Artemus Ward once remarked that the problem with the world isn’t what we don’t know: “It’s the things we know that just ain’t so.” That observation aptly describes the current debate over standards and competition in the high-tech field. Everyone knows that in video recorders Beta was better than VHS, but that VHS somehow won an illegitimate victory in the marketplace. And who doesn’t have a Macintosh fan for a friend who knows Macs are better than PCs, but lost out because of a market failure.
In Winners, Losers and Microsoft, economists Stan Liebowitz and Stephen Margolis put the lie to these long-standing high-tech legends. Sifting through the historical record, they find that the best product is, after all, usually the winner. This well-researched, fact-filled volume has obvious implications for the ongoing prosecution of Microsoft, as well as other attempts to regulate high-tech network-based firms. Though written before the release of Judge Thomas Penfield Jackson’s decision, Liebowitz and Margolis thoroughly undercut the economic basis of the judge’s conclusion.
The primary target of the book is the economic theory of “path dependence,” or “lock-in,” under which some (including Judge Jackson) argue that inferior products can survive—and thrive—because consumers are locked in to their use. Each individual consumer may want to move to a competing product, but is unable to do so because of fear that other consumers would not also make the change. The classic (though far from high-tech) example of path dependence has long been the typewriter keyboard arrangement now in use (known as “QWERTY,” from the first five letters on the second row). QWERTY was developed for the typewriters of the nineteenth century and according to popular myth would have been replaced long ago were it not for lock-in. During the 1930s, a competing system—the Dvorak Simplified Keyboard—was developed and was, according to the story, far more efficient. But few would learn Dvorak because no typewriters were made with that arrangement, and no typewriters were made because few would learn it. Its failure to become the prevailing keyboard indicates that the market doesn’t work, many economists say.
Nonsense, say Liebowitz and Margolis in their classic article, “The Fable of the Keys,” which appears in this volume. No reputable study exists showing that Dvorak was better. All evidence suggests that QWERTY is as good as or better than its alternatives. Yet the legend dies hard. In the ten years since their research first appeared, no one has rebutted it, yet QWERTY is still routinely, and wrongly, cited as a “market failure.” Going beyond QWERTY, Liebowitz and Margolis examine several similar—and more recent—alleged cases of the lock-in effect, including the Beta-VHS and Mac-IBM contests. In each case, they find that quality factors explain the successes more easily than path dependence or other “economic things that go bump in the night.”
At the core of this volume is an empirical analysis of competition in software applications. Specifically, the authors look at six applications markets: spreadsheets, word processors, personal finance, desktop publishing, browsers, and online services. Is there any evidence, they ask, of lock-in in software? Or are market shares determined by product quality? Do Microsoft products get an unfair leg up?
But how can one empirically measure product quality? Here the authors come up with a creative approach: they use trade magazine reviews of each product to judge how each fared against its rivals over time. The results showed no evidence of lock-in: time and again, products rated more highly push out market incumbents—Excel over Lotus, Word over WordPerfect, AOL over CompuServe.
Is Bill Gates leveraging his Windows market power to give Microsoft applications an unfair edge? No, say the numbers. Microsoft products that dominate are those rated higher. Those that lose the quality battle—such as Microsoft Money—do not dominate. Moreover, the evidence shows that Microsoft’s success isn’t always based on its Windows home-field advantage: many of its applications successes have come first in Mac applications, where Microsoft has no such advantage. Importantly for the Microsoft case, the study also finds that Microsoft Internet Explorer’s gains against Netscape Navigator can also be explained by quality differentials.
Liebowitz and Margolis do find one clear effect of Microsoft’s involvement in a market. Rather than increase prices, as Judge Jackson would have one believe, markets with a Microsoft presence have actually seen prices decline four times as fast as those without a Microsoft entrant.
While comprehensive and fact-filled, Winners, Losers and Microsoft is written in an informal, and at times even humorous style, making it good reading for laymen as well as experts. It makes a convincing case that competition is not nearly so fragile in high-tech network markets as we’ve been told. Instead of many products competing at the same level, there is often a dominant one. But no product is immune from challenge by a better one. In the end, consumers get better things at lower prices.
James L. Gattuso is vice president for policy and management at the Competitive Enterprise Institute.