In the civil antitrust proceeding against Microsoft, a federal district judge has held Microsoft to have a monopoly in a very narrowly defined, purported product market of PC-based operating systems, and to have engaged in "monopolization", as proscribed by the Sherman Act. Microsoft plans to appeal to the United States Court of Appeals for the District of Columbia Circuit, which reversed an earlier judgment against Microsoft by the same district judge. (See Microsoft Press Pass for the decisions by the court, the trial and deposition transcripts and exhibits, and the superb post-trial papers submitted by counsel for Microsoft.)
Beyond the legal issues relevant to the appeal, which will range from market definition to the existence of injury to competition, it is worthwhile to consider the similarities and dissimilarities between Microsoft and real, undisputed monopolies. Monopolies whose products face little, if any, competition include: (1) the U.S. Postal Service; (2) the Baby Bell local telephone companies; and (3) the cable television monopolies, as well as (4) traditional industrial monopolies insulated from competition by high initial investment requirements and high transportation costs for heavy but low priced products (e.g., cement plants).
It is difficult to find common ground between the monopolies and Microsoft. Most significantly, the contemporary monopolies many of us deal with every day are protected from competition by law or by extraordinary barriers to entry, and consumers must accept what is offered by one supplier and pay the supplier's price. For example, it is a federal crime to compete with the U.S. Postal Service in the delivery of first class mail. In all probability, if competitive entry were permitted, mail could be posted for delivery within hours rather than days in large metropolitan areas. Furthermore, the prices of some services, including most first class mail, probably would go down, while the quality of service would go up. But because a powerful union would lose influence and union members with limited skills would become unemployable at their current salaries, it seems highly improbable that Congress ever will free the market for first class mail.
The Baby Bells and cable TV companies derive their monopolies from their ownership of local wiring systems in conjunction with practical and legal obstacles to the construction of competing networks. In the few municipalities which have been "overbuilt", i.e., where more than one supplier offers the same services over parallel wires, prices have been found to be substantially lower than in the majority of jurisdictions where there is no choice. Prices are high, and the quality of service often is mediocre. Recently the cable TV monopolies have come under competitive attack from the DISH network and other satellite TV providers. Excellent television service now is available to customers appropriately situated to install a dish, but there remains but one supplier of the local telephone dial tone.
Simply because a firm controls 100% of the supply of a product does not make it a monopolist. If consumers can substitute other products for the supposedly monopolized product, there is in fact no monopoly. The underlying principle is what economists call "cross-elasticity of demand." For example, the sole marketer of Tylenol is not a monopolist, because Tylenol is merely one of many pain remedies. I will cover the issue of "relevant product market" in more detail in the future, since it is perhaps the most critical error in the district court's decision.
A monopolist has two fundamental goals: (1) to restrict investment and (2) to raise prices well above "marginal cost" in order to achieve the greatest possible return on investment. Neither goal can be accomplished in a competitive environment. If a company facing actual competition or the prospect of competition fails to invest in research and development, new products, new facilities, and capable employees, it soon will be outflanked by other firms that do. Similarly, if a company facing competition prices its products significantly in excess of its marginal costs of production, a competitor will undercut its prices and take away business.
The non-competitive prices and restricted investment of real monopolies can be quite obvious. For example, in the pre-satellite TV era, most cable TV companies failed to upgrade their electronic equipment for 10 years or longer and continued to offer the same 36-channel service while hundreds of new cable channels became available. Meanwhile, prices steadily increased. Local telephone rates have been set by state agencies and postal rates by a federal commission at levels well in excess of the rates that would prevail with free competition.
When accusations of racial discrimination were leveled against mortgage lenders, Peter Brimelow suggested a brilliant but simple objective test to determine whether the charges were meritorious. He pointed out that if a lender had discriminated in favor of whites who were poor credit risks and against blacks who were good credit risks, this could easily be demonstrated by comparing the default rates of the lender's respective white and black customers. If there had been racial discrimination, there would be a higher default rate among whites than among blacks. ("The Hidden Clue", by Peter Brimelow & Leslie Spencer, Forbes Magazine, January 4, 1993, p. 48.)
A similar sort of backwards analysis may be applied to the Microsoft case. If Microsoft is a monopolist, it will have raised prices well above marginal cost and it will have restricted investment to increase profits. However, there is no evidence that Microsoft has achieved either fundamental goal of monopoly. To the contrary, Microsoft has been a leader in the reduction of software prices, and it has continually invested heavily in new and improved products.
There simply is no evidence that Microsoft has restricted investment in any product area in which it has a significant market share. Indeed, Microsoft was able to out-compete the dominant firms selling spreadsheet, word processing and database applications because Lotus, WordPerfect, and Ashton-Tate failed to invest sufficiently to keep their products up-to-date or to cut prices to competitive levels. Instead, they behaved as if they had unassailable monopolies, charging more than $500 per program in 2000 dollars for software that required many hours of study to learn complicated "secret handshakes" unique to each vendor. The former "big three" applications vendors tried to ignore Windows, while computer users were flocking to it for its ease of use, and they only invested heavily in Windows versions after it was already too late to reverse movement to Microsoft applications.
In contrast, in the software categories in which Microsoft has become a leader, it has consistently invested in improving its products, often adding features that formerly required the purchase of supplemental programs. In PC operating systems, Microsoft competes against its own prior releases. Thus if Windows 98 had failed to offer significant advances over Windows 95 and if Windows 98SE had failed to offer significant advances over Windows 98, there would have been no incentive for users to upgrade.
Furthermore, Microsoft has been a leader in the reduction of prices for computer software. As Microsoft's operating systems have grown more complex, the prices actually have declined in real dollars. OEM prices for Windows are said to average about $65, and the 1999 consumer upgrade to Windows 98SE cost less than $100 for a program vastly more powerful but with more stable performance than Windows 95/98 and including fully integrated Internet Explorer 5.
The district judge objected to Microsoft's integration of Internet Explorer, the best available web browser, into Windows at no extra charge, holding it to constitute a "tying agreement" that unreasonably restrained trade. Since the Court of Appeals reversed a similar ruling against Microsoft in 1998, the prospects for another reversal seem very good indeed. Under the district court's curious rationale, GM would be forever barred from including radios and air conditioning as standard equipment in its cars, because at one time they were expensive aftermarket parts supplied mostly by small companies, who lost their markets when the features became standard equipment. Consumers benefited when Microsoft included IE with Windows just as they benefited when the auto companies made radios and air conditioning standard equipment. If for some reason a Windows user prefers Netscape, Opera, or some other browser, it may be downloaded and installed and designated as the user's default web browser.
High prices have held back such technically competent computer products as the Apple Macintosh and IBM's OS/2. I might be writing this on a Macintosh had Apple priced its products as aggressively as Microsoft did in the 1980s, when Apple computers were arguably superior to PCs. Instead, Apple behaved like a monopolist by restricting investment, raising prices, and licensing rights to build Apple clones only on a limited basis, even though it faced rigorous competition from Intel/Microsoft-based products. Such non-competitive behavior in a competitive market drove Apple into a decline from which it only recently has emerged with a new line of innovative, though still overpriced computers.
New entrants or rejuvenated competitors pose a constant threat to Microsoft's market share in the lines in which it competes. Like most consumers, I have no particular loyalty to Microsoft and would abandon its software in an instant if better and/or cheaper alternatives were to become available. For example, when the preview edition of Netscape 6 was announced in early April, I downloaded and installed it, only to find it so far short of a finished product to be unusable. But Netscape 6 will improve, as will IE, Opera and other programs in the continuing war among Internet browsers.
The barriers to entry can be amazingly small in the software business. Witness Linux, the hottest operating system for web servers and a growing presence on desktops. It was written by a Finnish college student who maintains and distributes the code as freeware. To write software, all you need is a PC, and a fully-equipped, 700 Mhz PC more powerful than the fastest supercomputers of a few years back now costs less than $2,000, in large part because of vigorous competition among CPU, memory chip, video board and hard drive manufacturers.
Thus, Microsoft has nothing in common with real monopolists other than large market shares, which can easily be reversed in the cutthroat software business. (E.g., WordPerfect, Lotus, Ashton-Tate.) If Microsoft has the power to set artificially high prices or to restrict its investment, it is not doing so. Instead, it clearly is competing aggressively on both quality and price and continually developing new and improved products.
[John Wall, an attorney in private practice, received a B.S. in Electrical Engineering from M.I.T. and a J.D. from Stanford Law School.]