, or members of the Dartmouth class of 1982) is not at all the one that was alleged to demonstrate monopoly power. The ad load is not zero; Facebook is ad-supported, as are other social-media platforms like TikTok, YouTube, and X.com. But it’s limited.
I suppose another possible “technical” issue might be Boasberg’s determination that the FTC was bound to show that the Instagram and WhatsApp acquisitions (and whatever else went into the “monopoly broth” of the alleged pattern and practice) required a showing that the commission’s authority to seek an injunction in the case “serves a . . . forward-facing role: enjoining ongoing and imminent future violations,” and hence required showing that the old acquisitions are, or are about to, violate Section 2 now. Again, there’s that pesky reference to statutory law and precedent.
And there are practical questions about the point of it all. There’s a cost to unwinding a merger many years after it is consummated. As Randy Picker has pointed out, there’s both an ex-ante cost (projecting uncertainty on other mergers, even after they’ve been screened by enforcers) and an ex-post cost (imposed on unwinding operations, “unscrambling the eggs,” etc.) Ex-post merger review may be important to competition and consumers when harm is demonstrable and likely to persist at the time a case is brought (as in the Evanston Northwestern Healthcare case, brought under Section 7 of the Clayton Act).
Merger review, at the HSR stage and thereafter, is always about what’s likely to come. What, after all, is the point to investing millions of dollars in limited agency resources, at risk, if one cannot show that harm, or the risk of harm, is ongoing?
So What Else?
Other post-shutdown agency news has included entries in the wins, losses, and settlements columns. I’ll recommend Bilal Sayyed’s short piece on the Gateway Services settlement, for one. Another remedies memorandum in the Google Search case dropped just this past Friday (I’ll get to it later this week), and discussion of the Netflix/Warner Bros. announcement is heating up (on that, here’s John Yun and here’s Eric Fruits). Stay tuned.
