Cory S. Capps and Leemore Dafny provide their round-one comments on the draft Merger Guidelines.
To read more from the ProMarket Merger Guidelines Symposium, please see here.
Cory: Leemore, as I eagerly read the draft Merger Guidelines, I wonder what they may accomplish. Clearly, the Agencies (the Federal Trade Commission and Department of Justice Antitrust Division) are informing the business and antitrust communities that they intend to investigate more transactions and file more merger challenges. This should work in the desired direction: if mergers have lower expected returns due to increased financial and time costs of review, a higher probability of litigation, and possibly a greater likelihood of being blocked, then we should expect to see fewer mergers. Lower expected returns may deter mergers of lesser strategic value to the parties, including mergers with few efficiencies and/or small or no anticompetitive effects. Deterring the former is probably good, while deterring the latter is not clearly so. Given that merger review and litigation costs have a heavy fixed component, any deterrent effect will be less for larger transactions. Raising the bar on a large swath of transactions is a blunt instrument that may not, on its own, deter the mergers the Agencies most want to deter. However, if the draft Merger Guidelines spark a change in how judges weigh theories of harm and evidence—an uncertain proposition to me at this early stage—then the deterrent effect could become more targeted.
Second, the Agencies seem to be apprising courts that the case law governing mergers, and possibly the economics, should be viewed as much more skeptical of mergers than litigation outcomes have reflected over the last few years, or perhaps the last few decades. However, the Agencies already apprise courts overseeing merger challenges of relevant legal precedents—in complaints, opening arguments, closing arguments, and briefs. For example, in St. Luke’s-Saltzer, the Ninth Circuit Court sided with the FTC’s skepticism of the efficiency defense in general and in that specific case, and cited Heinz, Brown Shoe, and Procter & Gamble, all three of which the draft Merger Guidelines reference multiple times. Will additionally putting those citations and arguments into the Merger Guidelines make that more convincing or effective? I have my doubts, but I’m curious what you think.
Leemore: I view the Merger Guidelines, old and new, as a living document that requires periodic revision and updating, even if the laws they seek to enforce are unchanged. This draft refreshes and expands on the 2010 Horizontal Merger Guidelines—both by citing case law and by providing a more comprehensive and detailed list of types of harm that may render a merger anticompetitive. The draft Merger Guidelines capture the Agencies’ view—shared by the Biden administration—that Type II errors (anticompetitive transactions proceed) are a greater risk to the competitiveness of the economy than Type I errors (transactions that are not anticompetitive are abandoned or blocked). Given the current state of reduced competition in multiple sectors of the economy, I think that’s a sensible approach. It’s not without casualties or costs, as you observe, but the costs of under-enforcement are real, too.
I have no idea whether citations to case law in the draft Merger Guidelines will make it easier for judges who wish to adopt this view to do so, but there isn’t much downside if there is “free disposal”—that is, if the citations don’t constrain enforcers.
Now, I do see some concrete areas where the draft Merger Guidelines break new ground and reflect a change in the Agencies’ public stance. Two topics that stand out are serial acquisitions and non-horizontal mergers.
Let’s start with serial acquisitions. The draft Merger Guidelines throw down the gauntlet when it comes to sequences of transactions that may accumulate to an anticompetitive outcome: Guideline 9 explicitly rejects a narrow interpretation of Section 7 of the Clayton Act that any transaction should be considered on its own. In healthcare markets, this is particularly meaningful for private equity rollups of physician groups and facilities and for hospital systems’ smaller acquisitions.
First, many such transactions fall below the transaction size threshold above which the Hart-Scott-Rodino Act requires premerger reporting, so it’s hard for Agencies to investigate them before they close. Second, this statement reminds would-be acquirers that the Agencies can sue to unwind deals after they are consummated. Third, the statement amplifies the risk of such a challenge by clarifying that the cumulative share realized via serial acquisitions will be considered. A recent report found that a single private equity firm has a 30 percent or higher share of physicians in one or more specialties for over one-quarter of local markets. The report also corroborates other studies showing price increases following private equity acquisitions. One of the (few) advantages of targeting consummated transactions is that anticompetitive effects no longer need to be predicted, just parsed out from among the various factors driving observed outcomes. Relatedly, many hospital systems have slowly acquired more physician practices and subsequently raised prices for both physician services and inpatient care. Guideline 9 makes clear that the pace and size of individual acquisitions will not impede the Agencies’ pursuit of a challenge.
Cory: I agree on serial acquisitions. For a sequence of related transactions, the piecewise size and order do not affect the economics and it makes sense to assess them cumulatively—at least when the acquisitions are near in time or part of a sequence where the acquirer controls the order. For example, suppose three firms have the following shares: A—13%, B—6%, C—4%. If A adds B and then C, there’s no violation under a 200 point HHI increase threshold. But if A adds C then B, the HHI increase exceeds 200. Viewing this as A acquiring both B and C—the actual final outcome—there’s no issue on sequencing.
I’ve seen acquisition sequencing come up in a case and the court chose to examine two proximate acquisitions separately. I’ll add that serial acquisitions often leave a track record, whether pro- or anti-competitive, that would likely weigh heavily in enforcement decisions and litigation. Interestingly, Guideline 9 doesn’t directly call for retrospective evaluation of serial acquisitions to inform the competitive effects analysis of the next transaction; instead the focus is on the acquirer’s practices and incentives. More generally, the draft Merger Guidelines have changed the language regarding past transactions to remove references to “natural experiments” (compare Appendix 2 of the draft Merger Guidelines with Section 2.1.2 of the 2010 Horizontal Merger Guidelines). Meanwhile, the draft Merger Guidelines have added a proviso that an absence of observed adverse effects from a consummated transaction does not rule out lessened competition. This proviso seems oriented towards challenging a consummated transaction, rather than looking to outcomes of past mergers in the same industry or by the same party for guidance in evaluating a prospective merger.
More generally, the 2010 Horizontal Merger Guidelines and the draft Merger Guidelines both include a statement to the effect that “[e]vidence of observed post-merger price increases or worsened terms is given substantial weight” (compare Section 2.1.1 in the former with Appendix 1 in the latter). However, whereas the 2010 Horizontal Merger Guidelines state clearly that “[t]he Agencies look for historical events, or ‘natural experiments,’ that are informative regarding the competitive effects of the merger,” the draft Merger Guidelines now say only that the Agencies “may” consider such evidence for the related but distinct purpose of “assess[ing] the presence and substantiality of direct competition between the merging firms.” This seems to mean the Agencies would give little weight to evidence that similar transactions in the past or in different geographies do not result in higher prices or other adverse effects.
I believe you also mentioned non-horizontal transactions, Leemore?
Leemore: I did indeed, as I am wont to do. An interesting subsection in Guideline 7 (“Mergers Should Not Entrench or Extend a Dominant Position”)is the final paragraph, entitled “Extending a Dominant Position into a Related Market.” It charts new territory regarding combinations of entities in different, but related, markets. It explains that if tying/bundling/otherwise linking sales across markets could extend a dominant position held by one of the merging entities, the Agencies will consider the risk to competition in the related market (i.e., the tied market). This is distinct from weakening, excluding, or foreclosing rivals within the dominant firm’s market, which Guidelines 5 and 6 address. Per Guideline 7, even if the merging firms would not find it possible or optimal to foreclose rivals after combining—e.g., based on the structure of the industry at hand or the predictions of “vertical math”—the Agencies will still consider the risk to competition if the transaction could extend the dominant firm’s market power to another market.
This subsection clarifies that non-horizontal and non-vertical (e.g., “cross-market”) mergers may lessen competition and give rise to merger challenges. Imagine that an academic medical center buys a community hospital (or system of hospitals) outside its catchment area, or a leading personal computer manufacturer purchases a printer manufacturer. These combinations in “adjacent” or related markets might not violate the other guidelines, but they could facilitate entrenchment of a dominant merger partner or extension of dominance into a related market. This subsection is a welcome addition in light of research showing that cross-market mergers are driving higher spending in the U.S. hospital industry and similar concerns in the pharmaceutical and technology industries.
Cory: I agree this subsection of Guideline 7 has the language most likely to be applicable to cross-market transactions, though there is no direct mention of the term. Interestingly, it could be read as providing a safe harbor for cross-market transactions in which the larger of the merging parties has no more than a 30 percent market share, one of the two potential indicators of a dominant firm under Guideline 7.
Leemore: While I doubt the authors intended to put forth safe harbors, I do think Guideline 7 (“Mergers Should Not Entrench or Extend a Dominant Position”) would benefit from some guardrails to enhance its effectiveness. Guideline 7 states, “When one merging firm has or is approaching monopoly power, any acquisition that may tend to preserve its dominant position may tend to create a monopoly in violation of Section 7” (emphasis added). It is among the boldest of the draft guidelines and leans toward the call by some commentators for a presumptive ban on mergers and acquisitions by dominant firms. Is that the intent? Either way, this is an area for the Agencies to clarify as they work on the final Merger Guidelines.
Cory: The draft Merger Guidelines do seem to have a mantra: potential entrenchment and extension of market power are to be thwarted to the maximum extent possible. I count 36 variants of entrench and 17 instances of extend in the draft Merger Guidelines versus one entrench and zero extend in the 2010 Horizontal Merger Guidelines (wherein Section 1 states, “The unifying theme of these Guidelines is that mergers should not be permitted to create, enhance, or entrench market power or to facilitate its exercise.”). Whether there is a limiting principle comes down to what “entrenching” or “extending” a dominant position means. Guideline 7 defines entrenchment as “any mechanism consistent with market realities that lessens the competitive threats the merged firm faces” and gives five examples. Extension is described as something that “could enable the merged firm to . . . substantially lessen[] competition in the related market” and the examples, as you note, are “tying, bundling, conditioning, or otherwise linking sales of two products, excluding rival firms.” In practice, the Agencies have struggled to win vertical merger cases and a challenge based on extending dominance seems like an even steeper hill.
I share your concern that parts of Guideline 7 could imply an overly strong presumption against expansion through acquisitions by firms that meet the definition of “dominant.” The new Merger Guidelines should not erect barriers to entry! Suppose, for example, that a large technology platform other than Meta had entered against Twitter by acquiring one of the various small Twitter alternatives. Would it advance competition for the Agencies to challenge such an acquisition?
Cory: Will you assign the 2010 Horizontal Merger Guidelines or the draft Merger Guidelines as reading for your classes this fall?
Leemore: Bullet points are appealing to busy students, so I’ll go with the handy two-page summary on pages two–three of the draft Merger Guidelines.
Author Disclosure: Leemore Dafny, Bruce V. Rauner Professor of Business Administration, Harvard Business School and Professor of Public Policy, Harvard Kennedy School. Leemore is also a Partner with Bates White Economic Consulting and serves as an expert in litigated and non-litigated merger cases on behalf of state and federal agencies as well as on behalf of private entities with a financial interest in the Merger Guidelines.
Author Disclosure: Cory Capps, Partner, Bates White Economic Consulting. Cory frequently serves as an expert in litigated and non-litigated merger cases on behalf of state and federal agencies as well as on behalf of private entities with a financial interest in the Merger Guidelines.
Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.
