In contrast to these earlier presumptions, recent economic studies show that the weight of the evidence indicates that horizontal mergers do not reduce costs in a significant percentage of high-concentration mergers. We are surprised that the drafters of the Guidelines did not credit this more recent empirical evidence. Even if some horizontal mergers lead to cost savings, there are theoretical economic reasons to doubt that such savings can be shown to be merger specific and verifiable in advance of the merger, or that they are passed on to consumers. The benefits from a potentially expensive and usually unpredictable efficiency inquiry are therefore not likely to result in a significant social benefit. Moreover, as many economists have pointed out, the Guidelines already provide merging parties with an efficiency “standard deduction” embodied in the Guidelines’ HHI thresholds.
In sum, the Guidelines have not been completely faithful to their promise to take textualism, legal precedent, and modern economics seriously. While the Guidelines narrowed the efficiency rebuttal, there was no cause to retain it.
Textualism and the Efficiency Rebuttals for Horizontal Mergers
The Guidelines’ efficiency rebuttal section cannot be justified by a textual analysis of the Clayton Act. Textualism is a method of statutory analysis that starts with, and in almost all cases ends with, the precise words of a statute. It gives the words and phrases in a statute the plain, fair, straightforward, and ordinary meanings they had when the statute was enacted, and it almost always ignores everything else, including a statute’s legislative history. Courts that apply textualism would not create rebuttals that are not found in the text of the statute. Nor could they inject their own policy preferences into the interpretation of a statute. A majority of current Supreme Court Justices often use textualism to interpret statutes.
A textualist approach to the merger efficiency rebuttal question entails a plain and straightforward reading of the Clayton Act and asks whether an efficiency rebuttal, defense, or exception appears in the words of the statute. It does not. The Clayton Act forbids all mergers that “may be substantially to lessen competition or to tend to create a monopoly”; it does not include any form of an efficiency rebuttal.
It would have been simple for Congress to include an efficiencies rebuttal or defense in Section 7 of the Clayton Act. For example, the 1936 Robinson-Patman Act amendment to the Clayton Act contains one: This statute forbids price discrimination whose effect “may be substantially to lessen competition or to tend to create a monopoly” (the exact language contained in Section 7). But Section 3 provides an explicit efficiency defense, if defendant can show that its pricing was caused by a lower “cost of manufacture, sale or delivery.…”
Because Congress in 1936 included an efficiencies defense in Section 2 of the Clayton Act, but not in Section 7 of the original Clayton Act or when it passed the Celler-Kefauver Amendment in 1950, Section 7’s lack of an explicit efficiencies rebuttal or defense, would decide the question for a true textualist, even if one personally considered it desirable from a public policy perspective.
Although a sound textualist analysis should stop at this point, we briefly consider further each half of Section 7’s prohibition against mergers that “may be substantially to lessen competition or to tend to create a monopoly.” There is, at best, very limited support for an efficiency rebuttal.
The May “Tend to Create a Monopoly” Clause
To a textualist, this clause is clear. A plain reading of this part of Section 7 precludes an efficiencies rebuttal so long as the merger tends to create a monopoly with no consideration as to whether that monopoly would be likely to be “efficient” (i.e., to have lower costs).
The plain language of Section 7 forbids all mergers that would even tend to create a monopoly—let alone actually create one—regardless of whether the resulting firm might have lower costs, and without even examining whether prices were predicted to rise, fall, or remain stable. Perhaps the plain language of this clause is what caused the drafters of the Guidelines to provide that the enforcers will not recognize an efficiencies defense for mergers that may tend to create a monopoly.
If this is the case, the Guidelines should be explicit about their reasoning. For example, there is not even a footnote to explain why an efficiency rebuttal is precluded for mergers falling under the “tend to create a monopoly” clause yet remains available to those analyzed under “may be substantially to lessen competition.” It is odd for the Guidelines not even to have a section on this part of the statute.
The “May Be Substantially to Lessen Competition” clause
Section 7 contains a potentially plausible textual argument for an efficiency rebuttal for mergers that only “may be substantially to lessen competition” but do not “tend to create a monopoly.” A textualist analysis of the “may be substantially to lessen competition” clause would center around what the word “competition” meant in the English language dictionaries and legal dictionaries from the 1900–1950 period. Fortunately, Justice Scalia’s treatise on textualism provided a list of such sources that he considered “useful and authoritative.” These definitions are the same ones we use today. Each dictionary defined competition in terms of rivalry or acts of rivalry. None defined “rivalry” in terms of a net balancing of harm, or efficiency, or defined more “rivalry” as achieving lower costs through the merger.
Suppose a merger would reduce the number of rivals in a market from four to three. Using a straightforward definition of “rivalry,” a textualist could conclude that the amount of rivalry (i.e., competition) would likely decrease, even if the merged firm might experience lower costs, because on average three firms compete less—engage in less rivalry or competition—than four.
However, it is possible that the drafters of the Guidelines had in mind a situation where a merger of two smaller firms resulted in cost savings that made the new entity a stronger rival—and therefore could increase rivalry although with fewer competitors. But if this is the logic behind the Guidelines’ efficiency rebuttal, why not simply provide that defendants must come forward with evidence that their merger fits this potential exception? Indeed, if this is the thinking behind the efficiency rebuttal, the Guidelines should explain how this theory can be reconciled with the fact that Congress chose not to include an explicit efficiencies rebuttal in the statute, in contrast to Congress’s decision to include one in another part of the Clayton Act. We see little evidence that the Guidelines are truly taking statutory textualism seriously when they contain an efficiency rebuttal for the section of Section 7 that affects most horizonal mergers.
Legislative History and Concern for Efficiency
In contrast to textualism, a traditionalist approach to statutory interpretation does interpret laws in light of their legislative history, especially when the statutory language is ambiguous. A traditionalist analysis shows that when Congress debated the antimerger laws, it showed almost no concern for missed or foregone efficiencies from mergers. Nor have scholars advocating an efficiency orientation to the legislation produced any such evidence.
Professor Derek Bok’s seminal analysis of the Celler-Kefauver Act’s legislative debates observed that “none of the justifications for mergers by big companies were accorded any significance by Congress. Efficiency, expansion, and the like were ignored or simply brushed aside in the deliberations.” He concluded: “There is little basis for concluding that the achievement of lower costs as such should give rise to favored treatment under Section 7.”
Our reading of the legislative history accords with that of Professor Bok. The view of the Celler Kefauver Amendment’s proponents, and thus of the majority in Congress, was that this law, by preventing many mergers, probably would help, not hurt, corporate efficiency. The Clayton Act’s legislative history is thus consistent with the textualist reading of the law, which showed no concern for efficiencies. Nor do the Guidelines cite Congressional intent to justify the Guidelines’ efficiency rebuttal.
The Supreme Court and the Efficiency Rebuttal
The Guidelines do explicitly highlight that the Supreme Court has declined to endorse an efficiency defense. But the language in these cases show that the Supreme Court’s reasons for disapproving an efficiency defense would apply equally to an efficiency rebuttal. In Philadelphia National Bank, the Supreme Court stated:
We are clear, however, that a merger the effect of which “may be substantially to lessen competition” is not saved, because, on some ultimate reckoning of social or economic debits and credits, it may be deemed beneficial. A value choice of such magnitude is beyond the ordinary limits of judicial competence, and in any event has been made for us already, by Congress when it enacted the amended Section 7.
Although the Court thought economies from mergers were possible, it opted to forego the extraordinarily difficult task of balancing market power and efficiencies because Congress was much more concerned with the problem of market power. Indeed, Richard Posner, who acknowledged that he essentially drafted the Philadelphia National Bank opinion when he was a law clerk, is a longtime critic of an efficiencies defense for these same reasons: “I would not allow a generalized defense of efficiency”. He reiterated this view in 2015, and the Supreme Court adopted identical reasoning in FTC. v. Procter & Gamble Co.: “Possible economies cannot be used as a defense to illegality. Congress was aware that some mergers which lessen competition may also result in economies but struck the balance in favor of protecting competition.”
This language from Procter & Gamble confirms that the Supreme Court elected not to consider efficiencies at all. If one were to maintain that the Court thought that efficiencies could be used as a rebuttal as part of the overall competitive analysis, but simply not as an absolute defense, then why did it not say as much? Instead, the Supreme Court’s analysis of efficiencies in Brown Shoe makes it even less plausible that the Court would have endorsed an efficiency rebuttal: “Congress appreciated that occasional higher costs and prices might result from fragmented industries and markets. It resolved these competing considerations in favor of decentralization.”
Here, the Court essentially is explaining that even if mergers can lower costs and prices Congressional intent mandates that mergers that increase concentration should be blocked. The Brown Shoe analysis leaves no room for an efficiency rebuttal. And no subsequent merger case decided by the Supreme Court has overturned this precedent.
So why has Supreme Court precedent been ignored? In Baker Hughes, Judge (now Justice) Clarence Thomas wrote that in “the wake of General Dynamics, the Supreme Court and lower courts have found Section 7 defendants to have successfully rebutted the government’s prima facie case by presenting evidence on a variety of factors other than ease of entry.” But Judge Thomas did not suggest that the Supreme Court meant to overturn precedent and include efficiencies in this consideration. Likewise, General Dynamics itself does not involve or discuss an efficiency rebuttal. Yet, much more recently, Judge (now Justice) Kavanaugh boldly declared in his dissent in United States v. Anthem, Inc. that efficiencies must be considered: “The case law of the Supreme Court and this Court, as well as the Government’s own Merger Guidelines, establish that we must consider the efficiencies and consumer benefits of a merger together with its anti-competitive effects.
Judge Kavanaugh is simply incorrect in his claim the Supreme Court ever mandated that a merger’s efficiencies must be considered “together with its anti-competitive effects.” Thus, the Guidelines drafters would not be justified in following Judge Kavanaugh, and the Guidelines do not offer any other explanation for citing Supreme Court precedent, but not actually following its mandate.
Lower Court Opinions and the Justifications for an Efficiency Rebuttal
The Guidelines’ drafters also cannot rely on lower court precedent to justify including an efficiency rebuttal. Ironically, the role of efficiencies in merger analysis was first introduced by earlier versions of the Guidelines, not by the lower courts. “The lower courts [then] enthusiastically relied on the merger guidelines” to support an efficiency rebuttal.
The earliest lower court opinion to accept an efficiency rebuttal was FTC v. University Health, Inc. in 1991, drawing support from the 1982 Merger Guidelines. To reconcile the 1982 Guidelines position with the Supreme Court, the lower court speculated that despite precedent, the Supreme Court might have considered an efficiency defense if a good case was presented to it.
Several Circuits then followed University Health in allowing an efficiency rebuttal. But each case merely cited to the others and to the then-current version of the merger guidelines to support the proposition that they must consider efficiencies. Conspicuously absent is any independent textualist reasoning, legislative history, or economic analysis. For example, the D.C. Circuit in FTC v. Heinz merely noted: “Although the Supreme Court has not sanctioned the use of the efficiencies defense in a section 7 case . . . the trend among lower courts is to recognize the defense.” The Sixth Circuit, without reference to Supreme Court merger cases, found that the goal of antitrust was to maximize consumer welfare and noted that efficiencies are part of welfare. Finally, the Ninth Circuit discusses at length the lack of an efficiencies defense in the Supreme Court decisions, does not point to one in the statute, but nonetheless concurs with sister circuits that defendants “can rebut a prima facie case with evidence that the proposed merger will create a more efficient combined entity and thus increase competition.”
Only three lower court opinions have both entertained an efficiency rebuttal and found significant efficiencies: FTC v. Butterworth Health, New York v. Deutsche Telecom, and JetBlue/Spirit. In all three cases, the efficiencies were not dispositive. (We infer from this that there are very few significant horizontal merger efficiencies.) At the appellate level, no defendant has successfully been able to rebut a prima facie case by demonstrating significant efficiencies.
It is also perplexing that many lower courts describe what they are doing as incorporating an efficiency defense rather than an efficiency rebuttal. For example, in Illumina Incorporated v. FTC, the court noted that the FTC’s rejection of Illumina’s efficiency claims as not cognizable was supported by substantial evidence, noting that an “efficiency defense is very difficult to establish.” Indeed, in many instances courts use the terms “defense” and “rebuttal” interchangeably.
Finally, it is revealing that we have not been able to locate any lower court opinions that justify an efficiency rebuttal using textualist analysis or the legislative history of the Clayton Act or that have seriously analyzed the Supreme Court precedent. Nor did any of these decisions contain any economic analysis similar to that which will be presented in the following section.
Economics of the Efficiency Rebuttal
While the legal arguments should be decisive, even advocates of an efficiency rebuttal have been unable to demonstrate reliable support for horizontal merger efficiencies in modern economic theory or recent empirical analyses.
Welfare Economic Assumptions of the Consumer Welfare Standard
Current leadership of the FTC and DOJ have been very critical of the consumer welfare standard, both generally and as it relates to merger enforcement. Yet, by including an efficiency rebuttal in the Guidelines, they implicitly endorse the consumer welfare standard without a sound economic basis for doing so. Eric Posner states the problem as follows:
The social costs of a merger encompass all the harms that result when one firm acquires another. Economists have always counted just one—the deadweight loss that results if the merged firm raises prices (net of any producer benefit in the case of the total surplus test)….[N]umerous other social costs could result from a merger; many of those social costs motivated Congress to enact section 7 and its amendments and continue to influence public policy debates. Why these social costs were disregarded in enforcement is a question for another day, but the answer is likely related to the mismatch between the field of industrial organization–which focuses on the relationship between market structure and prices–and merger policy, with its broader scope.
Unpacking Eric Posner’s criticism and sharpening the point, requires a brief explanation of some key welfare economics concepts. Unlike businessmen or physicists, when economists use the term “efficiency,” they mean greater social welfare. But when the Guidelines discuss efficiencies, they historically have referred to cost savings. To conceptually link cost savings and economic efficiency requires two normative economic assumptions. The first assumption is that the consumer welfare standard controls. It mandates that the only relevant effects of anticompetitive conduct are those on consumer surplus and producer surplus (which is Posner’s point about deadweight loss in the quotation). This assumption has two consequences for antitrust policy. The first is that Congressional concerns about democracy (as in media mergers) or macro stability (as in bank mergers) or concerns for small business or income transfers are ultra vires. This was Judge Bork’s original motivation for introducing his version of the consumer welfare standard.
The second consequence in the merger-efficiency context is that the consumer welfare standard allows us a priori to rationalize all cost savings as a social benefit, even if the source of the cost savings may harm social welfare, for example in the case of cost savings from layoffs.
Advocates of the consumer welfare standard must also link consumer and producer surplus to social welfare. This is done through a second assumption, the Kaldor-Hicks compensation principle. This principle states that greater economic surplus results in situations where all individuals who gain could compensate all the individuals who lose and still benefit. However, modern welfare economists abandoned the Kaldor-Hicks efficiency assumption decades ago because it is inconsistent and has undesirable ethical consequences. The current FTC and DOJ leadership are correct that the consumer welfare standard is flawed, yet they let it slip in through the back door in the Guidelines’ efficiency section.
Horizontal Mergers and Cost Savings
Setting aside welfare considerations, and even accepting Bork’s consumer welfare standard, the Guidelines’ efficiency rebuttal still cannot stand up to scrutiny. A major problem is that over the years it has become apparent that there is little evidence that horizontal mergers either often or on average lead to cost savings. Evidence comes from varied sources, ranging from the trade press to careful econometric work, using modern difference-in-differences techniques. Professor John Kwoka’s notable meta-study found that “there is, in short, no good evidence that mergers generally result in substantial and verifiable cost savings, notwithstanding claims to the contrary.” Nancy Rose and Jonathan Sallet thoroughly review the literature and conclude, “Overall, the results from efforts to directly measure merger-induced efficiencies provide little support for the propositions that horizontal mergers are either motivated by or effective in producing significant economic efficiency gains.” Professor Shilling stated: “A considerable body of research concludes that most mergers do not create value for anyone, except perhaps the investment bankers who negotiated the deal.” Carstensen and Lande conclude that “there is a very large and respectable body of findings suggesting that, generally and overall, significant mergers lead to a small or relatively large net negative effect on efficiency.” Our own review of the literature shows that the few papers supporting horizontal efficiencies contain major flaws and are not reliable.
Even when cost savings do exist there are strong a priori economic reasons to believe that the cost savings will not be either merger specific or verifiable in advance of the merger. As stated in the 1968 Guidelines, economies of scale and scope can generally be achieved through internal expansion. In 1983, Fisher and Lande wrote that “it would be extremely difficult for merging firms to prove that they could not attain the anticipated [so-called] efficiencies or quality improvements through internal expansion . . . .” Indeed, firms that have existed in a market for some time are unlikely to continue to have unexploited economies of scale or scope. A recent paper by Professor Kaplow forcefully makes this case.
Other types of cost savings also often cannot be shown to be merger-specific. The 1997 Merger Guidelines specifically state that so-called efficiencies “relating to procurement, management, or capital cost are less likely to be merger-specific. . . .” If a merger would result in cost savings, then some cost savings would often be achievable through contracting or licensing. Only the excess of merger cost savings over contracting/licensing cost savings would be merger-specific. Since the quantification has to be done ex ante, it usually will be difficult to carry out with any degree of certainty.
Verifying predicted post-merger cost savings (as required by the 2023 Guidelines) is also unlikely. Claims of dynamic cost savings are notoriously subject to difficulties of proof. The 1997 Guidelines state that claimed efficiencies “relating to research and development are potentially substantial but are generally less susceptible to verification.” Professor Brodley came to a similar conclusion: “innovation efficiency appears resistant to measurement because it is difficult to assess the efficiency consequences of alternative innovation decisions. Thus, if a particular transaction is foreclosed by antitrust law, the firm presumably would not abandon all research effort, but instead would take a different innovation path.” Likewise, transactional cost savings are difficult to verify. According to Professor Kwoka, vertical so-called efficiencies “pose substantial problems of identification and measurement.”
Finally, even if there are cost savings from a horizontal merger that meet all of these hurdles and are judged to be “cognizable,” they must be passed on to consumers at least in amounts sufficient to offset an expected unilateral price increase. But any cases where efficiencies are found to be cognizable and sufficient to save an otherwise anticompetitive merger are difficult if not impossible to come by. As one district court recently stated, “The Court is not aware of any case, and Defendants have cited none, where the merging parties have successfully rebutted the government’s prima facie case on the strength of the efficiencies.”
Another potential consideration is that crediting a merger’s cost savings would be at least partly redundant because the Guidelines already assume a “standard deduction” for efficiencies in its evaluation of concentration. This is because the threshold level of increased concentration before a challenge is triggered, is premised on the existence of merger efficiencies. Frederick Warren-Boulton, while he was Deputy AAG in the Antitrust Division, made this presumption clear in 1985: “a standard deduction is implicit in a policy that allows mergers that increase concentration to some extent, even without a showing of any efficiency gains.”
On the other hand, there is little doubt that the Guidelines’ invitation to present an efficiency rebuttal will add material costs and time to agency merger reviews and will reduce enforcement. Alleged cost savings will have to be analyzed and evaluated, and the degree to which these cost savings are likely to be passed on to consumers will have to be estimated. So, are these administrative and judicial costs justified? Following the approach by Steve Salop and John Kwoka, the FTC and DOJ could err in two ways: “errors of commission (Type I errors),” in which a merger with qualifying efficiencies is blocked, and “errors of omission (Type II errors), involving the failure to challenge a competitively harmful merger.” In our view, eliminating the rebuttal would result in a small probability of an error of commission because there is no evidence that a significant number of transactions ever result in cost savings that meet the requisite standards, and when they do, the “standard deduction” may mitigate this potential Type I error. Moreover, if the FTC and DOJ expend their resources analyzing efficiencies, they will be less able to challenge other potentially competitively harmful mergers, creating costs of omission, delays, and unpredictability. Thus, we conclude that economic analysis also supports completely dropping the efficiency rebuttal in the Guidelines for horizontal mergers.
Conclusion
The Guidelines are a significant improvement in many ways over their predecessors, but, by including the efficiency rebuttal, they were only partially faithful to their commitment to incorporate textual, judicial, and modern economic support. No efficiency rebuttal in horizontal merger cases is justified by the text or legislative history of Section 7 of the Clayton Act. It runs counter to the words and the spirit of Supreme Court precedent. And while some lower courts have adopted an efficiency rebuttal, they provide little reasoning or support for their opinions, instead simply relying on earlier versions of the Guidelines. Finally, modern economic analysis also does not support the Guidelines’ approach, and the Guidelines themselves do not provide any sound economic reason to have one.
