Mergers harm competition when they lead to higher prices or exclude rivals anticompetitively. However, they can also help firms achieve lower cost production or to produce better products or services. We describe these things as merger “efficiencies,” and opinion about them is all over the place. On one side, hard-line Chicago Schoolers like Robert H. Bork believed that nearly all mergers were efficient. On the other, Neo-Brandeisians believe that none of them are.
Most opinions are somewhere in between, and the empirical literature indicates that is where they should be. The world of mergers would be far simpler if we simply knew that mergers are either always or never justified by efficiencies. But the actual world is more diverse and stubborn.
Here are some relevant facts: The United States experience 15000-25000 mergers annually. The large fluctuation reflects changes in the level of economic activity, but also in enforcement policy. Of these, roughly 10%-15% are large enough to be reported to the Agencies, and fewer than 1% overall are challenged. For example, in 2024 the most recent fiscal year reported, the FTC challenged 18 transactions and USDOJ 14. The vast majority of mergers are too small to have any measurable impact on competition, but many of them do enable the merging firms to reduce costs or improve their product or service. Mergers are frequently a less expensive route to entering a new market, enabling firms to rely on established business assets rather than creating them from nothing. Some mergers allow owners in financial trouble to sell distressed assets, enabling buyers to acquire them at a low price. When asked about merger motives, firm CFOs identify operating economies and savings for nearly 90% of cases. Of course, CFOs might lie, even on an anonymous survey.
Merger “retrospective” studies look at what happens in the years following a merger. They focus mainly on price changes. Even in concentrated markets mergers result in lower prices nearly as often as in higher ones. These studies are very sensitive to sample selection. The antitrust-focused studies tend to be limited to mergers in markets with small numbers of firms. However, in broader studies that include mergers at all concentration levels, performance improvements tend to dominate.
Individual metrics are another matter: the correlation between price changes and market concentration is extremely rough. As noted in a previous post, the amount by which concentration increases is usually a better metric than the absolute amount of market concentration. At best, however, the decision whether or not to challenge a merger is often little more than an informed guess.
Offsetting this uncertainty is the fact that the standard for challenging mergers is not stringent. They are condemned when their “effect may be substantially to lessen competition.” The courts interpret this language as requiring evidence only of probabilities, not certainties. So the test is whether, more likely than not, a merger will have anticompetitive effects.
Ironically, one important Supreme Court decision that acknowledged merger efficiencies was Brown Shoe, which concluded that as a result of the merger the post-merger firm would be able to offer equally good shoes at lower prices, or better shoes at the same price. The Court concluded that
The retail outlets of integrated companies, by eliminating wholesalers and by increasing the volume of purchases from the manufacturing division of the enterprise, can market their own brands at prices below those of competing independent retailers.1
However, the Court condemned the merger for that very reason. That is, the merger “harm” that it recognized was lower prices harming competitors, not higher prices that would harm consumers.
Often we have a clearer picture of a merger’s efficiency effects than we do of its threats to competition. Determining competitive effects requires an assessment of the merger’s effects on its market. But markets are notoriously difficult to define. Further, the impact of a merger depends on a variety of behavioral assumptions that do not always obtain. By contrast, assessing efficiencies involves looking at the internal workings of the individual firm. For example, if the question is whether a merger of two small medical practices will save money by reducing equipment and staffing costs, the answer will largely be the same whether the practice is located in Ithaca or in New York City, which is 250 times larger. As a result, when efficiency effects are clear they provide a more reliable picture of a merger’s effects.
The Supreme Court has never approved the idea that there should be an efficiency “defense” to an otherwise lawful merger. Instead, the Supreme Court has bundled the efficiencies inquiry into the merger law’s test for competitive harm. The Supreme Court first took that approach in its Winslow decision in 1913, holding that a merger was justified as “simply an effort after greater efficiency.”2 Then in United States Steel (1920), it held that because of a merger’s “resultant economies and benefits,” condemnation risked “injury to the public interest.”3
More recently, in the Brunswick decision (1978) the Supreme Court rejected the claim that a merger was unlawful when Brunswick was buying up struggling bowling allies and then investing substantial cash in them to make them perform better. Assuming that was true, condemning a merger because it led to a better product would be “inimical” to the goals of antitrust.4 In its Cargill decision (1986) the Supreme Court addressed the problem of post-merger price decreases more directly.5 The plaintiff, a meat packer, complained that a very large competitor bought a meat packing facility and that the effect would be that this firm would lower its prices, thus reducing the plaintiff’s profits. Under the plaintiff’s theory, the defendant “would be in a position to do this because of the multiplant efficiencies its acquisition” would produce. The plaintiff’s case closely tracked the Brown Shoe decision, which had condemned a merger for that very reason. However, the Supreme Court concluded that it would be inimical to antitrust goals to condemn a merger because the result was lower prices rather than higher ones.
The government’s Merger Guidelines take a compound approach. First, they do consider efficiencies to be a “defense.” Following the Supreme Court, however, they also integrate the defense into the law’s harm requirement. Under the Guidelines, the government must first show that the merger threatens competition. At that point the defendant can offer an “efficiency defense” only by showing offsetting efficiencies that are sufficient to reverse any harm that the merger might cause. For example, if the government’s evidence shows that a merger is likely to generate a 3% price increase, the defendants would have to show offsetting efficiencies, or cost savings, that would completely reverse this increase, leaving consumers unharmed. As the Merger Guidelines put it, the defendants must show that “no substantial lessening of competition is threatened by the merger in any relevant market.”
Viewing efficiencies this way answers one argument that has been presented against efficiencies, which is that the Clayton Act does not contain an explicit efficiency defense.6 Under the Guidelines’ approach to efficiencies, the absence of an explicit efficiency defense is irrelevant because the requirement of harm to competition has not been satisfied. The argument is frivolous in any event: most statutes do not list the things that can be claimed as defenses. In any event, the merger statute does require proof of probable harm, and lower prices is not the right kind of harm. The 2023 Merger Guidelines do not agree with Brown Shoe’s view that making lower-cost shoes is a merger harm.
Brown Shoe Co. v. United States, 374 U.S. 294, 344 (1962).
United States v. Winslow, 227 U.S. 202, 217 (1913).
United States v. U.S. Steel Corp., 251 U.S. 417, 438, 457 (1920).
Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 488–493 (1978).
Cargill, Inc. v. Monfort of Colorado, Inc., 479 U.S. 104, 114–115 (1986).
See, e.g., FTC v. Penn State Hershey Med. Ctr., 838 F.3d 327 (3d Cir. 2016) (expressing skepticism that such a defense “even exists”).