What should we do with a magic wand? The revision of the merger guidelines offers an opportunity to update and improve the foundational texts of U.S. merger control, as well dangers. The guidelines’ status as a critical enforcement asset is a function of judicial confidence that the guidelines reflect the teachings of experience and a consensus that will endure across Administrations. Moreover, it is not at all clear that the text of the 2010 guidelines is really holding back federal merger enforcement: other frontiers — including agency resources, judicial skepticism, and the HSR timeline — all seem to be much more significant constraints on the agencies. Accordingly, too free a hand in revising the guidelines risks undermining a key asset of the merger-control system, without offering much prospect of an offsetting benefit. The deep puzzles and gaps in merger law, including causation and the treatment of efficiencies and upfront remedies, likely cannot be solved by guidelines. But there are a number of ways in which the guidelines could usefully be filled out or adjusted. Among other things, brief discussions of future (including potential and nascent) competition, platform markets, data, and the relationship between merger review and conduct would help the public, courts, policy-makers, and merging parties understand how the agency thinks about some high-profile puzzles that have been the subject of extensive enforcement experience. By adding some explanation of these topics, while relying on litigation and legislation to develop the substance of merger law, the agencies can ensure the guidelines remain a reliable, trusted, and up-to-date guide to the execution of one of their most important functions.

Daniel Francis[1]


For a long time, I’ve had a favorite question for interviews and fireside-chat-type settings in which I get to ask the questions rather than answer them. It goes: “If you had a magic wand and could change just one thing about the antitrust system — its laws, precedents, enforcement practices, institutional arrangements, anything at all — what would it be and why?” I’ve found that, every time I hear someone answer it, I always learn something interesting, and my sense of antitrust possibility is broadened. So, when I was generously invited to contribute a short essay to this collection on the merger guidelines, I thought I might take a shot at answering my own question. What would I do to the merger guidelines with a magic wand?

The answer, it turns out, is underwhelming. I would make a handful of modest tweaks: mainly to say something brief regarding a handful of topics on which the guidelines are currently silent, or a bit thin. The modesty of my proposals is a function of three fundamental beliefs about the merger guidelines and the state of merger enforcement today. First: I believe that one of the greatest assets of the federal merger enforcement program is the fact that the merger guidelines — including but not limited to the quantification of the structural presumption — are accepted as strongly persuasive by courts, even in cases where courts disagree with an agency about the merits of an individual merger challenge. And I believe that that judicial confidence in the guidelines, in turn, is a function of the widespread belief among judges that the guidelines reflect existing law, economic consensus, and a continuity of experience across Administrations of both parties, not a partisan or transitory project, or an effort to change established principles of merger law. This judicial faith is an immensely valuable, but fragile, component of our enforcement system. As such, I would be reluctant to make changes — even substantive improvements — to the extent that they risked undermining the status of the guidelines as a trusted, bipartisan centerpiece of the merger enforcement project. That status is too valuable to the antitrust enforcement system to be lightly imperiled.

Second: I don’t think the text of the merger guidelines is a particularly important constraint on merger enforcement in the United States today. Certainly, there are some things that I would tweak, clarify, and improve. Like any important document, the guidelines can benefit from some re-evaluation after more than a decade of experience. But I really don’t think that what the merger guidelines (horizontal or vertical) do or don’t say is on the list of, say, the five or ten most significant limitations on merger enforcement today.[2] At the top of that list is resources: the agencies need more funding and more dollars, and without both of those it’s hard to see how the agencies can adequately cover their beat. When you have your hands full covering the more striking two-to-one and three-to-two cases, it’s hard to justify letting some of those deals go unchallenged in order to burn staff time and dollars on more experimental cases where the prospect of harm is harder to see, and where the chances of success are much lower. (My point is not that agencies shouldn’t bring experimental cases to develop the law, but rather that the costs of doing so are very high when you’re struggling to cover even the most troubling deals.) In second place I’d put the views of the federal judiciary regarding the meaning and application of Section 7. In third place, by a narrow margin, I’d put the tight time constraints imposed by the HSR Act (especially following substantial compliance with a second request) which force agencies to make rapid up-or-down decisions with incomplete information, risking kneejerk decisions to file, or forgo, suit. There is plenty of room for reasonable minds to differ about whether and to what extent it’s a good or a bad thing that the agencies face resource constraints, skeptical courts, and tight deadlines, but I don’t think it’s plausible that the text of the merger guidelines exert a limiting force of anything like the same magnitude.

Third: I think the most pressing difficulties and ambiguities in substantive merger law — including confusion around the causal standard (“may be”), the treatment of uncertainty, and the analytical role of efficiencies and upfront remedies — can only be fixed through some combination of litigation and legislation. I don’t think guidelines are a plausible solution to those problems. The Supreme Court hasn’t taken a substantive merger case since 1975 and Congress hasn’t touched Section 7 since 1950: and, with the best will in the world, I don’t think agency guidelines can fill the costly voids that have resulted. The Court or Congress, or maybe both, must help us out.

The rest of the essay is in two parts. Part I offers some general thoughts about the function and limitations of the guidelines; Part II sets out and explains some specific choices that I would make when waving a magic wand at the guidelines.



There’s room for reasonable disagreement about what exactly merger guidelines should do: that is, what kind of function they are supposed to perform. Are they supposed to interpret existing law, maybe including specific cases? Describe agency practice? Explain the economic analysis of likely effects from mergers? Set out a vision of what the law should be in an effort to move the judiciary? Some combination of these things? Something else?

My own view is that guidelines should primarily aim to describe agency practice to the public, to courts, and to businesses. They are an explanatory document to tell the world what the FTC and DOJ does when undertaking merger review. This involves: identifying the central questions that agency staff will ask during a merger review; helping readers to understand the kinds of evidence that will be the focus of agency attention; and explaining how that evidence will be analyzed.

This necessarily involves or reflects some interpretation of law — that is, a view about what merger law requires — because agency practice is and should be a function of what the law is. Other antitrust guidelines cite cases sparingly, and there is probably room for some of this in merger guidelines too.[3] And of course any agency view about what the law is will reflect some quantum of the agency’s perspective about what the law should be: its sense of the “best reading” of existing law. It could not be otherwise.

But each step that the guidelines take into “interpretation of law” — and particularly into a self-aware effort to change or extend existing law — is fraught with danger. Because the antitrust agencies don’t just issue the merger guidelines as an informational service to the world in a here’s-what-we’re-doing-this-week kind of way. They issue the merger guidelines partly to help win cases. In particular, it is very helpful for a litigating agency to be able to reassure courts during merger litigation that a challenged transaction stands condemned under an established analytical framework that has not been cooked up for this litigation in particular; is recognized as a sensible, balanced, and economically sound product of evidence and experience; and does not represent a partisan or transitory project that will be turned on its head every four years. In other words, the persuasive force of guidelines in court is a function of the extent to which courts really believe that the guidelines represent a genuine pre-commitment by the agencies to a consensus-based analytical framework.

The existing merger guidelines have been pretty successful in this regard.[4] When courts mention the guidelines, the references are overwhelmingly supportive: they are treated as highly persuasive at least.[5] They are even cited in conduct cases,[6] and in private merger challenges.[7] And even when courts reject agency claims that an individual transaction is anticompetitive, they virtually always accept the persuasive force of the guidelines.[8]

This judicial faith is a huge structural asset for the agencies and for the antitrust enforcement project — particularly given the importance and difficulty of providing trusted guidance to judges who may be fairly or completely new to antitrust or merger cases. Among other things, the quantification of the structural presumption itself is a critical resource for helping to explain to judges why a particular transaction is beyond the pale. But, while this judicial faith is an asset, it is a fragile one. And it may be increasingly vulnerable, as the federal judiciary appears to be taking a closer and more skeptical look at the work of the administrative state.[9] The FTC in particular has had some strikingly unpleasant experiences in recent years.[10]

In light of all this, I think it is fairly clear that persuasive force of the guidelines does not reflect a judicial view that, as a matter of law, the agencies know best, or deserve substantive deference, when it comes to interpreting Section 7. (Or at least, it’s hard to think that view would survive being teed up and litigated in that form.) Rather, it is a function of the status of the guidelines as a reflection of broadly settled principles, themselves accepted as authoritative by the judiciary, refracted through years of experience across Administrations of both parties. In fact, the D.C. Circuit said almost exactly this in Anthem: “Although . . . the court is not bound by, and owes no particular deference to, the Guidelines, this court considers them a helpful tool, in view of the many years of thoughtful analysis they represent, for analyzing proposed mergers.”[11] The changes to the 2010 guidelines, for example, reflected evolutionary changes in agency practice and developments in methods of economic analysis since the previous iteration.[12]

What happens if that confidence is shaken? What happens if courts come to believe that the guidelines represent an effort to reinterpret or reengineer principles of antitrust law, and to introduce a step-change break with previous experience rather than a reflection of it? Almost certainly the answer is that the guidelines will simply cease to be a point of judicial persuasion. And, given that there are a host of background reasons — including psychological ones — for courts to find for defendants than for plaintiffs who bear the burden of proof (see also the customary fate of plaintiffs in rule-of-reason litigation[13]),[14] then the loss of trust in the guidelines framework would almost certainly be bad news for the agencies and for U.S. merger enforcement.

To be clear: I am not suggesting that antitrust can be fully neutral, or that elections shouldn’t have consequences for antitrust as they do elsewhere. Rather, I am suggesting that judicial confidence in the merger guidelines is a function of the fact that they are not seen as a partisan document. There would be nothing democratically illegitimate about guidelines that could not be expected to survive an election. But I think courts would start to tune out a document of that kind pretty fast, and the result would be the loss of a critical enforcement asset, and almost certainly more merger losses for the agencies in the end.

Most glaringly, the quantification of the structural presumption in particular is a critical tool for winning merger cases. A plausible story about exceeding the presumption (ideally by a margin!) is a tremendous comfort to agencies when deciding to bring a case, and surely also to courts when deciding to impose liability. And while the presumption itself is grounded in PNB,[15] the quantification is not. The identity between the Supreme Court’s “undue” market share and “significant” increase in concentration, on the one hand, and specific HHI numbers, is wholly a creature of judicial trust in the status of the guidelines, and it is only as robust as that trust.

Public remarks suggest that the agencies seem to have some hope that courts will lean more, not less, heavily on the presumption in litigation. In a speech at Georgetown, for example, AAG Kanter recently emphasized the importance of the structural presumption in winning cases: specifically calling out the Heinz baby foods decision as one in which the FTC’s remarkably strong structural case did plenty of heavy lifting.[16] Heinz itself was an extreme case, but I don’t think the point was intended to be limited to such whopping whoppers as a 3:2 with heavy 2:1 vibes (the merging parties in Heinz competed intensively for the “second slot” on the supermarket shelf[17]) and a post-merger HHI of almost 5,300.[18] Rather, I take the point to be that the agencies hope to win more cases by pointing to market structure.

But if the agencies hope that the structural presumption — and specifically the quantification of it in the guidelines — will be doing more work in litigation, then it becomes all the more important to protect the status of the guidelines as a reflection of reasonably clear, reasonably settled principles of law and economic analysis. If courts come to see the next version of the guidelines as a break with “existing law” (however defined) and agency experience, then the litigation value of the presumption will be diluted, not augmented, and more losses seem likely to result. For this reason among others, my use of a magic wand would be a cautious and careful one.

There’s a fair question about what “existing law” really is in the context of merger litigation. The Supreme Court’s last substantive merger decision was in 1975[19] — just after the Expediting Act was amended to substantially narrow the fast-track route to the Court[20] — so merger law today is a creature of the lower courts, with the D.C. Circuit having played a particularly active role in developing it. But I’m not sure that anyone can say with a straight face that “existing law” can be found exactly where the Court left it: Brown Shoe, PNB, Von’s, and all the rest of it. Modern merger law owes at least as much, and maybe more, to Whole Foods, HCA, Baker Hughes, Anthem/Cigna, Yamaha Motor, AT&T/Time Warner, and all the rest of the complicated tapestry — as well as some influential district court cases (H&R Block, Staples, and so on). Merger guidelines that aim to reset the clock back to 1975, or earlier, will probably not be persuasive: either to the lower courts that have taken the project onward since then, or to the current Supreme Court.[21] Accordingly, any revision of the merger guidelines must be able to live fairly comfortably in the realm of modern case law — recognizing that there is a fair margin of room in its interpretation.

I don’t mean by any of this to suggest that I think merger law is in the right place in every respect, or even every important respect. I think that Congress’s intended meaning of “may be” in Section 7 has been pretty clearly lost[22]; Section 7’s prohibition on transactions that “tend to create a monopoly” has been largely ignored by everyone; and the fact that we don’t have reasonable clarity about the role of efficiencies in merger analysis, after more than a century of life under the Clayton Act, is completely wild. But I don’t think any of those things can really be fixed with guidelines.

In Congress, for example, Senators Lee and Klobuchar have made helpful efforts to advance the conversation about revisions to our core antitrust rules, including Sections 2 and 7.[23] I think that conversation is a critically important one, and I regret that the effort to evaluate and update our core antitrust rules seems to have been overtaken by other things.

Of course, developing the law is a central part of the agencies’ role too. The agencies have fought lengthy and successful campaigns to develop the law on hospital mergers, pay-for-delay, the scope of state-action immunity, and a host of other important issues. But the victories on those frontiers weren’t created with guidelines. They were won through litigation — often over many years and across changes in Administrations — as civil prosecutors.[24] And I think the frontier of merger enforcement, if it is to be moved by the agencies, will have to be moved that way too.

In revisiting the guidelines, then, I would aim to work within a plausible reading of existing law. I would aim to fill out the existing text to address some topics that aren’t currently covered, and to adjust existing language to reflect the learning of the last dozen years. In so doing, I would aim for a document that could hope to survive not just the next Presidential election but the next few. (More than that is probably too ambitious.) And I think, maybe perversely, that that may be easier right now that it has been in a long time, because I think it’s true that there is a “new bipartisan consensus” is emerging about antitrust enforcement — even if it’s not the one that some commentators perceive.

My own sense is that a new bipartisan antitrust consensus is forming between the extremes of hard Neo-Brandeisianism and hard Chicago-ism. I take the foundations of that consensus to be something like the following: (1) a recognition that vigorous, principled antitrust enforcement is a necessary precondition for competitive free markets; (2) a recognition that antitrust enforcement should receive more attention, support, and resources than it has in some years (including in particular more funding for enforcement); (3) that on some important margins the content of federal antitrust doctrine has become a little blunter — that is, a bit more pro-defendant — than it should be; (4) in cases where prima facie cases of competitive harm are clear, courts and agencies should not be waved off by airy, broad, or under-proved justifications from defendants; and (5) antitrust enforcement should focus on specific acts and practices that can be shown to harm market participants, including consumers and workers, by reason of lessening competition. I don’t think this consensus supports proposals like digging up the Robinson-Patman Act[25]; authorizing collusion simply because a big player is on the other side of the table[26]; or refocusing competition policy on big-is-bad lines.[27] (And I don’t think there is anything like a consensus, let alone a bipartisan one, about the nature of the specific problems in digital platform regulation or about the desirability of any particular solution. That’s one reason I would prefer to focus on core antitrust reform while we think through the platform regulatory issues: soon with the benefit of experience from the European Union’s own natural regulatory experiment. But that’s another story.)

All of this adds up to a tremendous opportunity for the agencies — or anyone else who comes into possession of a “Magic Wand of Merger Guidelines Amendment.” The goal would be to write within the (somewhat flexible) frame of existing law and within the emerging bipartisan consensus that favors reinforced, vigorous, principled enforcement in the interests of consumers and workers, preserving judicial trust while more fully explaining how the agencies think about merger review.


In light of all that, here are ten quick thoughts about what I would, and wouldn’t, do with the magic wand, in no particular order.

  1. Keep the Horizontal and Vertical Guidelines Separate. As a matter of structure, there are some appealing reasons to integrate vertical issues and horizontal issues in the same document. For one thing, there’s plenty of overlap among the issues (market definition, evidence, entry, and so on). For another, and perhaps more importantly, verticality and horizontality are features of individual theories of harm, not discrete categories of transactions. Accordingly, in an ideal world, I would probably integrate the two sets of guidelines into a single document.

But the world is not ideal, and these are unusual times. The recent experience with the vertical guidelines suggests that vertical guidance may be controversial in a way that horizontal guidance need not be.[28] And if the vertical guidance is destined to be a battlefield, with the prospect of party-line votes for some time to come, then there is probably something to be said from keeping that controversy separated from horizontal guidance, in the hope of preserving the horizontal document, at least, as a point of bipartisan consensus.

  1. Revise the Thresholds (A Little), and Further Clarify Market Definition. Some commentators have suggested that the current concentration thresholds — which trigger the structural presumption at a post-merger HHI of 2,500 and a delta HHI of 200, and which articulate a presumptive safe harbor with a post-merger HHI below 1,500 or a delta HHI below 100 — are set too high.[29] This raises the question of whether they should be adjusted downward.

I think the answer is: yes, modestly, but with limited expectations about the difference this will make. It is worth remembering that the thresholds got moved in the first place to reflect existing agency practice.[30] As Professors Rose and Shapiro have pointed out, “[t]he average post-merger HHI for litigated mergers from 2010 to 2020 was 5805, and the average increase in the HHI was 1938.”[31] It is also worth remembering that the raising of the thresholds may, among other things, reflect a trend in modern practice — noted by Commissioner Wilson and Keith Klovers — toward narrower market definitions,[32] and the increasing use of targeted-customer markets.[33] Moreover, even before the thresholds were moved, courts were not always impressed with a structural case that marginally cleared the mark.[34]

So it’s not really clear that tweaking the thresholds a little is likely to do very much. In practice, after all, it is very seldom the case that structure alone carries the day: some combination of expert evidence, customer testimony, and support from ordinary-course documents is virtually always required.[35] Heinz perhaps represents an exception, as noted above: but that was a remarkably clear structural case that we could think of as halfway between a 3:2 and a 2:1.

There are other reasons, too, to think that a (re)turn to structure will not be a silver bullet for enforcement. For one thing, the agencies have not historically been shy about emphasizing concentration, and the importance of the presumption, when bringing cases. I don’t think previous Administrations have been leaving many structural arguments on the table in merger litigations. For another thing, leaning harder on structure increases the load on market definition, and this might not be a development that would delight enforcers. An array of recent merger losses, from U.S. Sugar to Jefferson/Einstein to Evonik, have re-emphasized the role of market definition as a rock upon which a merger challenge can founder all too easily.[36] For a third, emphasizing the importance of structural evidence, including HHI levels, must be done delicately in order to avoid undermining the viability of unilateral-effects theories.[37]

Indeed, given the centrality of market definition to litigation, and the difficulty of the exercise itself in many cases, some further clarifying language on market definition would be desirable. Among other things, as Professors Rose and Shapiro suggest, it would be helpful to clarify that any candidate market that includes the merging parties and satisfies the hypothetical monopolist test is a valid market for antitrust analysis, even when a broader market definition would also do so.[38] And, as others have suggested, it is probably past time to explain cluster and bundle markets separately and clearly in the guidelines.[39]

Although market definition is likely to be more important in practice, a modest downward adjustment of the presumption thresholds toward 2010 levels — as one way among others of reasonably tightening the framework — seems appropriate. I doubt an adjustment of this kind will make much difference to outcomes,[40] but a downward correction may help to support a bit more enforcement in a sensible subset of cases. My intuition is that doing so will make a small practical difference only in cases that exceed the current (i.e. 2010) concentration thresholds, but only by a modest amount: in such cases, the change will help to reassure a court a little by increasing the margin of comfort. In other words, I think the practical effect of a modest downward adjustment, to the extent there will be one, will be to give a little more confidence to the court in a subset of cases that exceed the 2010 thresholds, and that is probably a good thing on balance. (There may be some room, while adjusting the thresholds, to emphasize the importance of delta HHI, in particular, in understanding the strength of a structural case.[41])

I would not reduce the “red zone” thresholds below their pre-2010 levels. It is one thing to conclude that the relaxation of 2010 has not been vindicated by time, but quite another to dilute the force of the structural presumption by attempting to expand its scope. I doubt that any courts would be persuaded by such a move, and I fear it would undermine the force of the presumption itself across the board.

Some have suggested that the agencies may be wrongly focusing on transactions at very high concentration levels, missing concerns that arise at lower levels.[42] Whether or not this is true, I don’t think this is really a function of the HHI thresholds. Rather, I think it reflects resource constraints and the plentiful supply of transactions that merit investigation. If there are enough very-high-concentration transactions to keep staff engaged, then the center of gravity of enforcement will move to higher concentration levels, with fewer resources left over to challenge the less egregious cases. Again, resources, not guidance, are doing the work. Nevertheless, to the extent that some harmful transactions may be going unchallenged on the margin for want of resources, a modest adjustment in the thresholds now can preserve the possibility of additional enforcement when resources arrive in the future. The balance of error costs seems to tilt in the right direction: it is hard to think there is much risk that the federal courts will rush to over-enforce Section 7 in a slew of weak cases simply because the thresholds have come down by a few hundred HHI points. A modest course correction seems prudent and sensible.

  1. Add Some Discussion of Nascent and Potential Competition. The current merger guidelines are pretty light in their treatment of potential competition (that is, competition among businesses that are not yet in the same market, but where they may be in future) and nascent competition (that is, competition from a business that is in the relevant market, but which is new, emerging, or otherwise growing in competitive significance).

The agencies have amassed considerable experience challenging transactions that raise such issues: in fact, they have been bringing such cases for decades.[43] But they have been particularly prominent on the docket in recent years. In just the last few years, enforcement actions like Harry’s/Edgewell, Billie/P&G, the FTC’s Section 2 case against Meta (involving the acquisitions of Instagram and WhatsApp), Visa/Plaid, Otto Bock/Freedom, Sabre/Farelogix, Illumina/PacBio, Illumina/Grail, and CDK/AutoMate have all turned to some extent on future competition dynamics.

Accordingly, it is probably time to introduce some explicit language into the guidelines to explain how the agencies will think about nascent and potential competition, consistent with existing law. This discussion need not be lengthy or detailed: it would probably be enough to articulate some fundamentals. This could include, among other things:

(1) a statement that the agencies will examine the transaction’s effects on competition over some reasonable time horizon (say, five years);

(2) a statement that, as a merger is reviewed against a counterfactual in which the transaction does not take place, rather than the status quo ante, a transaction may substantially lessen competition within the meaning of the statute even if it does not lead to a less competitive market than existed before the deal, including by eliminating future competition that may otherwise materialize[44];

(3) a statement that the agencies consider that, under existing law, a transaction’s effect “may be” substantially to lessen competition even if the chances of the parties becoming direct competitors cannot be quantified with precision[45] (and, in particular, it may be worth explaining that, under existing law, the agency may conclude that a transaction harms competition even if it cannot conclude that specific, quantified impacts on the outcomes of competition, such as price, are more likely than not[46]); and

(4) a statement that the agencies will analyze effects on future competition by reference to “any reasonably available and reliable evidence,” as in § 4 of the current horizontal guidelines, including the same kinds of evidence used to analyze effects on current competition.

  1. Explain the Agencies’ Approach to Serial Acquisitions. The agencies might take the opportunity to explain how merger review will handle a series of acquisitions in or affecting a single relevant market. It is clear that a business may substantially lessen competition, or tend to create or maintain a monopoly, through serial acquisitions rather than an individual transaction.[47] Indeed, serial acquisitions might make a promising focus for efforts to develop some meaning for “tend to create a monopoly.” (This does not, of course, mean that a particular transaction should be presumed unlawful simply because there is an incipient trend toward consolidation, or indeed vertical integration, in a market.)

In particular, the guidelines might usefully:

(1) state that the agencies consider that, under existing law, a series of acquisitions may substantially lessen competition or tend to create a monopoly even if no individual acquisition might, considered individually, do so;

(2) explain the circumstances under which the agencies will aggregate a series of transactions together for analytical purposes; and

(3) explain the role, if any, of analysis of the merging parties’ conduct in merger review (for example: will the agencies consider, including for the purposes of applying Section 2, whether the transaction forms part of a series of exclusionary practices? or will the agencies regard this as a matter falling outside the scope of merger review and/or the merger guidelines?). Certainly it is probably not desirable to have the merger guidelines turn into Section 2 guidelines: but if the agencies can briefly state the relationship between merger analysis and the merging parties’ conduct, that might help parties understand the relationship between merger enforcement and the broader antitrust function — and the significance, if any, of other practices for the fate of a merger review.

  1. Avoid Industry-Specific Discussions. The agencies evidently have particular interest in certain industries or kinds of markets. Recent policy activity suggests interest, in various ways, in: “tech” markets[48]; pharma markets[49]; and labor markets.[50] As such, there might be some interest in including industry-specific discussions in the guidelines.[51]

My instinct is that this kind of industry- or category-specific discussion is probably undesirable in guidelines, at least for now, for a few reasons. First, guidelines are most likely to endure if they stay out of the weeds of individual markets and practices and focus on principles of general applicability. Otherwise, changes affecting those markets — changes in Administrations, in industry practices, and in substantive case law — are more likely to undermine the accuracy and applicability of the guidelines, aging them faster. General, trans-industrial principles are less likely to be overtaken by events.

Second, it’s not clear that there is much to be said, at least right now, with both specificity and confidence about such markets. “Tech” markets are wildly diverse: e-commerce platforms, apps, ridesharing platforms, social networks, search engines, and so on are much more diverse than they are similar. (However, as noted below, there may be some value in illuminating the application of the merger guidelines to multisided platform businesses — whether tech or otherwise — by focusing on the antitrust economics of platforms, rather than anything about the “tech industry.”)

There are plenty of open puzzles in the merger analysis of labor markets — including puzzles relating to market definition and to the measurement of shares and concentration —that would present challenges for guideline-drafters aiming to capture analytical consensus. Despite what is sometimes supposed, however, the agencies have been thinking about labor competition for some time and have opposed transactions on that basis.[52] Perhaps an easier focus might be expanded treatment of the analysis of buy-side competition. the agencies have extensive experience in thinking about the impact of mergers on buyer competition, and there is plenty of room to add to what is currently Section 12 of the horizontal guidelines.[53] Such a discussion could touch on, but would not be limited to, labor cases.

As for pharma markets: it is not clear whether the FTC’s recent review of pharmaceutical competition has produced a concrete change or evolution in the FTC’s analytical framework. It would be helpful to have that explained clearly in some forum, probably in a speech or (if there is something new to tell) in a separate document focused on pharmaceutical competition. But the merger guidelines are probably not the right forum for a deep dive into pharmaceutical markets.

  1. Briefly Describe the Agencies’ Approach to Data in Merger Review. I am generally skeptical of claims that the presence of data (even if “big”) or the use of algorithmic tools is likely to turn the fundamentals of merger review upside down. But there is tremendous interest, and perhaps some confusion, about the role and significance of data in merger review — and about the role and significance of tools that use data, such as algorithms. So the agencies could probably articulate some basics that would inform the public and policy-makers about how these things are considered in merger review, reflecting the agencies’ long, long record of thinking about data in competition cases.[54]

These might include:

(1) a statement that data may play a variety of roles in a merger review, including as product or service, an input, or a complement;

(2) a statement that analysis of entry (including barriers to entry), will include an analysis of whether particular forms of data are necessary to compete effectively, either in general or for the business of particular key customers, whether the availability of certain kinds of data may facilitate entry, or whether lack of access to such data may hinder entry;

(3) a statement that access by the merging parties to data (including, but not limited to, as a result of the transaction) will be considered in determining whether and to what extent the merged firm will be able to engage in price discrimination, and whether it is appropriate to define a price-discrimination market accordingly; and

(4) a statement that the agencies recognize that whether, and to what extent, access to data may confer or contribute to market or monopoly power is a function of the unique circumstances and competitive realities in an individual market — merely having a lot of data, even unique data, does not mean market or monopoly power!

In other words, the merger guidelines could usefully emphasize that the agencies routinely do, and will, consider data as part of the regular analytical framework for merger analysis, and that regular merger-review principles govern their assessment.

Likewise, given the tremendous interest in (and anxiety regarding) the possibilities of algorithmic coordination, the agencies might reassure the public — and warn businesses — that they consider the role and importance of algorithmic pricing and decision-making when evaluating competitive effects. For example, when evaluating whether and how a market may be vulnerable to coordination, the agencies will examine whether algorithmic tools may play a role in facilitating such coordination.[55]

  1. Briefly Describe the Agencies’ Analysis of Multisided Platforms. The merger guidelines would surely benefit from a short discussion of platform businesses. The agencies have extensive experience in evaluating antitrust issues in connection with platform businesses, from social networks to real estate portals.[56] Again, the discussion need not be lengthy, but it could usefully clarify a few things — particularly in the wake of the Supreme Court’s notorious AmEx decision.[57] In particular, the agencies could take the opportunity to state an understanding that the AmEx decision, on its face, is confined to a very narrow set of circumstances: indeed, not only is the decision limited to transaction platform businesses, but to particular services that exhibit the features emphasized in AmEx itself.[58]

These might include, for example:

(1) a statement that, in cases involving platform businesses, the agencies will define markets on the basis of demand-side substitutability, pursuant to the same principles applied in other markets, except in the narrow set of circumstances in which the products or services supplied on multiple sides of a platform constitute transactional services (e. are jointly and simultaneously consumed in fixed proportions by customers on multiple sides of a platform), such that in economic substance only a single product or service is being supplied to transacting customer pairs;

(2) a statement that, in evaluating transactions involving platform businesses that may harm competition in one relevant market, the agencies will apply the general rule that efficiencies and benefits on another side of the platform will be cognizable only if inextricably linked with the harms and sufficient in magnitude to ensure that the transaction does not substantially lessen competition overall (recognizing that “[i]nextricably linked efficiencies are most likely to make a difference when they are great and the likely anticompetitive effect in the relevant market(s) is small so the merger is likely to benefit customers overall”[59]);

(3) a brief discussion of how merger analysis will be conducted when products or services are supplied at a zero or negative price (perhaps including the outlines of how the agencies think about the application of the HMT to such markets, recognizing that scholarship and practice are continuing to evolve in this respect); and

(4) a statement that a multisided business may compete with other multisided businesses, single-sided businesses, or both.[60]

  1. Close the Merger-Conduct Loophole? For a long time, it has not quite been clear whether and to what extent Section 7 might permit a merger to be challenged on the ground that the effect of the transaction may be to substantially lessen competition by giving the merged firm the ability and incentive to engage in collusive or exclusionary conduct that would itself be unlawful under the antitrust laws. In other words: should merger review analyze whether the deal might make a separate antitrust violation more likely?

I used to think that this kind of thing ought to be outside the bounds of merger control.[61] For one thing, it is not obvious why merger control should be a vehicle for pre-emptively prohibiting something that would, itself, be independently prohibited. For another, it may be very hard to prove that a merged firm — particularly one with an unimpeachable record of compliance — would choose to violate the law. For a third, I am not aware that the agencies have ever brought any merger challenges on such a theory, suggesting that the problem may not be a particularly important one in practice.

But, over the 12-year history of the current guidelines, I have reconsidered that view. One reason is textual: the text of Section 7 makes illegality contingent on harm to competition or tendency to create a monopoly: it does not limit the proscribed means to those that would be independently lawful. Another is doctrinal: much of the conduct we are accustomed to testing for in merger review might violate Sections 1 or 2. For example, coordinated-effects analysis under the guidelines expressly includes an assessment of whether the transaction enables or encourages the formation of a cartel.[62] Likewise, the kind of foreclosure on which vertical merger analysis is customarily focused could be challenged as an unlawful refusal to deal.[63] A third reason is practical: post-merger behavior that violates the antitrust laws may simply go undetected or unpunished. It seems unwise and inefficient to conclude that a transaction will create the ability and incentive to engage in illegality but to blithely insist that we will spot and prevent it if necessary. (It is also not obvious that illegality of conduct, including in particular forms of foreclosure following a vertical transaction, is a binary that we can accurately judge during merger review.) A fourth reason is conceptual: it makes no sense to analyze whether a deal might harm competition while excluding from that analysis a set of behaviors that are so reliably harmful to competition that we prohibit them directly.

More generally, I no longer fear a flood of speculative claims by the agencies, in part because there have been none at all since the language about exclusionary conduct was added in 2010. It will be, and should be, pretty hard to prove that a transaction will give the merged firm the ability and incentive to, say, form a cartel or engage in anticompetitive tying. The antitrust agencies do not have a history of rushing to assert weak theories that undermine the force of their complaints. In cases where the evidence really does indicate that a transaction will encourage harmful and unlawful conduct, the agency should be free to follow it, and the guidelines should state that possibility clearly and crisply.

So I think the guidelines should be clear — clearer than they are — about the fact that, in appropriate cases, merger review will involve an appraisal of whether the merged firm will have the ability and incentive to harm competition including through collusive or exclusionary means that are or may be independently unlawful. In the vast majority of cases, this will be a theoretical prospect only. But not always. Indeed, in Teva/Allergan, the FTC expressly looked for — and did not find — evidence of whether the merged firm would have the ability and incentive to engage in anticompetitive bundling.[64] I think the agency was right to investigate that possibility.

  1. Fill Out the Vertical Discussion. I generally think the existing vertical guidelines are pretty solid, but I’d expand them a bit. I would consider adding, among other things:

(1) some more specific discussion of the analysis of mergers between suppliers of complements, which in the existing document are treated fairly briefly;

(2) a brief explanation about how the agencies will think, during merger review, about commitments offered by the merging parties to trading partners guaranteeing protection from adverse consequences — in the wake of AT&T/Time Warner and Illumina/Grail this seems likely to be a critical issue in future merger reviews (and, as the facts of the private merger challenge in Steves reminds us, a merged firm may be able and willing to foreclose around a pre-existing contract[65]);

(3) as much explanation as possible about how the agencies will evaluate the elimination of double marginalization (including whether the agencies will regard it as an efficiency for which the burden of proof is on defendants, or a component of the analysis of changes in incentives that the agency must undertake to prove harm); and

(4) a slightly fuller discussion of theories of vertical harm, in the interests of completeness and fair notice to merging parties.[66]

  1. Don’t Forget Everything Else. Finally, as I noted above, I think other things are much more important than guidelines as barriers to a suitably vigorous, suitably principled merger enforcement function. So before putting my magic wand down I would take one more run at some of the more pressing issues.

The most important constraint on merger (and conduct) enforcement is the shortage of resources. The agencies desperately need money and staff to meet a workload that has grown at a remarkable rate. From fiscal year 1979 to fiscal year 2021, for example, HSR filings increased from 814 to 3,644: an increase to more than 400 percent of the 1979 level.[67] But in fiscal year 1979 the FTC’s staffing was 1,746 FTE; in fiscal year 2021 it was 1,123: a fall by more than 35 percent from the 1979 level.[68] Those numbers speak for themselves. Changing the guidelines without increasing funding is just not going to increase overall enforcement levels: and given scarce resources, the most harmful (and/or most clearly illegal) transactions will still have to be prioritized, regardless of how the guidelines are recalibrated. If Congress wants more enforcement — including close oversight of tech, pharma, and hospital mergers — it will have to pay for it.

Another critical, if slightly neglected, frontier of merger control’s province is HSR. Although it’s important of course that the merger review process be brisk and efficient — and early termination should be restored for that reason! — agency staff could be given a little more time to analyze the small subset of cases involving a second request, following the parties’ substantial compliance with that request.[69] Having a little more time will help the agencies allocate staff more flexibly across merger matters, and will help to avoid the twin risks that (1) the agency is forced to file a complaint to protect its position when a little more time would have enabled staff to reach comfort, or (2) the agency is forced to let a harmful transaction close for lack of time to complete an investigation build a case. Merging parties and the agencies alike would be better served by a system that allowed the very large set of untroubling deals to close before the end of the waiting period, while that the small subset of most troubling deals is given the time and attention needed to get to the right result. Among other things: a system that effectively requires the agency and the parties to negotiate a timing agreement so that the agency can cover the necessary ground cannot really be anyone’s idea of an ideal process.

Failing an amendment to the Act, perhaps the agencies could think about requiring a little more information in the HSR notification form to help staff triage and spot issues earlier in the process: this would likely help agency staff as well as help businesses navigate the process more efficiently. For example, some basic questions designed to highlight vertical issues might help the agencies to spot vertical issues earlier in the review process, saving everyone some time.



So: that’s what I would do with a magic wand to the merger guidelines. It’s not a whole lot, because I generally think other frontiers are more important, or more in need of agency attention, and because I think the merger guidelines as they stand offer a decent guide to what the agencies actually do. But it’s not nothing, because think the guidelines are certainly important, and twelve years is long enough to merit a refresh and a polish. (We’ve all aged a lot since 2010…) And in light of the increased public attention to antitrust and competition policy, it is probably more important than ever that the guidelines offer an accurate, candid, and up-to-date explanation of what the agencies really do.

[1] Assistant Professor of Law, NYU School of Law; formerly Deputy Director, Associate Director, and Senior Counsel, FTC Bureau of Competition. I am grateful to a number of kind folks for reactions to an earlier draft, including Ian Conner, Mark Meador, Nancy Rose, Doug Ross, and Steve Salop. All views, and all errors are, of course, my own responsibility.

[2] I’m not sure I can think of a case in which a merger enforcement effort was wrongly stymied by the text of the guidelines. (The closest I can think of is Evonik, in which the district court took a rather narrow approach to the role of supply-side substitution, based on what appears to have been a misreading of the guidelines. FTC v. RAG-Stiftung, 436 F.Supp.3d 278, 293–94 (D.D.C. 2020). But even there, I doubt the fault could fairly be laid at the feet of the guidelines.

[3] See, e.g. U.S. Dept. of Justice & Federal Trade Commission, Antitrust Guidelines for the Licensing of Intellectual Property (Jan. 12, 2017); U.S. Dept. of Justice & Federal Trade Commission, Antitrust Guidelines for Collaborations Among Competitors (April 2000).

[4] See generally Carl Shapiro & Howard Shelanski, Judicial Response to the 2010 Horizontal Merger Guidelines, 58 Rev. Indus. Org. 51, 53 (2021) (“[T]he 2010 Guidelines have continued to be well accepted by the courts and to assist the case law’s (slow) incorporation of new economic learning and agency experience in analyzing the impact of mergers on competition.”).

[5] See, e.g. FTC v. Hackensack Meridian Health, Inc., 30 F.4th 160, 167 (3d Cir. 2022) (“We begin our analysis with the Merger Guidelines.”); United States v. Anthem, Inc., 855 F.3d 345, 349 (D.C. Cir. 2017) (guidelines are “a helpful tool, in view of the many years of thoughtful analysis they represent”); Saint Alphonsus Med. Ctr.-Nampa Inc. v. St. Luke’s Health Sys., Ltd., 778 F.3d 775, 784 n.9 (9th Cir. 2015) (“Although the Merger Guidelines are not binding on the courts, they are often used as persuasive authority”) (internal quotation marks and citations omitted).

[6] See, e.g. FTC v. Shkreli, 581 F. Supp. 3d 579, 627 (S.D.N.Y. 2022).

[7] See, e.g. Steves & Sons, Inc. v. JELD-WEN, Inc., 988 F.3d 690, 704 (4th Cir. 2021).

[8] See, e.g. United States v. U.S. Sugar Corp. No. CV 21-1644, 2022 WL 4544025, at *23 (D. Del. Sept. 28, 2022); FTC v. Thomas Jefferson Univ., 505 F. Supp. 3d 522, 539 (E.D. Pa. 2020); United States v. Baker Hughes Inc., 908 F.2d 981, 985 (D.C. Cir. 1990).

[9] See, e.g. West Virginia v. Environmental Protection Agency, 142 S.Ct. 2587 (2022); Alabama Ass’n of Realtors v. Dep’t of Health & Hum. Servs., 141 S. Ct. 2485 (2021).

[10] See, e.g. AMG Capital Management, LLC v. FTC, 141 S.Ct. 1341 (2021); FTC v. Shire ViroPharma, Inc., 917 F.3d 147 (3d Cir. 2019).

[11] United States v. Anthem, Inc., 855 F.3d 345, 349 (D.C. Cir. 2017) (emphasis added).

[12] Carl Shapiro & Howard Shelanski, Judicial Response to the 2010 Horizontal Merger Guidelines, 58 Rev. Indus. Org. 51, 57–58 (2021) (explaining the evolutionary nature of the 2010 changes and noting that “none of [the 2010 guidelines’] ‘new’ concepts were truly novel: All of them had been developing in the economics and antitrust policy literature during the interval between 1992 and 2010, if not before”).

[13] Michael A. Carrier, The Rule of Reason: An Empirical Update for the 21st Century, 16 Geo. Mason L. Rev. 827 (2009).

[14] See, e.g. Louis Kaplow, Replacing the Structural Presumption, 84 Antitrust L.J. 565, 615–19 (2022) (“[W]hen assessing the competing tangle of evidence, if a judge has the sense that the likelihood the government is right is somewhat over 50 percent — but it is all a mess — she may be more comfortable proclaiming and defending in a written opinion the proposition ‘not convinced’ than ‘convinced.’”).

[15] United States v. Philadelphia Nat. Bank, 374 U.S. 321, 363 (1963).

[16] AAG Jonathan Kanter, Respecting the Antitrust Laws and Reflecting Market Realities (remarks of Sept. 13, 2022); FTC v. H.J. Heinz Co., 246 F.3d 708 (D.C. 2001).

[17] Id. at 717.

[18] Id. at 716.

[19] United States v. Citizens and Southern Nat. Bank, 422 U.S. 86 (1975).

[20] See 15 U.S.C. § 29. See also, e.g. R. Hewitt Pate, Antitrust Law in the U.S. Supreme Court (May 11, 2004), https://www.justice.gov/atr/speech/antitrust-law-us-supreme-court (describing post-amendment certifications).

[21] Professor Hovenkamp wrote in 2005 that, “[w]hile antitrust casebooks continue to print 1960s-vintage merger decisions that have never been overruled, no one, not even federal judges and certainly not government enforcement agencies, pay much attention to them.” Herbert J. Hovenkamp, The Antitrust Enterprise: Principle and Execution (2005) 208. That seems rather too strong today — some of those cases are being cited a lot right now — but the core point that the soul of merger law has changed plenty in recent decades is a crucial one.

[22] See, e.g. FTC v. Steris Corp., 133 F.Supp.3d 962 (N.D. Ohio 2015).

[23] See, e.g. Tougher Enforcement Against Monopolists Act, S.R. 2039, 117th Cong. (2021); Competition and Antitrust Law Enforcement Reform Act, S.R. 225, 117th Cong. (2021).

[24] One prominent exception to my point here might be the illegality of wage-fixing, which was previewed in the HR guidelines before DOJ launched its recent round of enforcement efforts. U.S. Dept. of Justice & Federal Trade Commission, Antitrust Guidance for Human Resource Professionals (Oct. 2016). My own sense is that it was reasonably clear that naked collusion on prices among buyers of services, including of labor, was per se illegal before those guidelines were issued: what was novel was the exercise of prosecutorial discretion. Nevertheless, the example is at least arguably in a little tension with my point in the text.

[25] See Alvaro Bedoya, Returning to Fairness (remarks of Sept. 22, 2022), https://www.ftc.gov/system/files/ftc_gov/pdf/returning_to_fairness_prepared_remarks_commissioner_alvaro_bedoya.pdf.

[26] See, e.g. Journalism Competition and Preservation Act, S. 673, 117th Cong. (2021).

[27] Open Markets Institute, Restoring Antimonopoly Through Bright-Line Rules, ProMarket (Apr. 26, 2019) (arguing among other things that Congress should impose a “no fault” ban on durable monopolies, which should be dissolved and ordered “to freely license all their technologies”).

[28] See, e.g. FTC, Press Release, Federal Trade Commission Withdraws Vertical Merger Guidelines and Commentary (Sept. 15, 2021).

[29] See, e.g. Steven C. Salop & Fiona Scott Morton, The 2010 HMGs Ten Years Later: Where Do We Go From Here?, 58 Rev. Indus. Org. 81, 87–89 (2021).

[30] Herbert Hovenkamp & Carl Shapiro, Horizontal Mergers, Market Structure, and Burdens of Proof, 127 Yale LJ. 1996, 2003 (2018).

[31] Nancy L. Rose & Carl Shapiro, What Next For the Horizontal Merger Guidelines?, 36 Antitrust 4 (Spring 2022).

[32] See generally Christine S. Wilson & Keith Klovers, Same Rule, Different Result: How The Narrowing Of Product Markets Has Altered Substantive Antitrust Rules, 84 Antitrust L.J. 55, 87–89 (2021).

[33] See, e.g. FTC v. Wilh. Wilhelmsen Holding ASA, 341 F. Supp. 3d 27, 47–58 (D.D.C. 2018); Federal Trade Commission v. Sysco Corporation, 113 F.Supp.3d 1, 41 (D.D.C. 2015); United States v. Anthem, Inc., 236 F. Supp. 3d 171, 201 (D.D.C. 2017), aff’d 855 F.3d 345 (D.C. Cir. 2017).

[34] In Arch Coal I, for example, the district court noted: “The various measures of market concentration . . . presented by the parties thus reflect an increase in HHI ranging from 49 to 224 [with 49 the best measure appearing to be 49]. . . . The FTC has, therefore, satisfied its prima facie case burden. Nevertheless, it is important to note that this case is not one in which the post-merger increase in HHI produces an overwhelming statistical case for the likely creation or enhancement of anticompetitive market power. . . . [The demonstrated HHI] increases are far below those typical of antitrust challenges brought by the FTC and DOJ. . . . [The HHI increases in recent cases filed by the federal government] dwarf even the highest increase arguably present here. Indeed, between 1999 and 2003, only twenty-six merger challenges out of 1,263 (two percent) occurred in markets with comparable concentration levels to those argued here. Thus, although the FTC has satisfied its prima facie case burden, the FTC’s prima facie case is not strong. Certainly less of a showing is required from defendants to rebut a less-than-compelling prima facie case.” FTC v. Arch Coal, Inc., 329 F.Supp.2d 109, 129 (D.D.C. 2004) (citations omitted).

[35] See, e.g. United States v. Baker Hughes Inc., 908 F.2d 981, 992 n.13 (D.C. Cir. 1990) (“The government does not maximize its scarce resources when it allows statistics alone to trigger its ponderous enforcement machinery”).

[36] United States v. U.S. Sugar Corp. No. CV 21-1644, 2022 WL 4544025, at *19 (D. Del. Sept. 28, 2022) (“The Government has failed to identify the proper relevant market because its product market and geographic markets ignore the commercial realities of sugar supply in the U.S.”); FTC v. Thomas Jefferson Univ., 505 F. Supp. 3d 522, 557–58 (E.D. Pa. 2020) (“Because the Government has not proven that the Northern Philadelphia and Montgomery Areas are relevant geographic markets for GAC services, or that the Philadelphia Area is a relevant geographic market for Acute Rehabilitation Services, the Government has not identified a single relevant market to make its prima facie case”); FTC v. RAG-Stiftung, 436 F.Supp.3d 278, 292 (D.D.C. 2020) (“The Court agrees with Defendants that the FTC has not met its burden of establishing its prima facie case because it has not identified a relevant market within which to analyze the merger’s possible anticompetitive effects.”).

[37] Compare, e.g. United States v. H & R Block, Inc., 833 F. Supp. 2d 36 (D.D.C. 2011) with United States v. Oracle Corp., 331 F. Supp. 2d 1098 (N.D. Cal. 2004).

[38] Nancy L. Rose & Carl Shapiro, What Next For the Horizontal Merger Guidelines?, 36 Antitrust 4, 6 (Spring 2022).

[39] See, e.g. Krisha A. Cerilli, Staples/Office Depot: Clarifying Cluster Markets, Comp. Pol’y Int’l (Aug. 2016); Nancy L. Rose & Carl Shapiro, What Next For the Horizontal Merger Guidelines?, 36 Antitrust 4, 6 (Spring 2022); Carl Shapiro & Howard Shelanski, Judicial Response to the 2010 Horizontal Merger Guidelines, 58 Rev. Indus. Org. 51, 63 (2021).

[40] See generally, e.g. Carl Shapiro & Howard Shelanski, Judicial Response to the 2010 Horizontal Merger Guidelines, 58 Rev. Indus. Org. 51, 64 (2021) (“[W]e are unable to detect any change in merger enforcement associated with this additional transparency.”); Steven C. Salop & Fiona Scott Morton, The 2010 HMGs Ten Years Later: Where Do We Go From Here?, 58 Rev. Indus. Org. 81, 88 & n.18 (2021).

[41] See, e.g. Volker Nocke & Michael D. Whinston, Concentration Thresholds for Horizontal Mergers, 112 Am. Econ. Rev. 1915 (2022).

[42] See generally, e.g. John Kwoka, The Structural Presumption And The Safe Harbor In Merger Review: False Positives Or Unwarranted Concerns?, 81 Antitrust L.J. 837, 865–71 (2017) (noting that merger challenges between 1996 and 2011 shifted to a focus on higher-HHI transactions).

[43] The cases are numerous and many are well known. For a selection, see, e.g. Statement of the Federal Trade Commission, In the Matter of Nielsen Holdings N.V. and Arbitron Inc., FTC File No. 131 0058 (Sept. 20, 2013); Bureau of Competition Statement Regarding the Announcement that Thoratec Corporation Will Not Proceed with Its Proposed Acquisition of HeartWare International (July 30, 2009); Tenneco, Inc. v. FTC, 689 F.2d 346, 351 (2d Cir. 1982); Yamaha Motor Co., Ltd. v. FTC, 657 F.2d 971 (8th Cir. 1981); United States v. Marine Bancorporation, Inc., 418 U.S. 602, 624 (1974); United States v. Falstaff Brewing Corp., 410 U.S. 526 (1973).

[44] This concept is articulated in the current guidelines, but the implications for analysis of nascent and potential competition are not brought out clearly and succinctly. See U.S. Department of Justice & Federal Trade Commission, Horizontal Merger Guidelines (2010) § 1; see also § 4.1.2; § 5.2; § 6.4.

[45] See, e.g. Brown Shoe Co. v. United States, 370 U.S. 294, 323 & n.39 (1962) (stating that “[m]ergers with a probable anticompetitive effect were to be proscribed by this Act,” and quoting a Senate Report indicating that the language “may be” “would not apply to the mere possibility but only to the reasonable probability of the [proscribed] effect”); United States v. Baker Hughes Inc., 908 F.2d 981, 984 (D.C. Cir. 1990) (“Section 7 involves probabilities, not certainties or possibilities.”); Hospital Corp. of America v. FTC, 807 F.2d 1381 (7th Cir. 1986) (Section 7 requires “appreciable danger of [adverse] consequences in the future”); United States v. H & R Block, Inc., 833 F.Supp.2d 36 (D.D.C. 2011) (quoting authorities stating that a plaintiff must show a merger is “reasonably likely to cause anticompetitive effects,” that this is a matter of “probabilities, not certainties,” and that an “appreciable danger” of effects suffices)

[46] See, e.g. United States v. AT&T, Inc., 916 F.3d 1029 (D.C. Cir. 2019) (“[T]he court does not hold that quantitative evidence of price increase is required in order to prevail on a Section 7 challenge[.]”); Hospital Corp. of Am. v. FTC, 807 F.2d 1381, 1389 (7th Cir. 1986) (“Section 7 does not require proof that a merger or other acquisition has caused higher prices in the affected market. All that is necessary is that the merger create an appreciable danger of such consequences in the future. A predictive judgment, necessarily probabilistic and judgmental rather than demonstrable . . . is called for.”) (citation omitted); FTC v. Hackensack Meridian Health, Inc., Case No. CV 20-18140, 2021 WL 4145062, at *20 (D.N.J. Aug. 4, 2021) (“An anticompetitive effect means that the transaction would create an incentive for the merged parties to raise prices, lower quality, and innovate less.”) (emphasis added).

[47] See, e.g. United States v. Grinnell Corp., 384 U.S. 563, 576 (1966).

[48] See, e.g. FTC, Press Release, FTC to Examine Past Acquisitions by Large Technology Companies (Feb. 11, 2020).

[49] See, e.g. FTC, Press Release, Multilateral Pharmaceutical Merger Task Force Seeks Public Input (May 11, 2021).

[50] See, e.g. U.S. Dept. of Justice, Press Release, Department of Justice Antitrust Division and Federal Trade Commission to Hold Workshop on Promoting Competition in Labor Markets (Oct. 27, 2021).

[51] See, e.g. U.S. Dept. of Justice & FTC, Request for Information on Merger Enforcement (Jan. 18, 2022).

[52] For one example, see, e.g. FTC, Statement of the FTC Chairman Regarding Announcement that Aveanna Healthcare and Maxim Healthcare Services Have Terminated Their Acquisition Agreement (Jan. 30, 2020) (noting that both patients and nurses would continue to benefit from competition following the abandonment of the proposed transaction).

[53] See generally, e.g., Scott Hemphill and Nancy Rose, Mergers that Harm Sellers, 127 Yale L.J. 2078 (2018).

[54] See, e.g. Complaint, In the Matter of Verisk/EagleView, Dkt. No. 9396 (Dec. 16, 2014); Complaint, In the matter of CoreLogic, Inc., Dkt. No. C-4458 (May 20, 2014); Complaint, In the Matter of Reed Elsevier NV; Complaint, FTC v. Hearst Trust and First Databank, Inc., No. 1:01-CV-734 (D.D.C. filed Apr. 4, 2001).

[55] See generally, e.g. Ai Deng, Algorithmic Collusion and Algorithmic Compliance: Risks and Opportunities, in Douglas H. Ginsburg & Josh Wright (eds.) Global Antitrust Institute, Report on the Digital Economy (2020); Francisco Beneke & Mark-Oliver Mackenrod, Remedies for algorithmic tacit collusion, 9 J. Antitrust Enforcement 152 (2021); Michal Gal, Algorithms as Illegal Agreements, 34 Berkeley Tech. L.J. 67 (2019); OECD, Algorithms and Collusion: Competition policy in the digital age (2017); FTC Acting Chairman Maureen K. Ohlhausen, Should We Fear The Things That Go Beep In the Night? Some Initial Thoughts on the Intersection of Antitrust Law and Algorithmic Pricing (speech of May 23, 2017). See also, e.g. U.K. Competition & Markets Authority, Pricing algorithms: Economic working paper on the use of algorithms to facilitate collusion and personalised pricing (October 2018).

[56] See, e.g. First Amended Complaint, FTC v. Facebook, Inc., Case No. 1:20-cv-03590 (D.D.C. filed Aug. 19, 2021); Complaint, FTC v. Surescripts, Inc., Case No. 1:19-cv-01080 (D.D.C. Apr. 17, 2019); In the Matter of DraftKings, Inc., Dkt. 9375 (complaint filed 19 June 2017); Complaint, In the Matter of CoStar Group, Inc./RentPath Holdings, Inc., Dkt. No. 9398 (Nov. 30, 2020); Statement of Commissioner Ohlhausen, Commissioner Wright, and Commissioner McSweeny Concerning Zillow, Inc./Trulia, Inc., FTC File No. 141-0214 (February 19, 2015).

[57] For a fuller (critical) discussion, see Steven C. Salop, Daniel Francis, Lauren Sillman, and Michaela Spero, Rebuilding Platform Antitrust: Moving On from Ohio v. American Express, 84 Antitrust L.J. (forthcoming 2022).

[58] For a fuller discussion, see Daniel Francis, Proving Antitrust Effects (manuscript on file).

[59] U.S. Department of Justice & Federal Trade Commission, Horizontal Merger Guidelines (2010) § 10 n.14.

[60] See, of course, United States v. Sabre Corp., 452 F.Supp.3d 97 (D. Del. 2020) (vacated as moot) (missing this point).

[61] See D. Bruce Hoffman & Daniel Francis, Including Exclusion in the 2010 Horizontal Merger Guidelines, 10 Antitrust Source (October 2010).

[62] Horizontal Merger Guidelines § 7 (“Coordinated interaction can involve the explicit negotiation of a common understanding of how firms will compete or refrain from competing. Such conduct typically would itself violate the antitrust laws.”).

[63] See, e.g. Viamedia, Inc. v. Comcast Corp., 951 F.3d 429 (7th Cir. 2020) (refusal to deal challenge following acquisitions).

[64] Statement of the Federal Trade Commission, In the Matter of Teva Pharmaceuticals Industries Ltd. and Allergan plc, FTC Dkt No. C-4589 (July 27, 2016).

[65] Steves and Sons, Inc. v. JELD-WEN, Inc., 988 F.3d 690 (4th Cir. 2021).

[66] For some helpful reflections, see generally Steven C. Salop, A Suggested Revision of the 2020 Vertical Merger Guidelines (December 2021).

[67] https://www.ftc.gov/enforcement/premerger-notification-program.

[68] https://www.ftc.gov/about-ftc/bureaus-offices/office-executive-director/financial-management-office/ftc-appropriation.

[69] 15 U.S.C. § 18a(e).