I thank the ABA for inviting me to take part in this celebration of the Clayton Act’s important role in contemporary U.S. antitrust enforcement. I also appreciate being given the opportunity to go first. That first-mover position means my remarks may appear more original at the start of the conference than at day’s end. I’ll take any advantage I can get.
Antitrust enforcers work with the Clayton Act every day. But it is important to stress at the outset that the statute we enforce—and celebrate—today is not the 1914 version. The Clayton Act underwent significant changes over its 100-year history. Today’s version reflects an evolutionary process involving Congress, the courts, and the antitrust enforcement agencies that has focused merger enforcement on those mergers most likely to produce anticompetitive results.
But to understand how we evolved, let’s go back to the Bronze Age. The 1914 Act’s four original substantive provisions—sections 2, 3, 7, and 8—were in many ways the centerpiece of the Clayton Act. The drafters’ stated goal was to find an alternative to the broad prohibitions of the Sherman Act. At the time, there was widespread popular frustration with the rule of reason announced by the Supreme Court in the 1911 Standard Oil and American Tobacco cases. Many, including the Act’s drafters, perceived that rule as too lenient on anticompetitive behavior by big business. Even some in the business community viewed the Sherman Act standard as too vague to provide useful guidance. The substantive provisions of the draft Clayton Act were therefore intended to “prohibit and make unlawful certain trade practices which, . . . singly and in themselves, are not covered by the [Sherman Act] . . . and thus . . . to arrest the creation of trusts, conspiracies, and monopolies in their incipiency and before consummation.” In short, the creators of the Clayton Act sought to prevent competitive conditions from deteriorating even when competition was not clearly problematic at the time of the lawsuit.
The legislation had two main structural features. First, in sharp contrast to the broad prohibitions of the Sherman Act, the Clayton Act detailed certain specific problematic practices: price discrimination in section 2, exclusive dealing and tie-ins in section 3, holding companies in section 7, and interlocking directorates in section 8. Second, in order to arrest potential restraints “in their incipiency,” the Act banned these practices where their effect “may be to substantially lessen competition.” The intent was to consider likely future effect—not just palpable impact—in determining whether these practices were illegal.
Pairing these two structural features was unplanned; they resulted from the somewhat lost art of legislative compromise. The House version singled out specific practices as especially anticompetitive, but prescribed a different standard for determining liability for each practice. Specifically, the original version of section 7 prohibited only acquisitions with an actual effect on competition, while the original version of section 3 was an across-the-board ban on exclusive dealing and tying arrangements for sales of goods. Section 2, meanwhile, made price discrimination illegal when done with an anticompetitive intent. In addition, all four substantive prohibitions in the draft bill featured criminal penalties. But the Senate greeted the House bill with skepticism. In the ensuing negotiations the criminal penalties were dropped, and the test for illegality for sections 2, 3, and 7 became that the effect of the challenged practice “may be to substantially lessen competition.”
It turns out that the four statutory prohibitions of the initial Clayton Act were not of much lasting value to antitrust enforcers. Section 2, which bans certain forms of price discrimination, morphed into the Robinson-Patman Act of 1936. Today it is not much used outside of private lawsuits. Claims under section 3, which bans anticompetitive exclusive dealing and tying, today typically are alleged alongside claims under section 1 of the Sherman Act. Courts came to apply roughly the same standards for liability under these provisions, but section 1had greater utility because it applies to a broader range of agreements. For example, section 3 is limited to sales of goods but section 1 is not. Section 8 violations are rare.
Section 7, on the other hand, evolved into a cornerstone of current antitrust enforcement. But that is largely due to the 1950 Celler-Kefauver amendment, which expanded section 7’s reach substantially. The original Clayton Act only prohibited stock acquisitions with the probable effect of substantially lessening competition between the acquired and acquiring companies. The prohibition did not apply to acquisitions of assets or even to mergers. In fact, the Supreme Court went as far as holding that if, after a stock acquisition, the acquirer and acquiring firms merged into a new company then the FTC lost jurisdiction over the transaction. The Court reasoned that, because the proper defendant had disappeared because of the merger, there could be no more case.
Nevertheless, the two-part conceptual framework embodied in the original substantive provisions shaped American antitrust jurisprudence. Confronted with the concerns that animated the Clayton Act drafters, courts over time made use of the flexibility of the Sherman Act and came to interpret it much as the compromise solution embodied in the 1914 Clayton Act. Indeed, some Sherman Act decisions drew directly from the Clayton Act, recognizing it as an expression of the country’s “public policy.”
How did the Clayton Act influence Sherman Act enforcement? First, Sherman Act decisions came to incorporate a heightened concern for certain business practices that often produce anticompetitive results. The Supreme Court reaffirmed in cases such as Trenton Potteries that price-fixing by competitors and several other practices are categorically unreasonable, or, “unlawful per se.”
Second, although Standard Oil and American Tobacco articulated in 1911 a rule-of-reason approach for certain conduct, the Court left its definition to later cases. Just a few years after the Clayton Act was passed, Justice Brandeis began to give the rule of reason a more definite shape in Chicago Board of Trade. That decision explained that the “true test of legality is whether the restraint . . . may suppress or even destroy competition.” Chicago Board of Trade is routinely cited today for the proposition that the rule of reason should focus on “a practice’s likely anticompetitive effects.”
The broad texture of the Sherman Act allowed for reinterpretation of its prohibitions as courts came to better understand which practices were the most anticompetitive. The Sherman Act was applied to anticompetitive mergers and asset acquisitions before the Clayton Act was revised to accomplish that. And the Sherman Act was read to prohibit anticompetitive tying, exclusive dealing, and predatory pricing —not just of goods, but also of services. Although the probable-lessening-of-competition standard gave the Clayton Act some flexibility, the specific practices listed in its substantive prohibitions were far less amenable to judicial interpretation.
The courts could not do much with the Clayton Act’s original stock-acquisition provision. It took action by Congress to make section 7 viable. The Celler-Kefauver Act, passed in 1950 to amend section 7, righted several wrongs. Important to language purists in Congress, it unsplit an infinitive in the original act, correcting “may be to substantially lessen competition” to “may be substantially to lessen competition.” Curiously, however, the split infinitive remains in section 3.
More important to antitrust enforcement, the act closed the asset loophole by “amend[ing] section 7 so as to prohibit the acquisition of assets as well as stock of a competing corporation.” With this revision, Congress declared that lessening competition between the combined firms was not quite the issue; the act tied illegality to the lessening of competition in a “line of commerce” and “section of the country.” This change in phrasing also “ma[d]e it clear that the bill applies to all types of mergers and acquisitions, vertical and conglomerate as well as horizontal.”
Despite expanding the list of transactions to which section 7 applies, the revision kept the standard of probable lessening of competition from the original Clayton Act. Like the original Clayton Act drafters, the Celler-Kefauver Act drafters had the goal of “arrest[ing] restraints of trade in their incipiency.” To accomplish that, the Celler-Kefauver Act was “intended to permit intervention in  a cumulative process [of small acquisitions] when the effect of an acquisition may be a significant reduction in the vigor of competition” even though that acquisition does not itself constitute a “restraint of trade.” The legislative history to the 1950 amendment suggests that, more than any specific substantive change in the law, Congress simply wanted to signal to the courts that they should be more aggressive in stopping mergers to prevent further waves of consolidation. That helps explain why the act was popularly called the “Antimerger Act.”
As revised, section 7 of the Clayton Act is central to modern antitrust enforcement. The revision’s immediate success in fixing loopholes in the original section 7 can be seen by comparing enforcement statistics. Prior to 1950, companies could structure anticompetitive acquisitions as asset acquisitions or mergers, and thus escape Clayton Act liability. Consequently, there were relatively few successful section 7 challenges. During the whole period between the original Clayton Act’s passing in 1914 and section 7’s amendment in 1950, the government filed a mere sixteen (or so) cases under Clayton section 7, or less than one every two years. After the amendment, the rate of section 7 challenges dramatically accelerated: In the decade following the amendment, the government brought twenty-seven such cases, or nearly three a year.
The Supreme Court’s subsequent interpretation of section 7, as amended, gave the statutory text more definite shape. This had mixed results, at least at first. The first Supreme Court decision to interpret the 1950 act did not come until 1962, in Brown Shoe Co. v. United States. Relying heavily on the legislative history behind the act, the Court declared that the “dominant theme pervading congressional consideration of the 1950 amendments was a fear of what was considered to be a rising tide of economic concentration in the American economy.” The Court emphasized that the Celler-Kefauver was intended to arrest restraints of trade in their “incipiency.” Although the notion of incipiency was not original to the Celler-Kefauver Act, the Court arguably gave it a new interpretation in Brown Shoe by associating Congress’s concern about trends toward increased concentration with the incipiency concept. Accordingly, the Court held that the Celler-Kefauver Act provides “authority for arresting mergers at a time when the trend to a lessening of competition in a line of commerce was still in its incipiency.” That gave the courts wide latitude to find mergers illegal but provided no analytical framework through which to apply the concept.
Coming soon after Brown Shoe, the decision of United States v. Philadelphia National Bank set up a more workable structure to implement the incipiency inquiry in the context of horizontal mergers. The Court explained that the revised Clayton Act “requires . . . a prediction of [the merger’s] impact upon competitive conditions in the future.” Moreover, it is permitted to “dispens[e], in certain cases, with elaborate proof of market structure, market behavior, or probable anticompetitive effects.” Thus, the Court derived what has come to be known as the Philadelphia National Bank (or “PNB”) presumption: “a merger which produces a firm controlling an undue percentage share of the relevant market, and results in a significant increase in the concentration of firms in that market is so inherently likely to lessen competition substantially that it must be enjoined in the absence of evidence clearly showing that the merger is not likely to have such anticompetitive effects.”
This remains a sensible way of reflecting Congress’s concern with trends toward increased concentration. Where the merged firm would gain an especially large share of the market, and potentially an enhanced ability to control output and price—or market power—courts should be less tolerant of the merger. Moreover, the presumption, when paired with the joint DOJ/FTC Horizontal Merger Guidelines, gives useful guidance to businesses about how section 7 will likely be applied. The presumption—grounded in economics—serves as a useful screen for courts that are not experts at determining the anticompetitive effects of a merger. And the likelihood of any error is reduced by allowing defendants the opportunity to rebut the presumption.
Philadelphia National Bank simplified the required proof for many cases. But later decisions clarified that this simplified proof must still allow consideration of the facts specific to a particular transaction. In that respect, United States v. General Dynamics Corp. stands out. That decision made clear that defendants can prevail by showing that the market shares proffered by the government do not reflect the competitive significance of the merging firms. In that case, as you know, the defendant showed the acquired company’s coal reserves were low and thus the company’s recent sales share did not accurately measure its competitive position going forward.
The logic of General Dynamics applies not just to defending against a structural presumption—a merging party’s recent sales share also can understate its competitive significance. The central insight from General Dynamics is simple and rooted in the statutory text—merger analysis should be forward-looking. As such, it can apply in a broad array of cases.
A second critical improvement to section 7 came when Congress passed the Hart-Scott-Rodino Antitrust Improvements Act of 1976, or HSR Act. The Clayton and Celler-Kefauver prohibitions on anticompetitive acquisitions and mergers were intended to be forward-looking. The drafters envisioned a prohibition that stopped anticompetitive injury before it developed. Until the HSR Act, that was hard to do. The enforcement agencies lacked adequate tools to act before closing. Indeed, one study observed that in nearly 70% of government section 7 actions from 1956 to 1971, the challenged merger could not be stopped before trial.
The HSR Act’s reforms helped realize the forward-looking analysis intended under section 7 with three procedural innovations. First, the enforcement agencies must be notified of any large transaction before it takes place. Second, the merging parties must provide information to allow the agencies to analyze a merger’s competitive effects. Third, the act prohibits consummation of a merger for a period of time to give the reviewing agency a chance to conduct that competitive-effects analysis and, if necessary, file a complaint seeking to block the merger.
My remarks thus far have focused on the role of the courts and Congress building on the 1914 foundation we commemorate today. But the enforcement agencies deserve some credit too—in particular for formulating and periodically updating the Horizontal Merger Guidelines.
The Guidelines are key to contemporary merger enforcement. They help make predictable the exercise of the agencies’ prosecutorial discretion. This is especially important in the merger context where significant business consequences can result from government enforcement decisions. Because private enforcement is relatively infrequent, an agency’s decision whether to challenge a merger takes on unusual significance. The more we can provide guidance on what transactions enforcers see as problematic, the more we help the business community avoid lengthy, costly, and often unsuccessful efforts to combine.
Second, the Guidelines provide the courts with the agencies’ best understanding of how, in light of their expertise and experience, the requirements of section 7 of the Clayton Act should be implemented. After all, even though the Clayton and Celler-Kefauver Acts target specific practices, the probable-lessening-of-competition standard is open-textured and requires continuing judicial development.
For example, the hypothetical monopolist test articulated in the Guidelines gave structure to the market definition standard of “reasonable interchangeability,” which cases such as Brown Shoe had left rather indefinite. This test faithfully implements the reasonable interchangeability standard in a fashion that the courts have embraced.
Similarly, consider the HHI presumptions defined in § 5.3 of the 2010 Guidelines. Accordingly, a merger with an increase of 200 HHI points or more that results in a market with an HHI of 2500 or more is “presumed to be likely to enhance market power.” While in some sense these presumptions are a guide to the agencies’ exercise of their prosecutorial discretion, they also give a sense of how the agencies view the PNB presumption. Indeed, the inclusion of such presumptions in the most recent Guidelines make clear that the agencies believe in the continued vitality of the PNB framework. The Division’s two most-recent litigated section 7 victories—H&R Block and Bazaarvoice—show that courts also continue to accept the presumption as the governing standard in horizontal merger cases.
I’ll end with a word about incipiency. The PNB presumption is admirable in the way it provides a structure for a court to implement the notion of incipiency in the Clayton and Celler-Kefauver Acts. But the PNB presumption doesn’t take trends toward concentration fully into account, which was an obvious concern of the 1950 Congress in passing the Celler-Kefauver Act. The presumption’s focus is on the change in concentration produced by the merger. But I think prosecutorial discretion can play a constructive role in assessing where a market is trending and why that matters to competition. Where prior consolidation has occurred, the incremental competitive effect of the next merger may be worth a closer look.
What do I mean? Look at markets like beer, airlines, and healthcare. In each, past industry consolidation substantially increased later concerns about lessened competition and consumer harm. The advantage of hindsight reinforced the agencies’ worries about further consolidation, even if incremental.
A long history of consolidation preceded the announcement last year of the proposed acquisition by Anheuser-Busch InBev, the largest seller of beer in the U.S., of Grupo Modelo, the third largest. At the time, the two largest firms accounted for 65% of U.S. beer sales, and Anheuser-Busch alone had a 39% share. Many local markets were highly concentrated, and we saw signs of coordinated interaction. These developments amplified our concern about the proposed transaction. Allowing the largest and third-largest U.S. sellers of beer to combine would have made a bad situation worse. But we stopped that from happening.
Our recent challenge to the US Airways-American Airlines merger took into account a wave of mergers in the airline industry in the preceding decade, during which the number of major airlines had fallen from nine to five. We saw evidence that consolidation may have helped bottom lines at the expense of consumers. Given this industry context, the Division was right to seek a meaningful remedy that preserved, and even improved, competition. Since the settlement, which required the merging carriers to divest slots and gates at several highly concentrated airports, low-cost carriers have expanded their services to those airports and introduced a new competitive dynamic for many travelers.
Healthcare provides another example of creeping consolidation, and there the enforcement agencies have been active in sounding the alarm about the dangers of letting that trend proceed unchecked. It took a while for the courts to heed this warning. But they are increasingly doing so, as most recently evidenced by the FTC’s success in ProMedica Health System, where the Sixth Circuit upheld the FTC’s decision to block the acquisition by the dominant local hospital system of another local hospital. Importantly, the court rejected the defendant’s argument that the Philadelphia National Bank presumption was inapplicable in a unilateral effects case. The companies had 46% and 11% of the relevant market, and the FTC provided evidence that larger hospital systems charged higher prices.
Keeping a close watch on where markets are trending does not mean we should return to the days of Von’s Grocery. There the Court intervened long before the trend toward increased concentration should have been a cause for concern. Philadelphia National Bank’s presumption remains the best way for courts to implement the concern for increasing concentration that motivated the Celler-Kefauver Act drafters when they revised section 7.
Our antitrust laws and jurisprudence have come a long way in the last 100 years. But let’s remember also what has remained largely constant. The drafters of the original Clayton Act thought our nation’s antitrust laws needed two improvements—first, a heightened sensitivity to especially anticompetitive practices, and second, the ability to prohibit practices when their effects are reasonably likely to be anticompetitive. To a large extent both the Sherman Act and the revised Clayton Act embody these key tenets of the Clayton Act from a century ago. And that more than justifies today’s commemoration.