Issue Analysis 94 – Antitrust and Consumer Welfare

During the past decade, the American economy has undergone a massive restructuring that raises new questions about the effects of antitrust policy on consumer welfare. Information Age firms emerge from nowhere to dominate their newly-created markets, and older companies employ mergers to consolidate in shrinking markets or to morph their way into new markets. Federal antitrust officials find themselves criticized for high-profile lawsuits against Microsoft and American Airlines at the same time that they have approved major telecommunications mergers. State attorneys general have embarked on a new wave of antitrust activism, reflecting parochial political considerations as well as concerns that federal enforcement is inadequate.

Current enforcement contrasts with the approach of the Reagan and Bush administrations. Between 1980 and 1992, antitrust enforcement focused largely on price-fixing, mergers that created complete monopoly, and anticompetitive activities of government-granted monopolies. Other alleged offenses received less scrutiny because officials realized that growing international competition constrains aspiring monopolists. Policy was also influenced by scholarship demonstrating that mergers and other “suspicious” business practices often benefit consumers by lowering costs or improving product quality.1 Firms frequently earn large market shares by creating value for customers. In highly concentrated industries, the largest firms tend to be the most profitable, suggesting that they achieved their dominant positions by being more efficient.2

It would be a mistake to view the new antitrust activism as a wholesale rejection of the developments of the 1980s. We are not returning to the days when the government tried to break up mergers between two firms that served less than 10 percent of the market. A few contemporary antitrust actions turn back the clock, but most attempt to incorporate new ideas about anticompetitive behavior. This primer provides a short guide to help readers understand and evaluate ongoing antitrust activity.

Antitrust Basics

What is market power?

In “perfect” markets, firms face so many actual or potential rivals that they cannot influence the prices at which they sell. According to conventional economic theory, this situation leads to efficient price signals that ensure every resource is employed in the use that consumers value most highly. A firm that possesses “market power” – the ability to influence the price its customers pay – can distort marketplace decisions by distorting prices.

In reality, most firms have some ability to set their own prices, but not all market power poses antitrust problems. Antitrust policy is supposed to concern itself with firms that possess a significant amount of market power, and hence have a big ability to distort market prices.

Is all market power illegal?

Market power is not always illegal or anti-consumer. It is perfectly legal to possess market power, or even a complete monopoly, as long as it was not acquired through illegal, exclusionary practices. The most common, pro-consumer ways of acquiring market power are to introduce a new product or service, or to produce products and services at lower cost than competitors. As Judge Learned Hand noted in his Alcoa decision, “A single producer may be the survivor out of a group of active companies, merely by virtue of his superior skill, foresight, and industry…The successful competitor, having been urged to compete, must not be turned upon when he wins.”3 A dominant company might also be a “natural monopoly,” meaning that it is less expensive for one firm, rather than many, to serve the entire market.

Market power obtained from the government, through devices like exclusive utility franchises, agricultural marketing orders, and occupational licensing, is also legal, though not desirable. Government-granted monopoly reflects lobbying prowess, but not necessarily skill in satisfying customers.4

So What’s Happening Now?

Possession of market power is not itself a crime; a firm violates the law only if it achieves, protects, or expands its market power through various exclusionary business practices.5 A number of these practices – predatory pricing, tie-in sales, refusals to deal, control of essential facilities, mergers, and general nastiness – are especially relevant to current antitrust activity.

Predatory Pricing

Most people think they know predatory pricing when they see it, and they think they see it a lot: a dominant firm lowers its price, drives smaller competitors out of the market, and then raises its price again. But numerous analysts have demonstrated, and federal courts agree, that it is extremely difficult to distinguish predatory pricing from vigorous competition. U.S. Seventh Circuit Court of Appeals Judge Frank Easterbrook noted, “Studies of many industries find little evidence of profitable predatory practices in the United States or abroad. These studies are consistent with the result of litigation; courts routinely find that there has been no predation.”6

The reason is that even an honest competitor faces strong incentives to cut its prices down to marginal cost – the cost of producing an extra unit of product or of serving an additional customer. In fact, textbook economic theory says that all firms price at marginal cost in a “perfect” market. Therefore, any price at or above marginal cost could simply be a sign of competition, even if the price-cutting firm loses money.7 The Supreme Court recognizes this, confirming in 1982 that “Adopting marginal cost as the proper test of predatory pricing is consistent with the pro-competitive thrust of the Sherman Act…Establishing a price floor above marginal cost would encourage underutilization of productive resources and would provide a price ‘umbrella’ under which less efficient firms could hide from the stresses and storms of competition.”8

In predation cases, federal courts require not just proof that a defendant charged a price below cost, but also that the defendant had a reasonable prospect of recouping its losses later. The U.S. Supreme Court noted in one landmark case:

[T]he short-run loss is definite, but the long-run gain depends on successfully neutralizing the competition. Moreover, it is not enough simply to achieve monopoly power, as monopoly pricing may breed quick entry by new competitors eager to share in the excess profits. The success of any predatory scheme depends on maintaining monopoly power for long enough both to recoup the predator’s losses and to harvest some additional gain.9

Given this judicial attitude, it should come as no surprise that predatory pricing cases are hard to win. The Justice Department’s recent suit against American Airlines seeks to change this by establishing a new definition of the price that would be considered predatory. The marginal cost of carrying an extra airline passenger is pretty low – the cost of processing a ticket, a bit of extra fuel, and another meal or bag of peanuts. As a result, airlines can cut prices drastically without engaging in predation. The last time American Airlines was sued for predatory pricing, a jury took only two hours to return a verdict of “not guilty,” and it would have taken 15 minutes but for one juror’s doubts.10 The Justice Department, however, hopes to redefine the relevant cost as the cost of adding additional flights, rather than the cost of an additional passenger. This approach inflates the cost measure by allowing prosecutors to include the cost of an airplane, crew salaries, and many other items that are not really marginal costs.

The Microsoft case involves a different predation claim – that Microsoft seeks to protect its dominant position in computer operating systems by giving away its Internet browser. By attempting to monopolize Internet browsers, Microsoft is supposedly engaged in predatory activity that will prevent people from using browsers in the future to access programs located on the Internet that can run independently of the Windows operating system.11 It’s true that Microsoft is giving away browsers, and a Microsoft monopoly on browsers could in theory stymie Internet-based alternatives to Windows. But Microsoft has not managed to force Netscape from the market, America Online bought Netscape for $4.2 billion, and major computer makers have announced that they plan to offer the Linux operating system as an alternative to Windows.12 If this is predation, it’s not working.

It’s not hard to see why. If the victims of predation have a realistic prospect of producing a substitute for Windows, then they are sitting on top of an enormous profit opportunity. Netscape and other potential competitors thus have ample ability to attract capital that will let them withstand a predatory campaign. Even if individual companies fail, the software engineers designing their products do not lose their accumulated knowledge. These wandering knights should have no trouble attracting venture capital as they seek the Holy Grail of dethroning Microsoft.

Tying and Bundling

A tie-in sale, or “product bundling,” occurs when a firm refuses to sell two products or services separately, insisting instead that the customer must buy both if it wants either. Firms often use tie-ins as a way to ensure compatibility and quality; for example, a copier manufacturer may require that customers use only its brand of paper or its service department to fix problems. A tie-in can also help a company charge based on intensity of the usage. In the days when computers were programmed with punch cards, IBM required its customers to use IBM cards in IBM computers, which let the company see which customers used computers more intensively.

Bundling is no threat to consumers when there are alternative sources of both products. But it is a significant issue in the Microsoft litigation, because the Justice Department and 19 state attorneys general claim that Microsoft tried to acquire market power in the Internet browser market by bundling its browser with its Windows operating system, which is currently the dominant operating system. (Use of market power in one market to acquire market power elsewhere, a practice known as “leveraging,” is legal.) Microsoft argues that the browser is now a feature of Windows, not a separate product, and so it is not bundling two separate products.

The government case rests on three debatable claims:

Microsoft has a monopoly or near-monopoly on operating systems.

Windows and the Internet browser are two products, not one.

Bundling Windows and the browser harms consumers.

Does Microsoft have market power?

It is true that Windows has by far the largest market share among operating systems. Microsoft’s critics argue that this dominance occurs because software is prone to “network effects.” When network effects exist, a firm’s product may dominate the market not because it is really the best, but simply because individual customers expect a lot of other people will adopt it. They want to make sure they buy the system that is compatible with the one others choose, or that has the most software applications written for it. Later on, a better product may have difficulty displacing this dominant one because most people will not want to switch unless they are confident that most others will switch and a supply of software will be available.13 Recent research, however, has shown that Microsoft achieves large market share only when its products win favorable reviews by experts in computer magazines, and customers will quickly switch to competing software when something better hits the market.14 Therefore, any market power Microsoft possesses was earned through superior performance, and Windows could arguably be displaced quickly by a better product.

Are Windows and the browser one product or two?

Antitrust courts give firms wide latitude in designing their products. If the browser really is integrated with Windows, then it is doubtful that courts will find against Microsoft. Two prominent antitrust commentators note,

As a rule of thumb, courts have immunized design or development activity so long as the monopolist can show that its inventive behavior improved its existing product offerings, even though one of the firm’s aims was to exclude rivals. Courts and commentators have expressed concern that flawed judicial intervention might diminish dominant firm incentives to innovate and that no manageable test exists for distinguishing benign from pernicious behavior.15

To overcome this barrier, the government essentially argues that Microsoft’s integration of Windows with the browser is a sham that is not protected by courts’ traditional deference to companies’ product design decisions.

Are consumers harmed?

Even if Windows and Internet browsers are separate products, it is doubtful that bundling harms consumers in this case. Economic research has shown that bundling allows a firm to leverage market power if the products can be used in variable proportions.16 If the products are used in fixed proportions – such as one operating system per computer – then bundling gives Microsoft no additional market power. Microsoft can fully exploit any market power it has simply by charging the maximum possible price for Windows.17

Anyone really concerned with consumer welfare ought to note that software prices actually fall much more rapidly in markets where Microsoft competes. A recently published book by Stan Leibowitz and Steve Margolis shows that from the late 1980s to early 1990s, software prices in product categories where Microsoft competes fell by 65 percent, compared to 15 percent in categories where Microsoft did not offer a product.18

Refusals to Deal

Businesses refuse to deal with each other, often temporarily, for a wide variety of completely legitimate reasons. Contract disputes provide a common example. When several companies disagree about their obligations under a contract, they may put cooperation on hold while they bargain to resolve their differences. The threat of a refusal to deal can also make contracts easier to enforce, if it costs less to stop dealing with another firm than to go to court.

Refusals to deal figured prominently in the Federal Trade Commission’s recent case against Intel. Intel was involved in intellectual property disputes with several other firms that produced products complementary to Intel’s, and it stopped giving them proprietary information about forthcoming products. The rivals complained to the FTC, the FTC investigated, and Intel agreed to supply some of the information rather than face prosecution.

Intel’s rivals claimed that the firm’s dominant position in computer chips meant that they had to have its advance product information in order to compete effectively. This case exemplifies how success can backfire. If Intel had worse products and hence was a much smaller player, its withholding of information would have been viewed as a mean but uncontroversial negotiating tactic.

Control of Essential Facilities

Refusals to deal are also limited by antitrust doctrines, legislation, and regulation placing special obligations on firms that own “essential facilities.” A monopolist owning a facility that is impractical to duplicate must give other firms access to the facility if access is “essential” for the other firms to do business. Common examples include electric distribution lines, local telephone lines, and railroad terminal facilities. Some of these examples involve government-granted franchises, but the legal reasoning generally assumes that the facilities are unique and essential because they are natural monopolies, not simply because competition is banned.

For decades, telecommunications companies have been embroiled in controversy over refusals to deal. When AT&T was broken up in 1984, federal regulators compelled local phone companies to let long-distance companies use the local wires at a regulated rate to reach customers. The 1996 Telecommunications Act expanded this requirement, forcing local phone companies to lease parts of their network to competitors. In 1999, the Supreme Court clarified that the Federal Communications Commission could only mandate “unbundling” of network pieces whose use is indeed essential to other companies.

The essential facilities doctrine, along with similar legislative and regulatory initiatives, reflects a strong fear that the owner of the facility will “leverage” market power into related markets by cross-subsidizing or denying competitors access to the facility. To understand these concerns, one must understand the perverse incentives created by regulation. The telecommunications industry provides all too many recent examples.


In general, a local phone company has little incentive to cross-subsidize competitive services, unless it is subject to rate-of-return regulation. Under rate-of-return regulation, a company can earn higher profits by artificially inflating its investments in regulated markets. One way of inflating investments is to enter additional, competitive markets, but claim some of the assets used in these markets are actually used in the regulated markets. The most direct way of preventing cross-subsidies is to remove rate-of-return regulation, not heightened antitrust scrutiny or detailed regulatory rules.

Access Discrimination

In an unregulated market, the owner of an essential facility has little reason to refuse competitors access unless the refusal reduces costs or increases quality. If the facility really is a monopoly, the firm can make the most money by charging the highest possible price for access and then letting the most efficient companies use the facility. Nevertheless, incentives for access discrimination are likely to remain as long as phone companies face some form of effective price regulation. Regulation limits price increases for local phone service, but if the phone company can exclude competitors in other markets, it can then charge higher prices in those markets.

Fortunately, this problem is disappearing in telecommunications. Competition from cable television companies, satellites, and mobile telephony will soon undercut the rationale for price regulation. When competition regulates prices, the consumer-oriented justification for open access disappears. Unfortunately, other industries are not so far along. In local gas and electric distribution, government-granted monopolies and price regulation will likely continue to create incentives for access discrimination and a rationale for regulation to prevent it.


Federal officials have authority to challenge mergers, but only mergers that “substantially lessen competition.” For this reason, mergers in industries where there are plenty of capable competitors, such as oil and automobiles, pose little threat to competition. The automobile example especially illustrates how international competition has altered antitrust analysis. In the 1960s, the “Big Four” automakers were regarded as a threat to competitive markets. But by the 1980s, the remaining three U.S. automakers needed trade protection and government bailouts to survive the onslaught of lower-priced, higher-quality Japanese imports. Given these new realities, it’s no surprise that the Daimler-Chrysler merger won approval.

Mergers of firms that possess substantial market power are also benign if the merger itself creates no additional market power. This is one of the reasons that the Justice Department has tended to approve mergers of local telephone companies. Even though these companies have historically enjoyed government-granted monopolies, they rarely compete with each other, and so the mergers do not reduce competition.

That’s not to say that the federal government has taken a hands-off approach to mergers. In some cases, antitrust officials require divestiture of assets before they approve the merger. This happened in the MCI-Worldcom case, when MCI was forced by both U.S. and European regulators to sell its Internet business to a British firm. Department of Justice Antitrust Division chief Joel Klein justifies this as a “surgical” approach that removed the potential for market power.19 One Federal Communications Commissioner took a less optimistic view, arguing that the U.S. government simply acquiesced in a European extortion initiative that forced MCI to sell assets to a European firm at fire-sale prices.20

General Nastiness

Antitrust discussions in the media frequently feature obnoxious, warlike quotations from conversations and e-mails that suggest the defendant is out to crush its competition. But contrary to popular perceptions, neither nastiness, aggressive language, nor a desire to eliminate one’s rivals are against the law. The Seventh Circuit Court of Appeals commented in 1986, “Intent does not help separate competition from attempted monopolization and invites juries to penalize hard competition.”21 For this reason, we can safely ignore the warlike hyperbole of business discussions, however damaging it may be for a defendant’s public relations.

So why do plaintiffs and prosecutors highlight such language? Sometimes they hope to sway juries, but not all antitrust cases go before juries. Even in the absence of a jury, antitrust prosecutions take place in the court of public opinion as well as courts of law. Many companies now subject to antitrust prosecution are viewed as heroes of the high-tech revolution. If prosecutors can convince the public that these companies are nasty, they have a better shot at winning public support for activist antitrust.


In 1996, President Clinton declared, “The era of big government is over.”

Early in 1999, former Labor Secretary Robert Reich told USA Today, “The era of big government may be over, but the era of regulation through litigation has just begun.”22

Government has kept that promise in the antitrust field. Even before Reich uttered his prediction, DOJ Antitrust Division chief Joel Klein said in 1998, “We are especially interested in pursuing new doctrinal issues by re-engaging the federal courts…”23 For federal trustbusters, that means re-defining predatory pricing, searching for monopoly created by “network effects,” and scrutinizing companies that achieved their dominant positions through innovation. Many state attorneys general pile on, convinced that the more permissive antitrust policy of the 1980s was a mistake.

A common theme in many current cases is greater concern about the future direction of innovation, rather than an exclusive focus on short-run economic efficiency stemming from “market imperfections.” Consumers should welcome this shift, because research shows that innovation contributes much more to human well-being than short-run economic efficiency.24 Unfortunately, innovation-oriented antitrust policy requires enforcement officials to answer three extremely difficult questions:

What path will innovation in this industry take if we do not intervene?

What path will innovation take if we do intervene?

Which course of action offers more consumer benefits?

The payoffs for consumers could be big if antitrust officials are right – and the costs could be even bigger if they’re wrong.

1This scholarship is often identified with the “Chicago School” of law and economics, but many of the ideas were also pioneered by scholars at institutions like Harvard and Yale, and many of the insights are widely (though not wholly) accepted by antitrust economists and courts. Robert Bork fostered the acceptance of this “new learning” in The Antitrust Paradox (Basic Books, 1978).

2Yale Brozen, Concentration, Mergers, and Public Policy (New York: MacMillan, 1982); Brozen, “Bain’s Concentration and Rates of Return Revisited,” Journal of Law & Economics (Oct. 1971), pp. 351-70; Harold Demsetz, “Industry Structure, Market Rivalry, and Public Policy,” Journal of Law & Economics 16 (1973), pp. 1-9; Sam Peltzman, “The Gains and Losses from Industrial Concentration,” Journal of Law & Economics 20:2 (Oct. 1977), pp. 229-63.

3United States of America v. Aluminum Company of America, 148 F.2d 416 (2nd Circuit 1945). Unfortunately, Judge Hand regarded Alcoa not as a successful competitor, but as a vicious monopolist, largely because the firm kept building new capacity to meet future market demand.

4Government-granted monopolies are often justified on the grounds that they are “natural monopolies,” but a natural monopoly should normally be able to retain its market position without government protection. Many industries alleged to be natural monopolies have turned out to be quite capable of supporting competition, including gas pipelines, cable television, telephone service, and even local electricity distribution. See Robert Poole, Jr. (ed.), Unnatural Monopolies (Lexington Books, 1985); Jerry Ellig and Michael Giberson, “Scale, Scope, and Regulation in the Texas Gas Transmission Market,” Journal of Regulatory Economics (March 1993).

5Ernest Gellhorn and William E. Kovacic, Antitrust Law and Economics in a Nutshell (St. Paul: West Publishing, 1994), p. 93.

6Frank H. Easterbrook, “Predatory Strategies and Counter-Strategies,” U. of Chicago Law Review 48 (1981), pp. 313-14.

7Philip Areeda and Donald Turner, “Predatory Pricing and Related Practices Under Section 2 of the Sherman Act,” Harvard Law Review 888 (1975).

8Northeastern Telephone Company v. American Telephone and Telegraph, 455 US 943 (1982).

9Matsushita Electric Ind. Co. v. Zenith, 475 US 574 (1986).

10Mark T. Clouatre, “Comments: The Legacy of Continental Airlines v. American Airlines: A Re-Evaluation of Predatory Pricing Theory in the Airline Industry,” Journal of Air Law and Commerce Vol. 60 (Feb.-March 1995).

11″Direct Testimony of Franklin M. Fisher,” pp. 35-44 and 53-66, available at DOJ Microsoft case Web site.

12Steve Lohr and John Mouhoff, “AOL Lays Out Plan for Marriage to Netscape,” New York Times (Nov. 25, 1998); Thomas W. Hazlett and George Bittlingmayer, “Befuddled by ‘Internet Time,’” The Weekly Standard (July 5-12, 1999), pp. 23-26.

13In other words, network effects occur when the value of a product or service to one customer rises as more people choose to use it. Popular examples are telephones or computer software. The value of having a telephone rises if more people can be reached by phone, and the value of a software program rises if more people use it, because a single user can share files with a larger number of people. See Michael L Katz and Carl Shapiro, “Systems Competition and Network Effects,” Journal of Economic Perspectives 8:2 (Spring 1994), pp. 93-115; Stanley Besen and Joseph Farrell, “Choosing How to Compete: Strategies and Tactics in Standardization,” Journal of Economic Perspectives 8:2 (Spring 1994), 117-31.

14Stan J. Liebowtiz and Stephen E. Margolis, Winners, Losers, and Microsoft (Oakland: Independent Institute, 1999).

15Gellhorn and Kovacic, Antitrust Law and Economics, p. 145.

16John McGee, “Compound Pricing,” Economic Inquiry 25:2 (April 1987), pp. 315-89.

17Bundling might also harm consumers if it is part of a predatory strategy to eliminate future competition; this has already been discussed on page 3 above.

18Liebowitz and Margolis, Winners, Losers, and Microsoft; see also Stan Liebowitz, “A Defective Product: Consumer Groups’ Study of Microsoft In Need of Recall,” CEI On Point No. 25 (Feb. 9, 1999).

19Brock N. Meeks, “Pulling Back the High-Tech Reins, Antitrust Chief Sees a ‘Modest, Surgical’ Approach, MSNBC/ZDNN (July 11, 1999).

20FCC Memorandum Opinion and Order, CC Docket No. 97-211, separate opinion of Commissioner Harold Furtchgott-Roth.

21AA Poultry Farms, Inc., v. Rose Acre Farms, Inc., 494 US 1019 (1989).

22USA Today (February 11, 1999), p. 15A.

23Brock N. Meeks, “Pulling Back the Reins.”

24Joseph Schumpeter, Capitalism, Socialism, and Democracy (New York: Harper & Row, 1942), pp. 84-85; Oliver Williamson, “Economies as an Antitrust Defense: The Welfare Tradeoffs,” American Economic Review 58 (1968); Richard Nelson, “Assessing Private Enterprise: An Exegesis of Tangled Doctrine,” Bell Journal of Economics 12:1 (Spring 1981), pp. 93-111.