MicrosoftThe current Microsoft antitrust case, still in progress as this review is being written, has been both hailed and condemned as the most important antitrust action of the coming century.
Its potential significance has been compared to that of the Supreme Court’s 1911 Standard Oil decision, which not only applied for the first time the trust-busting power latent in the Sherman Antitrust Act of 1890 to break up John D. Rockefeller’s Standard Oil Company, but of at least equal importance enunciated the rule of reason on which judicial interpretation of the Sherman Act continues to be based. While none of this conference volume’s contributors develops this comparison, readers may come away from the book feeling that it is apt.; As noted in Lenard’s introduction and overview, there was general agreement among the conferees that the Microsoft case did not indicate any need for revision of existing antitrust laws.Order now
Yet the book identifies a paradox. Today’s computer industry has so many elements of natural monopoly–notably network effects and first-mover advantages–that the market will not work well if left to itself, yet it is far too complicated and fast moving to be regulated effectively. Antitrust is the only feasible policy option, but unless applied with skill and discretion, it may do more harm than good. In particular, consumer welfare is more likely to be enhanced by policy initiatives aimed at keeping the industry open to the introduction of major new technologies that might challenge the dominance of incumbents than by policies seeking to promote price competition among existing firms employing established technologies.
; The paper by Michael L. Katz and Carl Shapiro discusses the fundamental economics of software markets, focusing, not surprisingly, on network effects, but also identifying and discussing the effects of the high set-up costs and low–even zero–marginal costs of software, the durability of software systems, and their rapid technological change. This paper’s treatment of network effects is sharply challenged in Timothy J. Muris’ comment, which argues that network effects comprise a model for which very few, if any, valid empirical examples exist.
; Janusz A. Ordover and Robert D. Willig confront the basic question of whether or not there is any role for antitrust in high-tech markets. After reviewing competing arguments, they conclude that antitrust may be crucial in protecting long-run competition in innovation and that this should be its primary objective.
Their particular concern is with bottlenecks such as Microsoft’s monopoly control of the operating system market, which is a crucial component of a broader system of computer use, including applications such as word processing and access to the Internet via browsers. They propose a three-pronged test to ascertain whether monopoly control of such a bottleneck exists and if so whether it is being used to exclude or restrain potential competitors from other markets in the system: i.e., are short-run profits being sacrificed by exclusionary tactics in the hope of long-run recoupment through expansion of the initial monopoly to systemically related markets.
In his comment, Lawrence J. White maintains that the antitrust problems raised by Microsoft’s tactics are neither as new nor complex as Ordover and Willig suggest, but rather mirror one of the earliest issues in antitrust history–single railroad ownership and control of a bottleneck facility such as a monopolized stretch of track within a networked system of rail transportation.; Timothy F. Bresnahan utilizes an intriguing life cycle punctuated equilibrium model to analyze the nature of technological competition in the computer market.
Monopolies such as Microsoft’s monopoly of microcomputer operating systems and Intel’s of microprocessor chips may arise from first mover advantages in introducing a major new innovation, from patenting of such a system, or from network dynamic economies of scale. Whatever its source, the monopoly may persist for over a decade–aeons in the PC business (p. 158) in Bresnahan’s words. The monopolist will constantly be challenged by potential entrants and must keep abreast of minor technological improvements, but has overwhelming advantages in maintaining its position, including strategic entry barriers.
This should not be of great concern to antitrust enforcers, as users’ switching costs are likely to outweigh the advantages of adopting new products incompatible with previous ones. However, occasionally these periods of stability will be shattered by quantum improvements in technology, and in these periods of epochal competition (p. 163) a new dominant firm may arise. Bresnahan is highly skeptical that any sort of antitrust intervention can improve the industry’s performance, in large part due to powerful network forces creating and sustaining monopolistic positions and to the high switching costs that could be imposed on users by a policy artificially easing entry.
; Benjamin Klein elaborates on one of the characteristics of industry structure identified by Katz and Shapiro: the very low marginal costs of software. He notes that the marginal cost of Microsoft’s browser is not only low, but negative, since a browser steers Internet users to specific web sites which can generate fee revenue. Thus, Microsoft’s willingness to provide its Explorer browser free of charge to original equipment manufacturers, or to bundle it with sales of Windows 95 and charge OEMs for permission to remove its icon, was not entirely motivated by a desire to exclude Navigator, as alleged by the Antitrust Division.; Utilizing an extremely interesting set of time series price data for semiconductors, computers, software and services, telecommunications equipment, and telecommunications services, Kenneth Flamm explores the feasibility and implications of the convergence of telephone, television, and computer access via the Internet (the network computer).
On the basis of changes in relative prices of software, computer hardware, and communications hardware, he concludes that the short-run future of the network computer is in doubt, but convergence is quite possible. The threat to Microsoft’s monopoly of the PC operating system is dire. If network computing is the next major technological breakthrough, the operating system will move from the desktop to the network, and Microsoft’s market power therein will become irrelevant.; A nagging question remains.
Given the pace of technological development in the digital marketplace, should one bother to read this book now? The answer is yes- The appropriate role of antitrust in markets where the greatest welfare gains are expected to come from technological rather than price competition, and where product market boundaries are ephemeral, remains at least as pertinent and thought-provoking as it was in early 1998. Already biotechnology promises to be at least as economically significant as information technology. Standard Oil Company v. United States of America addressed the underlying antitrust issues facing the transition to an industrialized economy: United States of America v.
Microsoft Corporation is dealing with an equally fundamental transformation. Those concerned with this transformation will be stimulated and challenged by the informed and perceptive papers and comments found in this book.Bibliography
no bib needed