Wall Street Journal, February 26, 2001
By Gary S. Becker and Kevin M. Murphy. Mr. Becker, the 1992 Nobel laureate in economics, is professor of economics and sociology at the University of Chicago and a senior fellow at the Hoover Institution. Mr. Murphy is professor of economics at the graduate school of business at the University of Chicago. He is also a consultant to a company helping to defend Microsoft in its antitrust suit.
Rethinking Antitrust
The Clinton Justice Department, under the guidance of Joel Klein, reversed the more laissez faire Chicago School antitrust policy initiated by William Baxter at the beginning of the Reagan administration. Mr. Klein's activist policy culminated in the attempt to bust up Microsoft. The Microsoft case -- which comes before an appeals court today -- is the most visible example of what we believe is a more general and unwise shift in antitrust policy.
With a new administration in Washington, there is an opportunity to rethink antitrust policy once again, and point it toward a few simple and workable rules. The initial goal of the Sherman Act -- combating price-fixing agreements among competitors -- should again become a central focus of antitrust policy.
Troubling Aspects
Three aspects of the shift in policy trouble us most. These are: (1) an emphasis on "unfair competition" as opposed to evidence of harm to consumers; (2) a focus on avoiding "market dominance"; and (3) "market engineering" through the imposition of structural remedies.
Competitors always complain about the aggressive efforts of rivals -- which they call "predatory" -- regardless of whether those efforts make consumers better off. Even in cases where the allegedly unfair practices are explicitly designed to harm rivals, consumers will often benefit from the lower prices and other means by which firms attempt to displace competitors.
An omniscient court would prohibit low prices only when it has proof that the harm to consumers in the future will more than offset the gains which consumers now receive from lower prices. A proof of anticompetitive predatory pricing must establish that a firm is pricing low in order to exclude some competitors in the future, that competitors will in fact be excluded, that this exclusion will harm consumers, and that future losses to consumers outweigh the gains from present lower prices. In the real world, this is an impossible task for courts.
In fact, there is very little empirical evidence to support theories of predatory behavior, and we know of no historical example where economists are in broad agreement that alleged predatory behavior led to consumer harm. Under these conditions, attempts to prevent predatory pricing are likely to discourage a large number of procompetitive actions for each anticompetitive act that they deter.
Given the lack of evidence to support economic theories of predatory behavior, we believe that public policy should operate with the premise that even aggressive competitive behavior generally benefits consumers. The government should be especially skeptical of "unfair competition" arguments by competitors and would-be competitors, particularly when they allege predatory pricing behavior. Lower prices, predatory or not, produce undeniable and measurable benefits to consumers.
An old antitrust doctrine states that bigness per se is no offense. This doctrine recognizes that companies may become big by being more efficient, so that discouraging large market shares may handicap more efficient firms. Despite this wise tradition, rules that subject firms to increasing scrutiny and tougher standards of conduct when they achieve a dominant market position are becoming common in antitrust doctrine. The advocates of this approach justify such rules on the presumption that the likelihood of anticompetitive behavior increases when companies acquire a dominant market position.
By its very nature, however, such a presumption against "dominance" penalizes companies for achieving market success. The Microsoft case provides an apt illustration. Given its strong position in the market for computer operating systems in the mid 1990s, Microsoft could maximize short-run profits by charging relatively high prices for its products and allowing its market position to dwindle over time -- or it could charge relatively low prices and maintain, or even expand, its market share over time. All the evidence suggests that Microsoft followed the latter strategy, and clearly, up to this point, consumers have benefited from the low price strategy.
This tradeoff is very general: Companies usually have a choice between strategies with higher prices and lower market shares, or strategies with lower prices that encourage increases in their market shares. Public policies concerned about consumers rather than competitors would encourage companies to try to increase market shares with lower prices rather than exploiting their market position through higher prices. Unfortunately, current antitrust approaches are often hostile to companies that achieve market dominance, even when that position is achieved through efficiency and low prices.
To fix ideas, consider a rule that subjects firms to antitrust scrutiny when their market shares rise above a critical value, say 60%. When its share approaches this percentage, a company would try to avoid litigation and government harassment by taking profits in the form of higher prices rather than more sales, policies that will reduce effective competition, raise prices, and harm consumers.
Several aspects of modern economies also render traditional notions of market dominance obsolete and harmful to efficiency. Increased global competition means that dominant companies domestically will frequently face substantial competition from abroad if they try to exploit their positions by raising prices or lowering the quality of their products.
Moreover, competition in the fast-changing high tech and software sectors often leads to the displacement of companies with dominant positions by competitors who introduce more advanced technologies. IBM lost its pre-eminent position in computers because of the development of powerful personal computers, and the dominance of Microsoft's operating system is being eroded by the growth of the Internet.
By the time antitrust cases proceed through the litigation and trial phases, a dominant company under attack in one of these industries will usually have lost its pre-eminence, as happened to IBM after more than a decade-long antitrust case. Federal Judge Frank Easterbrook, an expert on antitrust policy, recently pointed out that "any practice in the information industry that survives long enough to be challenged in court, and for the court to reach a decision . . . must be efficient. If it were otherwise, there would have been a swift market response."
A good example is the software market, where consumer demand is greater when a product has more users since a larger number of users creates a network effect, i.e. it facilitates the sharing of files and data with other users and the development of complementary products. Under such network conditions, markets often move toward a standard program or platform, which may lead to a single pre-eminent company -- such as Microsoft's operating system or IBM's earlier dominant position with large computers.
Among the many troubling aspects of Judge Thomas Penfield Jackson's ruling in the Microsoft case, none is more presumptuous, and potentially more disastrous, than the proposed breakup of Microsoft into two separate companies. Structural antitrust remedies attempt to substitute the judgments of economist expert witnesses and the courts for market-determined outcomes. This is particularly problematic when it comes to remedies that attempt to alter the boundaries of a company or the architecture of a market.
Economists in fact know little, and judges know still less, about the economic determinants of company and market structure. They know essentially nothing about how to change these structures in ways that improve rather than worsen outcomes to consumers. Solutions such as breakups, restrictions on the markets which firms are allowed to compete in, or limits on vertical integration, are at best based on untested economic theories about how such remedies might overcome particular anticompetitive concerns. At worst they result from political power and media attention. Under these circumstances, there is enormous potential for unintended and adverse consequences.
For example, when deregulating the California power market, the state government hoped to create a more competitive environment by imposing severe restrictions on the ownership of generating capacity, on the types of contracts the utilities could sign with suppliers, etc. Recent experience, however, indicates that many of these restrictions -- such as the prohibition on long-term contracts for electricity supply -- have made the electricity market there perform much worse.
California regulators may not have been able to foresee the current problems, but they should have been dubious about their ability to engineer improvements in market structure. In "The Theory of Moral Sentiments," Adam Smith recognized, 250 years ago, that "the man of system. . . seems to imagine that he can arrange the different members of a great society with as much ease as the hand arranges the different pieces upon a chess-board . . .; but . . . in the great chess-board of human society, every single piece has a principle of motion of its own, altogether different from that which the legislature might choose to impress upon it."
Destroying Incentives
The proposed remedy in the Microsoft case also illustrates the potential for unintended consequences. Splitting the operating system and applications components of Microsoft into separate independent companies is justified as an attempt to eliminate the applications "barrier to entry." But it may destroy any incentive for Microsoft to set low prices and innovate on both classes of products in order to increase demand for the whole platform.
The remedy also places restrictions on the ability of Microsoft to transfer information and technology between the separated companies. While perhaps some anticompetitive actions will be prevented by such restrictions, these inefficiencies may be only a small component of all the adverse consequences that would eventually follow from such drastic market interventions.
In conclusion, we believe that antitrust policy should be based on one or two simple, well-documented rules that would continue to be effective in the complex modern economy. An obvious example is opposition to price-fixing among competitors. But the market system should not be weakened by attacks on "unfair" competition, by handicapping firms with dominant positions, or by imposing remedies that alter the boundaries of firms and the structure of markets. Such efforts are likely to reduce competition and raise costs, and make consumers and nations worse off.
