The Stigler Center’s 2023 Antitrust and Competition conference seeks to answer the question: what lays beyond the consumer welfare standard? In advance of the discussions, ProMarket is publishing a series of papers with proposed alternatives to the infamous consumer welfare standard. This piece is part of that debate.
Section 7 of the Clayton Act prohibits mergers and acquisitions “the effect of [which] may be substantially to lessen competition, or to tend to create a monopoly.” The test for prohibiting mergers thus refers to competition, not consumer welfare. The two concepts are different. Competition means rivalry among firms—the struggle for customers and other trading partners through price reduction, quality improvement, marketing, and related activities. Consumer welfare refers to the difference between price and willingness to pay of customers. Through a long and complicated process, the consumer welfare standard has made inroads on the competition standard in many areas of antitrust law, though without (in the case of merger law, my focus here) the sanction of the Supreme Court or the clear acknowledgment of any circuit court. And while the policy motivation for this transformation is clear enough, it has been taken for granted rather than adequately defended; it also lacks democratic sanction.
Robert Bork promoted what he called the consumer welfare standard but he meant allocative efficiency (that is, total surplus, which is the sum of consumer and producer surplus), a standard he borrowed from Oliver Williamson’s 1968 paper, Economies as an Antitrust Defense: The Welfare Tradeoffs. That paper introduced in highly qualified form the idea that mergers should be approved if the efficiency gains exceed the deadweight loss—a simple cost-benefit analysis that counts the impact of the merger on everyone (or everyone in privity with the merging parties) equivalently. The law does not take this view. A common but controversial view of the law is that a merger will be blocked if it reduces consumer welfare (surplus); this is equivalent to saying that a merger will be blocked if it increases prices or quality-adjusted prices (or, in the case of input markets, reduces prices, although sellers are not “consumers,” one of the many sources of confusion caused by Bork’s choice of words). One might think of the “consumer welfare test” as a “price test” as in nearly all cases regulators and courts focus on the predicted impact of the merger on consumer prices, and will usually approve the merger if they predict that prices will fall or not rise.
Williamson’s 1968 article did not actually endorse total (or consumer) surplus so much as cautiously advance it as a relevant consideration for the DOJ and FTC in their enforcement decisions. Williamson recognized that the law, as voiced by the Supreme Court, spoke in terms of competition, not surplus or efficiency. He also acknowledged that a test based on the tradeoff between cost savings and price effects omitted many of the harms associated with loss of market competition—the political influence of powerful firms, harms to equity, what he called “social discontent” with corporate power that spurred the various antitrust laws, and harms to innovation. These other harms associated with loss of competition, with the partial exception of harms to innovation, have been mostly forgotten—and with them, the idea that competition is the proper focus of merger enforcement.
Consider the 2010 Merger Guidelines, a statement of enforcement policy. Section 1 says “The unifying theme of these Guidelines is that mergers should not be permitted to create, enhance, or entrench market power or to facilitate its exercise,” but also that “Regardless of how enhanced market power likely would be manifested, the Agencies normally evaluate mergers based on their impact on customers.” Sections 2.1.3 and 5.3 say that a merger that significantly increases concentration and results in a highly concentrated market is “presumed” to substantially lessen competition. But section 10 says that “The Agencies will not challenge a merger if cognizable efficiencies are of a character and magnitude such that the merger is not likely to be anticompetitive in any relevant market”—meaning where “cognizable efficiencies likely would be sufficient to reverse the merger’s potential to harm customers in the relevant market, e.g., by preventing price increases in that market.” Section 6.1 further says that in markets with differentiated products, the agencies will, when possible, use upward pricing pressure and structural models to predict price effects. The talk is of competition but the analysis focuses on prices—or as the Guidelines admit, “efficiencies almost never justify a merger to monopoly or near-monopoly” (section 10) or, in other words, sometimes they do. According to the Guidelines, a statute that prohibits mergers that substantially lessen competition or tend to create a monopoly turns out to allow a merger to monopoly if it does not increase prices. The other possible harms associated with reductions of competition—with the very occasional exception of loss of innovation—are ignored. The Guidelines thus implicitly convert the statutory competition standard into a price standard even though the two standards are different.
In the 1960s, the Supreme Court had confirmed that section 7 prohibits mergers that substantially lessen competition, not just price-increasing or inefficient mergers. It deemed the price effect of the merger a subordinate consideration. But the Court lapsed into silence in the 1970s. Some circuit courts have maintained the Supreme Court’s original view, while others have fallen under the sway of the Merger Guidelines. This can be seen in the debate over the role of efficiency in merger analysis. The market share and Herfindahl–Hirschman index (HHI) tests were originally seen as tests of competition: more concentrated markets were inherently less competitive. Bork reinterpreted the HHI tests as tests of efficiency: mergers that fail HHI thresholds were likely inefficient. Then the notion arose that as the HHI test only created a presumption, defendants should be able to rebut the prima facie case by showing that the apparently inefficient merger was actually efficient—or, more specifically, actually likely to result in lower prices, as in section 10 of the 2010 Guidelines. While defendants have rarely found success with the efficiency rebuttal in the courts, arguments over whether merger-specific efficiencies will offset upward price impacts are now routine in merger litigation.
Similar ambiguities were seen in other areas of merger litigation. Market definition claims are vulnerable to challenge because they artificially segment economic activity into discrete packages by relying on artificial thresholds like the Small but Significant Non-transitory Increase in Price (SSNIP) test. The metrics used to calculate market shares for HHI calculations are similarly vulnerable. Defendants retain experts to cast doubt on the government’s case by highlighting these vulnerabilities, and a common way of doing so is to use methods, including upward pricing pressure and structural models, to show the merger will not raise prices. In this way, the focus of judicial inquiry subtly shifts from whether the merger will reduce competition to whether the merger will increase prices.
These trends have weakened merger enforcement. Price is only one of the harms from loss of competition. The others—like loss of innovation, or risks of collusion, or harms to political integrity—are much harder or even impossible to measure. Thus, the shift in focus from competition to price undermines the law against mergers, enabling rising concentration and falling competition even if courts are able to successfully block mergers with identifiable negative price effects.
Worried that the consumer welfare or price test has undermined merger enforcement, some commentators have called for a return to a competition standard. A merger that substantially reduces market competition should be blocked even if it does not increase prices. But they have struggled to articulate a workable test. Market share tests rely on arbitrary thresholds, are vulnerable to challenge on both theoretical and empirical grounds, have limited usefulness for all but the most homogenous markets, and are losing favor among economists. Qualitative tests like Brown Shoe’s raise questions about administrability. It is hard to believe that courts will disregard economic analyses performed by experts, and as long as prices are regarded as relevant, price predictions will be made, and the precision of those predictions (however questionable in practice) will continue to impress judges. A workable competition test would maintain the greater level of economic rigor that has accompanied the rise of the price test, while avoiding spurious claims that courts can measure real but highly diffuse harms from concentration like the political influence of powerful corporations or public discontent about their role in public life.
Economists have not settled on a single formal definition of competition as they have for consumer or total surplus; thus, one cannot simply convert an economic concept into a legal test. There are, however, models of competition, one group focusing on number of firms (e.g., Cournot) and another group focusing on oligopoly pricing (e.g., Stigler’s theory of oligopoly). Both types of model link market structure or market conditions to market power—the ability of the firm to charge prices above marginal cost, in other words, the margin or markup. All else equal, a merger should increase market power under either theory. The HHI tests were understood as competition tests: a merger that substantially increases HHIs to a high level in well-defined markets substantially lessen competition, meaning they either substantially limit consumer choice or enhance the ability of sellers to collude or set prices in parallel. Reduction in competition is equivalent to an increase in market power; the increase in market power can result in higher prices but it can result in other harmful effects as well, so the price effect of a merger does not by itself identify the full impact of the reduction in competition.
The aim of a competition test thus is to predict changes in market power. Market power cannot be directly observed, and so the question is whether a test less hazy than “competition” and less vulnerable to challenge than HHI can be used as a proxy for market power, and as a focus for economic analysis. Recent work in economics that has sought new ways to measure economy-wide increases in market power suggests one possibility: margin, that is, price minus marginal cost.
Margin and HHI are closely related, but margins can be calculated in a differentiated products model when a market cannot be defined, so a margin test would appear to have broader applicability. A margin test would block mergers (exceeding a size threshold) that substantially increase market power, whether or not they also increase prices, where the market power increase exceeded a predetermined threshold measured in terms of margin level and percentage margin increase (akin to the Lerner index). Margins cannot always be calculated, in which case regulators and courts must fall back on other methods for evaluating mergers. But the current methods for predicting the price effects of mergers, like upward pricing pressure, typically involve calculation of margins, so a margin test should not involve much more complexity than the status quo.
The margin test would be a more rigorous version of the HHI test without the efficiency defense. For example, a merger that merely reduced costs and did not increase prices would be blocked. A margin test, unlike the price test, would normally block a merger to monopoly that reduces prices. But a margin test would (like the price test) permit a merger between small firms that enable them to achieve economies of scale and challenge a dominant firm, causing industrywide margins (and prices) to fall—as Brown Shoe itself contemplated. Note that the margins of the merged firm could increase but the industrywide average margin would normally decline, indeed as HHI calculations could also indicate—fewer firms but more competition as market shares equalize. A margin test might produce some undesirable or inconclusive results in certain circumstances—for example, when fixed costs are high and margins are zero before and after a merger, as may be the case with some tech mergers. But that is true with all tests. Merger evaluation will never be reducible to an algorithm.
The policy case for a margin (or, more broadly, a competition) test rests on a number of premises that have been in disrepute for decades but have recently made a comeback and are now the subject of renewed public debate and academic research. These premises include the notion that market power leads to harmful political power, and that other policy instruments for constraining such power are practically or constitutionally impermissible. An important example is the role of corporations in influencing political outcomes. FCC media ownership rules reflect a traditional worry that media monopolies could interfere with political freedom in localities. Campaign finance regulations reflect a broader concern that any firm, if powerful enough, can distort political outcomes through money and other forms of influence. The relentless advance of technology has produced new forms of corporate political power—exercised most notably today through the moderation rules of giant social media companies—while the Supreme Court has cut back on laws restricting corporate speech. That leaves merger law as one avenue for curbing corporate influence—if it can be shown that market power can be, and frequently is, used to distort political outcomes, and not just by media companies.
The roles of equity and of such matters as “social discontent” with large corporations, to use Williamson’s term, have also received renewed attention in debates about merger policy. It is at least plausible that even mergers that reduce prices will increase corporate profits even more, as these mergers will be undertaken only if managers believe they are sufficiently profitable. Public confidence in big business has been falling for decades, while Americans continue (to the frustration of economists) to celebrate small (admittedly, inefficient) businesses, possibly because they don’t feel abused and pushed around by them. Another question is whether a margin test may lead to perverse results in individual cases because it can be gamed or because margins or related indications of market power are difficult to measure. One might worry that a margin test would result in excessively fragmented markets, but that is an empirical question as well. Today, many economists think markets are excessively concentrated. In any event, the margin test would not atomize the market, as Bork claimed about a competition test. It would not prevent concentration; it would prevent only concentration that would occur through mergers of large firms. Firms would remain free to grow organically.
It is important to see that the implicit normative basis of the market power standard contrasts sharply with the normal way that economists think about competition policy. An economist would normally look approvingly upon a merger to monopoly that is efficient. Most economists would not even require that it reduce consumer prices (as is clear in the theoretical economics literature, where the consumer welfare or price test is nearly always ignored: increased output for the consumer is offset by lost gains to the producer). The market power standard reflects the political economy vision that has repeatedly animated antitrust reform, a vision based on worries about the risks that economic power will be converted to political power, that gains retained by producers cause unfair distributions of wealth, and that corporate power causes other harms. It is not bigness as such that is bad, but concentrated power in private hands that is bad. It is this vision that the conversion of the competition standard to the price test has attempted to eliminate from antitrust law. The antitrust laws were not designed by economists and so do not reflect economic theory; we may ask why economists were allowed to repeal them.
Many commentators and even some judges have claimed that it is only a matter of time before the Supreme Court endorses the price test. But the opposite is more likely the case. The Court is economically conservative and mostly hostile to antitrust enforcement but it is also committed to textualism, and no one has yet explained how a statute that explicitly prohibits mergers to monopolies could authorize a merger to monopoly that reduces prices. The plain meaning of the statute reflects a policy of achieving an institutional aim—markets characterized by multiple competing firms—rather than particular normative goals such as total surplus or consumer welfare or low prices, to be determined on a case by case basis. The institutional goal rests on a judgment that market power causes harms that are greater than price or efficiency effects. This judgment may or may not be shared by IO economists or many antitrust lawyers and academics today, but it is not unreasonable from a broader perspective, and so should enjoy the allegiance of the agencies, the courts, and the antitrust community until Congress, after due consideration of public opinion and the evidence, changes the law.
Eric Posner is Kirkland & Ellis Distinguished Service Professor, University of Chicago Law School. This essay is based on a working paper, where all citations can be found. Eric A. Posner, The Invention of an Efficiency-Based Price Test as the Standard for Evaluating Mergers under Section 7 of the Clayton Act (2023). The author thanks Herb Hovenkamp, Filippo Lancieri, and Steve Salop for helpful comments.
Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.
