Issue Brief: Volume 6, Number 8

Antitrust in 2018: The Meaning of Consumer Welfare Now

Author: Herbert Hovenkamp, JD, PhD

The market dominance of American technology firms is giving rise to a call for antitrust intervention that is slowly but surely gaining momentum.


  • Modern antitrust policy is based on the consumer welfare principle (CWP), which holds that markets should yield two things for the benefit of consumers: (1) output that is as high as is consistent with sustainable competition and (2) prices that are correspondingly as low. Under the CWP, prices faced by consumers are paramount.
  • This approach to antitrust policy, however, is currently being called into question by two other schools of thought that deny the primacy of low prices as an antitrust goal.
  • On the right, the CWP faces a competing technical approach to antitrust policy marked by an emphasis not on consumers but rather on assessing the “general welfare” of consumers and producers in the marketplace. The more significant challenge to the CWP, however, is coming from the left, in the form of a new antitrust “movement” that has a particular aversion to large firms such as Amazon and Microsoft, and an interest in economic goals such as protecting small businesses and controlling runaway profits that can be at odds with promoting low prices.
  • This brief summarizes these three approaches to antitrust policy and enumerates why the current CWP remains best positioned to respond to antitrust concerns, while also suggesting how implementation of the CWP could be tweaked to better protect consumer interests.

An article from The Wall Street Journal earlier this year reports some of the statistics: Amazon accounts for 75% of electronic book sales; Google and Facebook captured 63% of online ad spending in 2017; Apple and Microsoft currently supply 95% of desktop operating systems.[1] These modern companies elicit comparisons, rightly or wrongly, to the monopolies of a century ago. “A growing number of critics think these tech giants need to be broken up or regulated,” in the way AT&T was, the Journal author observed. “Their alleged sins run the gamut from disseminating fake news and fostering addiction to laying waste to small towns’ shopping districts.” But the one thing they don’t run afoul of is the very heart of what actually informs modern antitrust policy: the consumer welfare principle (CWP).[2]

The CWP stands for the proposition that antitrust policy should encourage markets to produce two things for the benefit of consumers: (1) output that is as high as is consistent with sustainable competition, and (2) prices that are accordingly as low. The CWP opposes competition-limiting cartels at one extreme and less competitive firms that rely on higher prices at the other. Market structure is relevant to antitrust policy only insofar as monopolies and oligopolies harm consumers by reducing output, stifling innovation, or yielding higher prices.[3] Factually, the CWP can tolerate very large firms, such as Amazon and Microsoft, provided that their gains are passed on to consumers, and historically it has.

Whether the CWP will remain central to the future of antitrust is under question, though. The CWP is now navigating between two hazards, both of which threaten the importance of low prices as an antitrust goal.[4] On the right, it faces a competing technical approach to antitrust policy marked by an emphasis not on consumers but rather on “general welfare.” This standard dominated antitrust policy in the 1980s and early 1990s, and it is best identified with Robert Bork and his important work, The Antitrust Paradox, in which he permits efficiency claims as an antitrust defense, even when the challenged practice leads to higher prices and causes consumer harm.[5] For Bork, “consumer welfare” referred to the sum of the welfare, or surplus, enjoyed by both consumers and producers, or perhaps even by all of society. His general welfare understanding was built on a strong faith that various practices produced cost savings or other efficiencies, whether or not these were provable, as well as considerable doubt that a large menu of practices (e.g., predatory pricing, vertical mergers) caused genuine competitive harm. In the process he gave clout to an antitrust standard that is very difficult to administer and underdeterrent over a wide range of practices. As a result, many businesses favor it.

On the left, the CWP faces a challenge from a new antitrust “movement.” The goals of this movement have been variously defined as combatting industrial concentration, limiting the economic or political power of large firms, correcting the maldistribution of wealth, controlling runaway high profits, increasing wages, and/or protecting small business. None of those goals is new.[6] They have appeared and reappeared in the history of United States antitrust policy. Among the articulated goals of this movement, however, low consumer prices often goes unmentioned. Proponents, some of whom are referred to as “neo-Brandeisians” (after Supreme Court Justice Louis Brandeis), often regard low prices as an undesirable outcome, at least when they come from large firms at the expense of higher cost, smaller rivals. Overall, the movement is not enthusiastic about the use of economics in antitrust and appears to believe that economics should either be subordinated to political priorities or abandoned entirely.[7] Accompanying this belief comes very considerable suspicion about markets generally, quite aside from monopoly.[8] This movement exhibits strong ambivalence about innovation, particularly when the firms who engage in it become large,[9] and it believes that exclusionary strategies such as predatory pricing are a common device by which firms create dominant positions[10] or force targeted firms to merge.[11]

Of these three approaches to antitrust policy—consumer welfare, general welfare, and the new antitrust movement—the current CWP remains best positioned to respond to the problems of increasing market power going forward, although it would benefit enormously from some technical improvements that I will recommend at the end of this Issue Brief.

An Accurate Understanding of Consumer Welfare

“Consumer welfare” as it is properly used today refers to the welfare of consumers-as-consumers, pure and simple. Speaking objectively, consumer welfare is improved by high output and low prices, as well as high quality. The most explicit case for application of the CWP is in merger law under the Horizontal Merger Guidelines, which expressly embrace a consumer welfare principle to the extent that they tie merger policy to the effects on output and consumer prices.[12] But misunderstandings about definition—often the result of confusing the consumer and general welfare standards—have complicated the debate about how to improve antitrust policy and have affected even Supreme Court usage of the term. The Supreme Court has never categorically embraced any particular definition of consumer welfare, even though it has used the term several times.[13]

It should be said that no particular technical welfare test—whether consumer or general—answers antitrust’s hard questions about when a particular practice should be condemned. One must also have a substantive theory about when practices such as aggressive price cutting, tying, exclusive dealing or mergers are anticompetitive and when they are beneficial. The problem, however, is not merely that the general welfare test trades off presumed consumer harm against presumed producer benefits. It is that Bork gave the benefit of the doubt to efficiency claims while being extremely skeptical about claims of competitive harm.[14] In fact, for practically every practice other than naked price fixing, Bork emphasized their efficiencies or harmlessness, while rejecting nearly all theories of competitive harm. The result is that general welfare tests can tolerate a significant amount of market power in the economy.

There is at least a temporal link between Bork’s more general welfare test and the significant rise of monopoly power in the United States economy. The 2010 Merger Guidelines, unlike Bork, take the risk of high market concentration seriously. Once a prima facie case has been made, the Guidelines require strong evidence of efficiencies that could not be obtained except by the merger and that are of sufficient magnitude to render temporary any predicted price increase.[15] These are rarely found.

In contrast to the administrative challenge of evaluating claims of “efficiency,” the most significant benefit of embracing the modern consumer welfare test is that it makes assessing a transaction relatively straightforward. One only needs to know whether output has gone down or price has gone up. That is the only issue to be considered, and the size of the output reduction or price increase does not matter. Further, there is nothing to trade off. Relative to any general welfare test, the administrative cost savings from a consumer welfare test seem to be substantial.

The main challenge facing the CWP, however, is not the one coming from the technical and monopoly-friendly right, but the one from the political and large firm-allergic left. Ironically, both sides reject the importance of low consumer prices and high output.[16] They simply do so for vastly different reasons.

The Changing Landscape: A New Antitrust “Movement” and Its Flaws

Over the last fifty years, antitrust has become much more technical, particularly in areas such as merger enforcement and exclusionary behavior. Today the concern about market power concentration is robust, and many observers focus on a specific indicator: high price-cost margins.[17] As firms face fewer competitors, price-cost margins tend to rise. And recent literature suggests that market power, as measured by price-cost margins, is in fact rising.[18] The increase began to occur in the early 1980s, at about the same time as the Reagan-era antitrust revolution, with its underdeterrent, general welfare bent.[19]

Changes in antitrust policy may be a factor in this rise of price-cost margins, but they are far from the only explanation. For instance, during this same period, the economy became far more digitized and information-based. To the extent the rise in margins results from an increase in fixed costs, it is not an antitrust problem. Also, as a result of decades of anti-union politics and legislation, wages have been suppressed in favor of company profits. Most aspects of this are not an antitrust problem either, although there are strong arguments for paying more attention to labor market concentration, particularly in merger cases.[20] But to whatever extent antitrust policy has directly resulted in increases to price-cost margins, there is no viable reason for eschewing the modern CWP—which is much less tolerant of market power than the general welfare standard employed by the DOJ and FTC in previous decades—in favor of the neo-Brandeis movement’s unclear interpretations of consumer welfare.

Table 1

Top 10 Highly Concentrated Industries in 2012

One advantage of CWP is that its goals are empirically quantifiable. The correct rules for determining what is best for consumers are empirically determined, moving targets, which change not only with further economic theory and empirical study, but also with changes in production and transportation technology, as well as demographics.[21] To date, the strongest claim of the neo-Brandeis movement—its assumption that individuals in our society would really be better off if they lived in a world characterized by smaller firms and higher prices—remains unverified. The neo-Brandeisians still face the formidable task of providing empirical evidence that this is true in a society where everyone is a consumer and consumers vote mainly with their purchasing choices. The goals of this movement (e.g., fairness[22] and small business protectionism) are not measurable and are fundamentally inconsistent, although their contradictions—like the one between the competing priorities of small business protection and consumer welfare—are rarely exposed.

Also missing at this stage is any serious discussion of remedies, except for some very general statements to the effect that perhaps the best fix for Amazon is regulation.[23] On the one hand, the neo-Brandeis movement is highly suspicious of government, and particularly of its power over the economy. It observes, quite correctly, that government is prone to corruption and special interest domination.[24] At the same time, however, members of the movement argue for much more heavy-handed government regulation, and not on behalf of consumers.[25] Meanwhile, as this movement strongly emphasizes the role of politics in economic change, it pays little attention to changes in technology that provide at least as powerful an explanation.[26]

As decades of antitrust litigation have shown, antitrust is not good at balancing. The advantage of the CWP is that economics gives us a set of tools for assessing the conditions that are conducive to high output and lower prices, and thus for examining the practices claimed to challenge them. That is not to say that employing these tools is easy, but over the years we have been able to improve their usefulness.

In contrast, the broader goals identified by movement antitrust, including control of political power and wealth equality, job provision and wages, and protection of small business, are difficult to assess and weigh, and they often operate at cross-purposes with one another. The CWP, however, actually speaks to some of these goals, at least indirectly.[27] For example, although wealth equality and job creation are not separately articulated goals of antitrust under the CWP, competitive markets are very likely conducive to more appealing distributions of wealth than monopolized ones. Unless proponents of the antitrust movement provide metrics for their goals, it will remain difficult to justify a move away from the CWP as the guiding force of antitrust policy.

So what should antitrust do about rising price-cost margins? Should it entertain calls for radical change, or hold fast to the CWP? It is worth noting that two of the principal targets of movement antitrust today – Google and Amazon – are not significant contributors to this rising margins phenomenon. Google’s most common price to consumers is zero, and Amazon’s margins are among the lowest in all retailing.[28] Certainly, big business can cause harm to the lives of Americans in other ways than through competitive pricing. But these ways need to be articulated, supported by evidence, and then sorted into those things that are conceivably within the domain of antitrust and those that are not. Promiscuous application of the antitrust laws so as to make big firms smaller and prices higher could cause irreparable harm, not only to consumers, but also to the entire economy. Objectively, embracing an ideology of supporting lower output and higher prices would be a disaster for the American economy, which is in competition with other world economies that is fiercer than at any point in the postwar period. The danger that the political process will force government policy off the rails is real. The only workable option is to reinforce the CWP.

Recommendations for Improving Antitrust Policy—Without a Movement

The way to repair deficiencies in antitrust law today is not to resort to an undisciplined set of goals that provide no guidance and could do serious harm to the economy. Rather, it is to make ongoing adjustments in our technical rules of antitrust enforcement that reflect what research and experience have taught us. The antitrust laws can reach nearly every form of anticompetitive behavior, provided that they are interpreted flexibly.

Although this Issue Brief is a defense of the CWP, antitrust could protect consumer interests better than it has in recent years. The CWP is not without its flaws. It often appears to be underdeterrent to anticompetitive behavior because of its insistence on due process and rationality, administrability and clear proof.[29] For example, too many merger approvals have been followed by price increases, and current legal standards make predatory pricing almost impossible to prove. By contrast, though, movement antitrust often makes claims that are impossible to deliver, or adopts speculative, unprovable theories about competitive harm. As a result, movement antitrust tends to be overdeterrent.

But in light of the CWP’s record of underdeterrence, I offer the following recommendations. First, legal scholars should work to devise better remedies for mergers discovered to be illegal. Since judges tend to follow scholarship in antitrust more than in other legal areas, this is an important endeavor. At the same time, the CWP would benefit from new scholarship driven by the need to strengthen merger standards—especially for vertical mergers and large tech firm acquisitions of smaller highly innovative rivals—as well as the need to amend proof requirements for exclusionary practices. Additionally, the CWP should be adapted to enable appropriate evaluation of the labor effects of mergers. In general, anticompetitive practices affecting labor markets need to be taken more seriously.[30] While antitrust policy is certainly not the only reason wages fail to keep up with economic growth, its lack of attention in this area is at least a partial contributor. Finally, on a technical level, several practices, such as tacit collusion, predatory pricing law’s recoupment requirement, and the status of indirect purchaser plaintiffs, should be re-examined.

One place that CWP proponents and the neo-Brandeis movement can agree is that concentration does matter, although they currently disagree about how it should be included in the calculus of competitive harm. The antitrust concern with high concentration is a means to an end—namely, control of higher prices—rather than an end in itself. The important point, however, is that established antitrust tools are up to these tasks. Properly applying the consumer welfare principle, not jettisoning it, is the way forward for antitrust law.

About the Author

Herbert Hovenkamp, JD, PhD
James G. Dinan University Professor, University of Pennsylvania

Professor Hovenkamp is a recognized expert and prolific author in the areas of antitrust law and American legal history.  He holds a joint appointment at the University of Pennsylvania between Penn Law and The Wharton School. Prior to joining Penn, he was Professor of Law at the University of Iowa, and before that, taught at the University of California, Hastings College of the Law.

Professor Hovenkamp has been a Fellow of the American Academy of Arts and Sciences. He also has been a Rockefeller Foundation Fellow, Harvard Law School; Fellow of the American Council of Learned Societies, Harvard Law School; Faculty Scholar, University of Iowa; Presidential Lecturer, University of Iowa; and recipient of the University of Iowa Collegiate Teaching Award.

Called “the dean of American antitrust law” by The New York Times in 2011, Hovenkamp received the John Sherman Award from the Antitrust Division of the Department of Justice in 2008, awarded a few times a decade for “outstanding achievement in antitrust law, contributing to the protection of American consumers and to the preservation of economic liberty.”  He is co-author of the landmark 21-volume Antitrust Law, which has been cited more than 50 times by the Supreme Court and more than 1,000 times by federal courts.

[1]Greg Ip, “The Antitrust Case Against Facebook, Google and Amazon,” Wall Street Journal, January 16, 2018.

[2]This Issue Brief is based on the following two papers: Herbert Hovenkamp, “Whatever Did Happen to the Antitrust Movement?” available at and Herbert Hovenkamp, “Is Antitrust’s Consumer Welfare Principle Imperiled?” available at

[3]Herbert Hovenkamp and Carl Shapiro, “Horizontal Mergers, Markets Structure and Burdens of Proof,” 127 YALE L.J. 1996 (2018).

[4]The expansive, vague language of the antitrust laws, particularly the Sherman Act and the Clayton Act, gives rise to the feeling that the antitrust laws can do all things for all people, and over the decades that is precisely how some constituencies have viewed them.

[5]Robert H. Bork, The Antitrust Paradox: A Policy at War with Itself (1978).

[6]See Brandeis’ papers dating back to the 1912 Presidential election, collected in Louis Brandeis, The Curse of Bigness: Miscellaneous Papers of Louis Brandeis (Osmond K. Fraenkel, ed. 1934).

[7]E.g., Zephyr Teachout and Lina Khan, “Market Structure and Political Law: A Taxonomy of Power,” 9 Duke J. Const. L. & Pub. Policy 38 (2014).

[8]See, e.g., K. Sabeel Rahman, Democracy Against Domination, 10-13 (2017).

[9]E.g., Barry C. Lynn, Cornered: The New Monopoly Capitalism and the Economics of Destruction, Ch. 6 (2010).

[10]E.g., Laura Phillips Sawyer, American Fair Trade: Proprietary Capitalism, Corporatism, and the “New Competition,” 1890-1940 at 277-279 (2018).

[11]Lynn, supra note 9.

[12]U.S. Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines ยงยง1, 6, 7, 9, 10 (2010), available at

[13]Five majority opinions speak of consumer welfare. Four additional usages of the term are in dissents. In Reiter v. Sonotone Corp., the Supreme Court held that end use consumers had standing to pursue price fixing, making it the ultimate consumer welfare decision.

[14]Bork made no denying of the fact that he fundamentally disbelieved in the theory of oligopoly. As a result, mergers should be considered harmless unless they created a single-firm monopoly. He also took extremely benign positions on all vertical practices, concluding that the best rule for them should be virtual per se legality except in a small group of cases thought to facilitate collusion. And he believed that predatory pricing is so unlikely to succeed that the best rule for it should be per se legality.

[15]Guidelines, supra note 12.

[16]Protagonists in the neo-Brandeis movement sometimes even write as if low prices are the evil to be avoided.

[17]A price-cost margin increases when the difference between the price a firm charges to consumers and the marginal cost of producing a good or service (i.e., the competitive price) grows. Higher margins are indicative of greater market power.

[18]Jan de Loecker & Jan Eeckhout, “The Rise of Market Power and the Macroeconomic Implications,” NBER Working Paper No. 23687 (Aug. 2017).

[19]Ibid. at 16. (Noting the profit rate relative to GDP grew fourfold during the period 1980-2014.)

[20]The same consumer welfare principles that apply in product markets should be applied to labor markets as well – with a little modification to account for the fact that workers sell, rather than purchase, their labor. The best understanding of consumer welfare is as promoting markets in which output of both product and labor is as high as competition permits. For more, see Ioana Marinescu, “The Other Side of a Merger: Labor Market Power, Wage Suppression, and Finding Recourse in Antitrust Law,” Penn Wharton Public Policy Initiative Vol. 6, No. 3 (2018).

[21]The area that exhibits this most strongly is mergers, with enforcement policy going from severely overdeterrent in the 1960s to underdeterrent today.

[22]“Fairness” was an antitrust goal of the Democratic Party in 2016, according to its party platform. As an antitrust concern, fairness means nothing without a reference point or set of measurement tools. For reference, the full text of major party platforms back to 1840 can be found at

[23]Lina Khan, “Amazon’s Antitrust Paradox,” 126 Yale L. J. 710, at 797-801 (2017).

[24]Lynn, supra note 9, 24, 99-102.

[25]Refocusing antitrust policy so as to make political theory the driver would actually return us to cycles of special interest capture and protected local monopoly. Consider the Robinson-Patman Act of 1936, which prohibited price discrimination, effectively curbing the power of a firm’s size. The historical record of this law shows one of the strongest instances of legislative capture by a special interest group in the entire body of antitrust law. Because the statute applied only to “sales,” it fostered a great deal of vertical ownership integration.

[26]Brick and mortar booksellers, for example, have suffered, but their injury has resulted largely from a technology – direct electronic distribution – that has made them superfluous. It is not antitrust’s purpose to force distribution channels to maintain institutions that no longer perform a valuable function.

[27]Jonathan B. Baker & Steven C. Salop, “Antitrust, Competition Policy, and Inequality,” 104 Geo. L.J. Online 1 (2015). (Discussing, inter alia, the relationship between market power and inequality and offering several proposals for using the antitrust laws to address inequality issues.)

[28]To be sure, either company might be doing anticompetitive things.

[29]E.g., Khan, supra note 23 at 710-711. (Associating the consumer welfare principle with an excessive focus on short run concerns.)

[30]Marinescu, supra note 20.