Herbert Hovenkamp: Competitive Harm and the 2023 Draft Merger Guidelines

Herbert Hovenkamp provides his round-one comments on the draft Merger Guidelines.

To read more from the ProMarket Merger Guidelines Symposium, please see here.

The Department of Justice Antitrust Division and Federal Trade Commission have posted 2023 Merger Guidelines as a draft offered for comments. One feature is their call for stronger enforcement, which is badly needed in some areas. Another is their nostalgia. They rely mostly on case law and many concepts from the 1960s and 1970s. A third is their structuralism. A fourth is the labyrinth they create by stating eight substantive “Guidelines” and five interpretive ones. A merger can be challenged for violating any one of them. That makes clarity and focus critical.

This article addresses mainly Guidelines 1 and 2, making these points:

1. Antitrust thinking during the mid-twentieth century, from which the 2023 draft Guidelines draw much of their wisdom, regarded structural concentration as critical to understanding “performance.”
2. The legislative debates on the Celler-Kefauver Act in 1950 condemned market concentration, typically linked to high prices.
3. The Supreme Court’s Brown Shoe decision distorted that history, emphasizing concentration but largely ignoring competitive performance.
4. Mergers that eliminate competition “between” the merging firms should be actionable, but only when they increase prices.
5. The harmful effects of mergers presents a question of fact.
6. These Guidelines rely heavily on the Supreme Court’s Brown Shoe decision, but overlook that the Supreme Court itself subsequently corrected many of Brown Shoe’s errors.

Guideline 1 is concerned with structural concentration, or “the number and relative size” of firms in a market. It identifies concentration of HHI>1,800 as problematic (The HHI is the sum of the squared market shares of each firm in the market: thus a market of five 20% firms has an HHI of 20² X 5, or 2,000). This Guideline never discusses the relationship between structure and performance, measured by output, price, or innovation. Indeed, it does not identify any harm associated with concentration. By contrast, Guideline 3 discusses “coordination,” which can increase as concentration rises, and identifies concentrations of HHI>1,000 (equivalent to 10 equal firms) as problematic. The concentration harms that Guideline 1 contemplates must come from something other than inter-firm coordination.

That approach is at odds with the structuralism that dominated antitrust thinking in 1950, when the merger law was amended. A heavily empirical research program associated with Harvard University between 1930-1970 linked market structure to competitive performance. Its pioneers included Edward S. Mason, Joe S. Bain, Leonard W. Weiss, Donald F. Turner, and others. They helped fashion an antitrust policy concerned with the relationship of market structure, output, price, and innovation.

Although the Chicago School challenged structuralism, market structure continues to be important and is reflected in every set of Merger Guidelines. But the link between structure and performance was always critical. Under the “Structure-Conduct-Performance” (S-C-P) Paradigm that dominated 1950s-era antitrust law and economics, structure could predict higher prices, lower output, or reduced innovation. In supply markets, including labor, it can lead to suppressed buying prices or wages. Already in the 1930s Edward Mason identified the concern as “the raising of the price of the product” or “the deterioration of its quality.” Joe S. Bain’s work on entry barriers was dedicated to identifying factors preventing new firms from entering a market even as incumbents charged high prices. Economist Donald F. Turner, head of the Antitrust Division that drafted the 1968 Merger Guidelines, referred to the decisions “of a few dominant sellers in an industry to maintain the same high noncompetitive price.” Structure was important but contingent: it was a way of getting at the harms related to high prices.

The drafters of the 1950 Cellar-Kefauver amendments to the merger law were not economists or antitrust lawyers, but they worried about both concentration and high prices. Not a single participant disagreed. In its selective reading of the Congressional debates, the Supreme Court in Brown Shoe wrote as if the concern was concentration for its own sake. It acknowledged that such a policy might lead to “occasional higher costs and prices,” and affirmed the district court’s conclusion condemning the merger because it resulted in “lower prices or in higher quality for the same price.” Guideline 1 does not confront this critical issue.

While structuralism has been hotly debated in antitrust policy, aside from Brown Shoe those debates reflect the sensible view that antitrust is concerned with the harms caused by noncompetitive markets. We’re not just shooting at some abstraction – concentration – without understanding why it is bad. Guideline 1 needs an amendment to ensure that the structural concern is with mergers that reduce market output (measured by quantity, quality, or innovation) and increase price.

Guideline 2 is the only one that defines “competition.” Its focus is mergers that eliminate “substantial competition between firms.” This category includes, but is broader than, the “unilateral effects” analysis in previous Guidelines. Both the economic methodologies and the case law of unilateral effects have always been concerned with the merger’s impact on prices.

As these draft Guidelines state, competition “often involves firms trying to win business by offering lower prices, new or better products and services, more attractive features, higher wages, improved benefits, or better terms….” Further, “[i]f evidence demonstrates substantial competition between the merging parties prior to the merger,” it may substantially lessen competition. Of course, all firms in the same market compete with each other. The draft Guidelines focus this relationship by considering evidence of things like customers’ substitution between the merging parties, or the impact of one firm’s recent entry on another firm. But customers in highly competitive markets also switch among sellers. An Appendix does observe that one of the various metrics is the effect of such a merger on prices. What is missing, however, is a clear statement that a merger will be challenged on this basis only when it predictably leads to higher prices or lower output – that is, when it represents an exercise of market power. Without that, it fails to give much guidance.

The Clayton Act states a cause-effect test (“where the effect may be”). Questions about causation are eminently factual. The causation question here is whether the merger “may … substantially … lessen competition.” Fact assessments such as these are governed by the Supreme Court’s Daubert and related decisions for addressing scientific evidence. Daubert itself involved expert evidence of causation. At this writing Daubert admissibility is discussed in nearly 300 reported federal antitrust decisions. Under those standards, expert conclusions must be evaluated by their methodology, not their conclusions, and it is critical that experts keep their methodologies responsible and up to date. Merger jurisprudence has done exactly that, and judicial control of expert economist testimony in litigated merger cases is the norm, as Daubert compels.

For example, conclusions expressed in all versions of the Guidelines about the concentration levels that trigger enforcement are ones of fact and may need periodic revision. The revisions are a product of empirical testing and re-testing to see if we have identified the right spot. In this case, the 2023 draft was correct to lower the HHI level that triggers close scrutiny to 1,800; merger policy has been underdeterrent against mergers in highly concentrated markets.

The draft Guidelines treat as a presumption of law the Supreme Court’s Philadelphia Bank conclusion that a merger creating a firm with a 30+ percent market is unlawful. Once again, the question whether a merger of that magnitude harms competition is factual. Nothing in the statute speaks to that question, and certainly not to the 30 percent number. Here, the Philadelphia Bank Court relied on the House Report on the 1950 amendments, which stated that mergers should employ the same test that applied to other Clayton Act provisions. Both the Report and the Court relied on the Standard Stations case, which had applied Section 3 of the Clayton Act to condemn contractual exclusive dealing on market shares lower than 30 percent. The Court effectively extracted from the House Report a criterion for legality that had absolutely nothing to do with horizontal mergers. One thing the House Report and the Supreme Court’s treatment reveal is that neither had a good understanding of the practices they were examining. Today, the economic science is much better and Daubert requires its use. This makes responsible expert testimony on such issues all the more important to proper identification of competitive harm.

More questionable is the treatment of general economic effects as if they were matters of law, thus placing them beyond empirical review. The draft Guidelines quote a statement from Brown Shoe, that “Internal expansion is … more likely to provide increased investment…, more jobs and greater output.” Whether internal growth is superior to mergers and produces greater output are empirical questions, contingent on the circumstances. New entry requires the addition of new assets into a market, and the possible ruin or waste of older ones. In concentrated markets it can also lower post-entry prices, making it unprofitable. By contrast, a merger reassigns assets to a different user and typically keeps them in production. Most successful mergers very likely create value, although some also create market power. Comparative effects cannot be evaluated categorically, once for all mergers.

Also dubious are the draft Guidelines’ statements that a “trend toward concentration” calls for harsher scrutiny. Why firms become larger both horizontally and vertically is another empirical question that preoccupies industrial organization economists. One important cause of higher concentration is technology change, as economists fully realized by the 1960s. As markets moved from hand-made to machine-made cans or nails, or from buggies to automobiles, the size of a minimum viable firm grew substantially. Markets accordingly became more concentrated. Another important question is whether increased concentration came from internal growth or merger. The growth of a market’s lower cost firms increases concentration even if the number of firms remains unchanged.

 The “trend toward concentration” provisions in the draft Guidelines are a harmful disconnect between stated merger policy and sound industrial economics. Even more mystifying is their statement that “efficiencies are not cognizable if they will accelerate a trend toward concentration.” Indeed, in such markets merger efficiencies are just as likely to be efforts to compete with firms that have already attained greater scale or scope.

One thing these draft Guidelines neglect is that the Supreme Court very largely cleaned up its own Brown Shoe mess. In its 1964 Continental Can decision, the Court made clear that the issue was not concentration for its own sake, but rather with harmful effects that might result, including “raising prices above the competitive level.” In its Marine Bancorp decision a decade later the Supreme Court chastised the parties for not undertaking “any significant study of the performance, as compared to the structure” of the market at issue. The Court relied on Bain’s study of the conditions under which new entry would discipline pricing in the market, an issue relevant to that potential competition case. The point was clear: one cannot do merger policy without evaluating the impact of the merger on output, price, innovation, or new entry. Structure is but a tool for making that assessment.

In a case whose economic facts resemble Brown Shoe, but which the 2023 draft Guidelines do not cite, the Supreme Court turned away a private merger challenger in Cargill vs. Monfort. The plaintiff was a competitor who claimed that a merger enabled the defendant to charge lower but non-predatory prices. The plaintiff’s claimed injury was that its profit margins would be “severely narrowed.” The Supreme Court held that lower prices that reduce a competitors’ profits are not a harm that the “antitrust laws were designed to prevent.” Granting relief would be “inimical to the purposes of these laws.” That decision alone should have motivated the drafters to consider effects on output and price more seriously, or at least explain that decision away. Cargill is an imposing signal that the Supreme Court is unlikely to condemn a merger because it leads to lower prices or higher output.

Author Disclosure: I have no financial involvements, nor am I consulting or accepting remuneration, from any party that may have an interest in this issue.

Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.