This Just In
Judge Amit P. Mehta’s memorandum opinion in the Google Search case has dropped. It’s 230 pages, and I’ve merely skimmed it. A careful discussion–from me or anyone else–will wait a bit. For now, the remedies are quite a bit more than Google had proposed, but at the same time, a good deal less than the U.S. Justice Department (DOJ) wanted.
Among other things, Judge Mehta rejected the DOJ’s proposed structural relief, and for good reasons. He does not order the divesture of the Chrome browser, in part because: “Plaintiffs do not satisfy the Circuit’s ‘clearer indication of a significant causal connection’ test for structural remedies.” He also doesn’t order the divestiture of Android (one of the DOJ’s contingent proposals).
More, much more, to follow. And commentary aside, I don’t know that anyone expects the case to be over and done with.
Also This, Too
A divided panel of the U.S. Court of Appeals for the D.C. Circuit has sided with Rebecca Slaughter in Slaughter v. Trump, dissolving the administrative stay ordered by the same circuit court on July 21, and denying the government’s motion for a stay pending appeal. In effect, we’re back to Judge Loren L. AliKhan’s district court decision and contemporaneous order, which enjoined the FTC:
from removing Ms. Slaughter from her lawful position as an FTC Commissioner or otherwise interfering with Ms. Slaughter’s right to perform her lawful duties as an FTC Commissioner until the expiration of her term or unless she is lawfully removed by the President for ‘inefficiency, neglect of duty, or malfeasance in office.
The panel decision was hardly a foregone conclusion, but neither does it come as any great surprise. I’ve written about this and related issues a bunch, most recently here at Truth on the Market in July. I won’t recapitulate that or my other writings on the matter.
For now, we’re back to a bipartisan commission, which has some advantages, not that those settle the constitutional questions that have been at-issue. I expect that the government will appeal; I expect that the Supreme Court will grant certiorari; and I expect that the Court will issue a decision that pays less deference to Humphrey’s Executor and, hence, the language of the for-cause removal provision in the FTC Act. But those are ex-ante expectations, not established facts. We shall see.
None of this is about me but . . . gee, you draft a lengthy post with all sorts of news pending, and then big news drops (and keeps dropping). Here’s what I was going to say before such late-breaking developments.
Who Moderates the Moderators?
My International Center for Law & Economics (ICLE) colleague Ben Sperry and I have been writing about the federal antitrust agencies’ various forays into content moderation. Both the Federal Trade Commission (FTC) and the DOJ Antitrust Division have been making noise about, as Ben and I said, “issues at the intersection of antitrust and content moderation (or, as some FTC statements would have it, private ‘censorship.’)” For more of that, you’ll find Ben here, here, and here; me here; and ICLE comments in response to an FTC “Request for Public Comments Regarding Platform Technology Censorship” (to which Ben, Geoff Manne, and I contributed on behalf of ICLE) here.
As Ben and I and not a few others have noted (here’s a comment by Tech Freedom; and here’s Corbin Barthold on a related GOP suit against Google) these forays have been at times dramatic—at times, even ominous—but they’ve produced more heat than light. That is, they haven’t been very good.
You have the links, so there’s no need to recapitulate our prior discussions. In brief, as far as antitrust law and the FTC Act were concerned, yes, in abstract, nonprice dimensions of competition (including demonstrated or likely effects on product quality) could figure in antitrust analysis (not that anyone had said otherwise), but quality-only cases are few and far between, for good reasons, and much more plausible in entirely different contexts.
Yes, material, but false or misleading, marketing statements could violate the consumer-protection prong of Section 5 of the FTC Act (not that anyone had said otherwise), but such violations seemed unlikely in the context in question.
And yes, the firms operating tech platforms could be susceptible to scrutiny under the FTC Act or the antitrust laws (encore), but content moderation decisions by tech platforms were protected against regulation by state or federal governments by the First Amendment to the U.S. Constitution.
There could be a way through that thicket, but neither the FTC nor the DOJ had sketched a convincing one with any specificity. Actual content-moderation policies vary, and that variation seems critical to the differentiated competition in which social-media platforms engage. While I might like or loathe some moderation policy or other, this seems very much a matter of one consumer’s meat being another’s poison. And vice versa. In this context, content-based discrimination by a private party would likely be constitutive of a product, rather than a signal of qualitative decline, much less unlawful exploitation of market power.
And then, of course—and no small matter—is the First Amendment.
Now comes a press release from the FTC announcing an Aug. 28 letter signed by Chairman Andrew Ferguson in his official capacity, but not obviously writing for the commission or any other members thereof, and not referring to any formal law-enforcement investigation or FTC vote. Ferguson writes to Google and Alphabet CEO Sundar Pichai “due to recent reporting that suggests Alphabet’s administration of Gmail is designed to have partisan effects, and accordingly to notify you that Alphabet may be engaging in unfair or deceptive acts or practices.”
The letter’s threatening tone is moderated only by its frequent use of “may,” in the sense of “could possibly be,” even if there’s no corresponding use or suggestion of “or may not.”
In an earlier piece about content moderation, I acknowledged that certain statements by Ferguson and Commissioner Melissa Holyoak could be “more a matter of rhetorical vigor than a policy signal.” But this sort of public letter from the commission chair to the CEO of a particular firm seems something more.
The gravamen of the complaint (in the colloquial sense) is that “according to recent reporting”—if not a finding by the FTC or any court, or any published, rigorous research—“Alphabet ‘has been caught this summer flagging Republican fundraising emails as ‘dangerous’ spam—keeping them from hitting Gmail users’ inboxes—while leaving similar solicitations from Democrats untouched….’ Likewise, commenters on the FTC’s request for information regarding Technology Platform Censorship have complained that Google is using a partisan approach in administering its spam filters. And finally, as you know, similar concerns have resulted in ongoing litigation against Google in other settings.” (internal citations omitted).
Here, “internal citations omitted” means that I’ve left out three footnotes. The first cites to a New York Post article with a somewhat misleading title; the second cites to a letter signed jointly by Sen. Tim Scott (R-Fla.) and Rep. Richard Hudson (R-N.C.), the chairs of the National Republican Senatorial Committee and the National Republican Congressional Committee, respectively; and the third cites a case before the 9th U.S. Circuit Court of Appeals.
The letter by Sen. Scott and Rep. Hudson was submitted in response to the FTC’s request for public comments I mentioned above. And hoping you’ll forgive a longish quotation of my own observation back in February:
Keeping up with market developments by various means—from rigorous empirical studies to issue-spotting workshops and RFIs—are basic parts of the FTC’s statutory mission. The FTC’s press release does not mention a commission vote authorizing the RFI, but perhaps none was needed.
At the same time, we might wonder whether an inquiry into “censorship” and how “technology platform” conduct “may have violated the law” appears to tilt the inquiry (and perhaps the legal playing field), even if it doesn’t say which provision of the antitrust laws or the FTC Act “may” be violated by, e.g., a platform’s denial or degradation of “users’ access to services based on the content of their speech or affiliations.” Encouraging input from “[t]ech platform users who have been banned, shadow banned, demonetized, or otherwise censored” (but no others) hardly seems a neutral solicitation of public comment on the potential costs and benefits of platform conduct.
Indeed, some of the commission’s commentary seems downright ominous. The FTC’s press release informs us that “Censorship by technology platforms is not just un-American, it is potentially illegal.” And similarly (very similarly), in a post on the technology platform formerly known as Twitter, Ferguson says that: “Big Tech censorship is not just un-American, it is potentially illegal.”
Once again, the rhetoric seems ominous. No, it does not announce a law-enforcement action, and it stops short (if just short) of alleging FTC Act violations. But it’s an odd letter, at least, and perhaps at best. As FTC chair, it seemed to many that Lina Khan often flirted with pre-judgement. Which is to be avoided.
Not incidentally, the third of the footnotes cites to a “notice of oral argument in the case,” in support of the claim that “as you know, similar concerns have resulted in ongoing litigation against Google in other settings.” But a notice of oral argument is not a decision on the merits of a case, or findings of fact in support of a decision on the merits. For what it’s worth, it’s a case in which the plaintiff appellants, the Republican National Committee (RNC), alleged that Google had directed fundraising emails by the RNC to users’ spam folders “without warrant,” and in a manner (or to a degree) they did not for fundraising emails by the Democratic National Committee (DNC).
With just a bit more specificity: the RNC filed suit in 2022 in the U.S. District Court for the Eastern District of California alleging “violations of California’s common carrier law, the Unruh Civil Rights Act, the California Unfair Competition Law, and the Federal Telecommunications Act, as well as claims alleging intentional and negligent interference with prospective economic relations, and negligence under California Civil Code § 2162. (See Id. at 17-25.)”
Google filed a motion to dismiss, and in August 2023, Judge Daniel J. Calabretta granted that motion: case dismissed.
The RNC then filed an amended complaint, with additional allegations that Google’s conduct violated California’s Unfair Competition Law (UCL). In July 2024, Judge Calabretta once again dismissed the RNC complaint—this time, with prejudice.
The FTC Act was not at-issue in either decision. The dismissals recognize UCL concerns with consumer harms and competitive effects, among other things, but that is not to say that the UCL or its “unfairness prong” is simply a state-law version of Section 5, whether that’s taken to mean the unfair methods of competition (UMC) prong, the unfair or deceptive acts or practices (UDAP) prong, or both.
But the court’s recognition of First Amendment concerns, and its application of the Supreme Court’s decision in Moody v. NetChoice to the alleged bias in spam filtering, is both salient and significant. It is, as the court recognizes, a constraint on both federal and state law intervention.
In brief, the citation to the notice of oral argument is not much of a citation. The case is, I gather, “ongoing,” so there’s that. The RNC’s opening brief appealing to the 9th Circuit is here. But it’s a case in which the plaintiff’s allegations were dismissed twice—the latter time with prejudice. The fact that the RNC has appealed to the circuit court substantiates only the fact of the appeal, not the facts alleged, much less a finding of illegality under any federal or state law.
The footnotes seem the sort of citations that second-year law students are taught to flag when serving as law-review editors. It happens (see, e.g., Gus Hurwitz and me discussing some of the commission’s citations in “The FTC’s UMC Policy Statement: Untethered from Consumer Welfare and the Rule of Reason”). But it, too, is undesirable.
I’m not suggesting that the claims were frivolous. I’ve not reviewed the record that was before the court, and I’m not a member of the California Bar or an expert in California commercial law. But as Ben and I said before:
The Supreme Court has repeatedly made clear that the First Amendment only applies to government actors and gives wide latitude to private actors to police speech on their property. Antitrust law, after all, is government action, itself subject to the First Amendment.
And so is the FTC Act. California state law too, I reckon.
Apart from my concerns about the rule of law, and the proper application of the antitrust laws (and the FTC Act), my only dog in the race is this: I have seen quite a bit of spam from the RNC and the DNC, and from individual candidates, my zero donations to either party (or their candidates, or associated PACs) in 2024, 2020, 2016 (and in the years in between) notwithstanding. Most of what I’ve seen getting through really is spam—and I don’t eat spam.
As far as I can tell, solicitations at-issue are worth less than nothing as policy discussion, even as there have been very real and important policy differences at play. I wish that Google’s spam filter–and those of all my other product and service providers—caught more of it, not less. Disparate impact? Intent? With regard to one variety of phishing rubbish rather than another? Tell me less.
Ferguson also expresses concern that:
actions by foreign powers to impose censorship and weaken end-to-end encryption will erode Americans’ freedoms and subject them to myriad harms, such as surveillance by foreign governments and an increased risk of identity theft and fraud.
These may indeed be legitimate—indeed important—areas of concern, among other legitimate areas of concern about foreign regulatory, legal, and extra-legal constraints on, and actions against, U.S. tech firms. Ferguson has good reason to characterize a number of these as “misguided international regulatory requirements.” But that doesn’t lend much clarity or specificity to his suggestion that Google “may” be in violation of the FTC Act.
Again, there could be a way through the thicket, but I still don’t see one.
That Was the August of Our Discontent
On July 30 (proto-August), it was reported that Roger Alford and Bill Rinner—two senior officials in the DOJ Antitrust Division—had “been fired for insubordination.” To a first approximation, nobody had seen that coming (if, surely, some small number of people inside the administration saw it coming in advance of the news).
I know Alford and Rinner only by reputation, and some of their work, but their reputations had been excellent and there had been no obvious tension between them and the head of the division, Assistant U.S. Attorney General Gail Slater. I do know (or knew) Slater a little bit from our overlap at the FTC, where I found her to be a smart, thoughtful, and congenial colleague. It was rumored that she, too, did not see the firings coming—a rumor that seems plausible, although I really have no evidence one way or another, besides the ever-uncertain grapevine.
One wondered, and insubordination was not the first conjecture that came to mind. At first, Alford and Rinner themselves seemed tight-lipped about it all. And then there was this: Roger Alford’s barnburner of a reflection on “The Rule of Law vs. the Rule of Lobbyists.” In it, in the voice of Sir Thomas More in “A Man for All Seasons,” he excoriated “people inside and outside government who offer and accept the silver cup and who care little for the antitrust laws. They consider law enforcement not as binding rules but an opportunity to leverage power and extract concessions. They have, shall we say, a loose relationship with the law”—lobbyists, one supposes, and, expressly, Chad Mizelle and Stanley Woodward at DOJ.
The Hill, quoting Alford as mincing zero words:
Although I am limited in what I can say, it is my opinion that in the HPE-Juniper merger scandal, Chad Mizelle and Stanley Woodward perverted justice and acted inconsistent with the rule of law ….
A DOJ spokesman called Alford’s remarks sour grapes—or worse, really:
He was fired from the department, and all should treat his comments for what they are – the delusional musings of a disgruntled ex.
That seems to me frank and, frankly, excessive, at least. Alford’s extended analogy to Sir Thomas More might be a tad purple—dismissal precipitating a return to a tenured professorship at Notre Dame Law seem to me decidedly to be preferred over More’s execution for treason. Still, I don’t personally put much stock in the line about “delusional musings of a disgruntled ex.” And—concerns about ad hoc convenience samples aside—I don’t get the impression that many antitrust lawyers and economists do either.
The high drama shouldn’t distract from the interesting and potentially useful window into friction between the division and DOJ leadership. As a formal matter, the U.S. attorney general is head of the department and does have administrative authority over the division. It’s not obvious that this authority should never be used.
At the same time, there are at least strong pragmatic reasons to think that such authority should be used sparingly—indeed, rarely, when it comes to decisions about specific investigations, cases, and settlements. High-level policy may be the realm of the AG and, ultimately, the president, but high-level policy can only say just so much about the proper outcome in a matter subject to the rule of reason (something agency leadership under the Biden administration sometimes forgot).
Technical expertise and familiarity with the details of investigations and cases are concentrated in the division, and good government should avoid not just politically motivated enforcement (and, worse, corruption) but the appearance of such to reasonable outside observers.
But there’s another puzzle here: why the drama either way about this matter? In broad strokes, of course I favor the rule of law, which seems to me critical to the functioning of our society. And I am not now, nor have I ever been, a lobbyist. Bemoaning lobbyists seems just fine to me. But bemoaning is not banning or, for example, a suggestion that lobbyists are tolerable only if ineffective (or that any actual effectiveness is necessarily a sign of corruption).
If Alford is right that he (and Rinner) were fired because DOJ leadership caved to undue outside influence—if this was really, as Alford said, illustrative of a “new pay-to-play approach to antitrust enforcement”—driving a poor settlement in a strong antitrust case, to the likely detriment of competition and consumers, then talk of corruption versus the rule of law becomes appropriate.
I don’t know with any certainty that he’s wrong. On the other hand, as I wrote at the end of July, the division’s complaint in the matter seemed lacking. Once more, and with feeling:
I’ll recommend two pieces written by another former FTC colleague (and ICLE academic affiliate) John Yun. Yun discusses the Antitrust Division’s now-settled case against Hewlett Packard Enterprise’s acquisition of Juniper Networks (here and here). John wrote—convincingly, in my view—that the case against the merger was deficient, partly because it seemed a weak structural case, and substantially because it failed to consider potential merger efficiencies and, specifically, the ability and incentive of a merged entity to more effectively challenge an industry-leading incumbent in Cisco.
And because even the structural case was debatable as such. As former FTC General Counsel Alden Abbott also wrote, the complaint seemed to rely excessively on structural features, which “modern antitrust views as only the starting point in determining whether a merger might reduce competition.” And for that matter:
HPE and Juniper point out that DOJ’s market definition ignores “a broad set of players” that are competing for WLAN business, as customers shift to AI and cloud-driven strategies. Indeed, the European Commission found that HPE/Juniper would “continue to face competition from a wide range of competitors, including strong and established players on each of the markets.”
I was never in the weeds on this one, and I cannot say that the merger could not have raised competition concerns. But reading the complaint and other commentary give ample credence to the concerns Yun and Abbott, among others, raised about the DOJ’s case. A complaint is not the whole of a case, but the complaint was lacking, the case appeared infirm, and the division might have done more harm to competition than good had it succeeded in blocking the merger.
And if that’s right—and without questioning Alford’s integrity or intent—I wonder whether this really illustrates a fight for the rule of law. And whether, in that war, this was really the hill to die on.
While we’re on the topic (or one of them), I have to wonder about Alford’s view that this was “about the battle between genuine MAGA reformers and MAGA-In-Name-Only lobbyists.” I do not doubt his commitment to the rule of law as a general matter. And I don’t doubt that it matters to AAG Slater or FTC leadership. Some real improvements seem in evidence in case- and fact-specific law enforcement in both agencies.
There have been improvements, too, in attention to process–which matters–and, for example, in an increase in early terminations (at least relative to the Biden era). Beyond antitrust, we’ve seen some grounded consumer protection enforcement at the FTC, as I’ve noted before. But I honestly don’t know what battle Alford has in mind.
I have read Slater’s description of “The Conservative Roots of America First Antitrust Enforcement,” doing my darndest to suppress the obvious—if surely unintended—association with the 20th Century’s America First Committee. And I’ve read FTC Commissioner Mark Meador’s “Antitrust Policy for the Conservative,” as well as various remarks by Chairman Ferguson, all attempting to delineate what I suppose might be had in mind by “Trump Antitrust,” “Trump/Vance Antitrust,” or, perhaps, “genuine MAGA reformers.”
If bits and pieces seem sensible enough, other bits and pieces do not. Among other things are the nearly global indictments of the tech sector (certainly in tension with simultaneous advocacy for a bottom-up, case-specific approach to enforcement); a strained and certainly overbroad attempt to revive a “big is bad” approach (and corresponding presumptions) or to continue them from the era of Lina Khan and Jonathan Kanter; a cavalier push to diminish (if not wholly repudiate) the actual contributions and proper application of economics; and a drive to further minimize consideration of merger efficiencies. (Further? Compared to what? Agency practice? Recent judicial decisions?).
The whole thing seems a mixed bag at best, and more likely a muddle. A piece-by-piece analysis or critique will have to wait for another day. But is there really a there, there? With regard to “genuine MAGA reformers” versus “MAGA-In-Name-Only” antitrust, I wonder whether Alford may be stuck on a distinction without a difference. The administration’s top-down commitment to the rule of law? A reference without a referent?
Be such things as they may, this was, evidently, a breaking point. Why the fault line ran through this case remains something of a puzzle.