Herb Hovenkamp provides his round-two comments on the draft Merger Guidelines.
To read more from the ProMarket Merger Guidelines Symposium, please see here.
An enforcement policy should identify the harms it seeks to control. The 2023 draft Merger Guidelines identify “concentration” as a central concern but are unclear about the harm. Concentration, which relates to the number of firms in a market and their size distribution, can result from mergers, high fixed costs, intra-firm economies, network expansion, differential firm growth, or small markets. On the latter, the grocery market in Ozona, Texas (pop. 2,731), is more concentrated than the national market for automobiles. Single-city hospital mergers can be unlawful.
Concentration can have harmful results. Mainly, it can facilitate anticompetitive coordination among firms or monopoly-induced higher prices—a central concern of the Clayton Act’s “lessen competition” language. Section 3 of the Clayton Act uses the same language for tying arrangements, and the harm is high prices. However, the harm the Brown Shoe case identified—lower prices injuring smaller rivals—had nothing to do with concentration. Lower prices result from competition and the reduced intra-firm costs of making and distributing shoes. Even chain grocers with 5% market shares who compete with each other have lower prices that harm small grocers.
The draft Guidelines describe other concentration-related concerns. One is “roll-ups,” or serial acquisitions threatening oligopoly. That is a smart addition to previous Guidelines. Another is mergers involving two-sided networks, an area that promises great benefits but also harms. This article focuses on two others: mergers that “entrench or extend a dominant position,” and mergers that “further a trend toward concentration.”
“Entrenchment” is concerned with the exclusion of competition. The original Section 2 of the Clayton Act, before it was amended by the Robinson-Patman Act, prohibited a form of predatory pricing, and Section 3 condemned anticompetitive exclusive dealing and tying. These were exclusionary practices. The sections use identical “substantially lessen competition or tend to create a monopoly” language, so it applies to mergers as well. Nevertheless, previous Guidelines have ignored it. Adding exclusion as a concern is wise, but the question is how to go about it?
One comment provided in this symposium describes “entrenchment” concerns as confronting new territory. The courts addressed it repeatedly, however, in the 1960s and 1970s. In Citizens and Southern, Justice William Brennan discussed it in a dissent. In Marine Bancorporation the Court acknowledged the idea but concluded that anticompetitive entrenchment was unlikely. A dissent by Justice Byron White disagreed. A third decision, P&G, condemned a merger on potential competition grounds, observing that Clorox, the acquired firm, was “solidly entrenched” in its market. Finally, the Pabst beer decision condemned a concentration-increasing merger. Justice John Harlan II concurred, believing that the post-merger brewer might become entrenched. Other decisions, like Ford-Autolite, indicated that a merger could increase entry barriers, although it did not speak of entrenchment. Numerous lower court decisions relied on entrenchment.
One feature of the entrenchment decisions is that almost all are mergers of complements, which are goods or services that are either used together or produced together. Except for vertical mergers (also mergers of complements), they are never addressed in any set of Merger Guidelines. Some discussion of complementary products occasionally appears for other reasons. They are woefully undertheorized and the draft Guidelines would have done well to discuss them as a category.
The decisions addressing entrenchment also suffered from the same anti-consumer bias that pervades 1960s and 1970s merger cases: treating cost savings or output expansions as the enemy. That should make them a rationale for approving rather than condemning a merger. Allis–Chalmers, cited twice in the draft Guidelines, condemned a merger between a company that made steel rolling mills and another that made their electrical hookups. A functioning mill needed both, which until then had required two different suppliers. The court condemned the merger because it made the post-merger firm the “only company capable of designing, producing and installing a complete metal rolling mill,” leading to “potential entrenchment” of its market power. The 1963 Ingersoll–Rand decision condemned a merger of two manufacturers of complementary, noncompeting lines of mining equipment. The government’s theory was that the merger injured competition by creating the only firm offering one-stop shopping. Similarly, the 1968 Wilson decision condemned a merger between a maker of gymnastics equipment and a producer of baseball and basketball equipment. The theory was that the firm’s ability to offer a broader line gave it an advantage in bidding for the business of schools, who preferred to deal with a single firm.
“Entrenchment” occurs in these cases because the merger reduces the firm’s production or distribution costs or results in a more attractive product. A merger of two components would not entrench if combining them offered no benefit. The draft Guidelines do acknowledge that the theory will not be used to challenge “simple improvements in efficiency,” citing a decision rejecting a merger challenge on the ground that it would “strengthen the capital position, the resources, [and] the scientific knowhow…” of the post-merger firm.
The draft Guidelines also cite as entrenchment concerns that a merger might increase entry barriers or switching costs, interfere with others’ use of competitive alternatives, deprive rivals of scale economies or network effects, or eliminate a nascent competitive threat. Any one of these is plausible under the right circumstances. The Guidelines would be greatly improved if they provided examples linking these situations to higher prices or other exercises of market power—that is, to actual competitive harm.
For example, “increasing switching costs” reduces the post-merger firm’s elasticity of demand, enabling a price increase. Depriving rivals of scale economies or network effects is simply a way of reducing competitive pressure, enabling a firm to increase its own prices. However, offering products in a way that consumers prefer can also increase switching costs and deprive rivals of scale economies. Not every instance of these practices is harmful, but only those that plausibly lead to higher prices or reduced quality. By relating the concerns to prices or product quality, the Guidelines could also provide a metric for evaluating them. A pervasive problem with these draft Guidelines is that because they are so indifferent to consumer harms resulting from market power, particularly higher prices, they are unable to provide useful metrics for evaluation.
The draft Guidelines would also apply a stricter standard to mergers that further a “trend toward concentration.” That concern was articulated in the Brown Shoe case, although it is not stated in the text of the Clayton Act. The General Dynamics decision pushed back without rejecting it entirely. The concern was stated in the 1968 Merger Guidelines, but then dropped until now. The draft Guidelines mention it for both horizontal and vertical mergers. This provision is particularly interesting because the Guidelines’ basic statement about concentration is largely untethered from a theory of harm. Their principal focus appears to be concentration per se, unrelated to higher prices, reduced output, or loss of innovation.
What harm results from a trend toward concentration? To answer that we need to know why markets exhibit it. A dominant reason is changes in technology, which often involves investment in larger plants with greater fixed costs. The migration of transportation from horse drawn to gasoline vehicles led to many fewer manufacturers, as did the emergence of machine-made cans and large gasoline refineries. Mergers become an important exit strategy for the firms left behind. Occasionally the trend works in reverse. For example, the digital computer moved from large mainframes in the 1960s and 1970s to smaller digital units, permitting many more firms to enter the market. It is no coincidence, however, that the great American merger wave accompanied a period of rapid technological change.
Another thing that increases concentration is uneven firm growth. Some firms have lower costs or are more innovative than others. Suppose a market starts with 10 equal size firms, each with a 10% share, or HHI of 1,000. (The HHI measures concentration as the sum of the squared market shares of each firm.). Suppose that subsequently two firms grow more quickly because they have lower costs or superior technology. Others decline. The array of firm sizes becomes 25, 25, 10,10,5,5,5,5,5,5. This new market still has ten firms, but now the HHI is 1,600, 60% higher (252, 252, 102, 102, 52, 52, 52, 52, 52, 52). In this case the lagging firms lost market share but did not go out of business. If some had shut down, the resulting HHI would be even higher. None of the increase in concentration resulted from a merger, but only from the fact that some firms were more successful than others. A merger of the two 10% firms in the illustration would be challengeable under the draft.
What is accomplished by a more aggressive standard for mergers exhibiting this trend? Certainly not the benefit of small firms, who will lose either a means of surviving or an important avenue of exit. One possibility, of course, is that mergers are making inter-firm coordination more likely, but the draft Guidelines do not mention it. Brown Shoe itself never tied the trend toward concentration to any measure of performance, and certainly not to higher prices. Neither do these draft Guidelines. To the contrary, they suggest that firms might be compelled to match one another’s vertical integration in order to secure supply or distribution. The trend to integrate vertically emerges when integration gives a firm production or transaction advantages over unintegrated rivals. In general, a policy calling for higher scrutiny in markets exhibiting greater amounts of innovation or technological change is difficult to defend.
One defense of the draft Guidelines occasionally offered in this symposium is that they reflect the best of modern economics, displacing older Chicago School views. Indeed, abundant economic evidence indicates that merger law is underenforced. One prominent article cited in this symposium identifies the problem as growing market power that enables firms to “command high prices.” Well and good, except that these Guidelines do not state control of market power or higher prices as an enforcement priority. In fact, de-emphasis of market power and price as competitive concerns is its most notable feature. Another piece doubts the existence of substantial merger generated efficiencies. But these Guidelines’ statement on efficiencies is not much different from the one in the previous Guidelines. In any event, an efficiencies defense is already extremely difficult to prove.
Other economists whom the Agencies regularly engage make a compelling case that we currently underenforce. An important book provides argument and data indicating that we underestimate the adverse price effects of approved mergers. Others argue that mergers often reduce the rate of innovation. Yet others conclude that high concentration and corresponding high prices are a problem. An important article makes a strong case that we should be paying more attention to increases in concentration rather than overall concentration levels, but it relates this to decreases in consumer welfare and higher price levels. Further, it suggests a fairly simple metric that could streamline horizontal merger policy in coordinated action cases.
One symposium author protests that the economic focus on market power and higher prices makes merger analysis difficult. But the list of diverse rationales offered in the 2023 draft Guidelines is hardly an improvement. Some of them mix harmful and beneficial factors with no metric for sorting them. In any event, much of merger analysis requires a market definition, even under these Guidelines. They offer no relief from the complexities encountered there.
Many economists believe that merger enforcement is underdeterrent. I agree. The stated harm, however, is nearly always market power and failure to control higher prices. On outcomes, they are in agreement with economists of the 1950s when Section 7 was amended. While these two groups differed severely on methodologies, they agreed about the goal. The Philadelphia Bank case, a year after Brown Shoe, cited no fewer than seven mid-century antitrust economists. Every one of them tied competition concerns to market power and higher prices and recommended even lower market share levels than the decision adopted. The 2010 Horizontal Merger Guidelines correctly stated that the “unifying theme” of the Guidelines is that “mergers should not be permitted to create, enhance, or entrench market power or to facilitate its exercise.” They were too conservative about the metrics, but correct about the goal.
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.
