 |
|
|
DDJ Editors' Note: In
most cases, a single review of an individual book is
enough. However, the recently released Winners, Losers
& Microsoft: Competition and Antitrust in High
Technology, by Stan J. Liebowitz and Stephen E.
Margolis, is sure to be a controversial book because it
addresses a flameable topic (the Department of Justice
antitrust lawsuit against Microsoft) and, much to the
dismay of its authors, has made headlines because of a
botched liason between the Independent Institute (the
book's publisher) and Microsoft. It should be stated at
the outset that there is no evidence whatsoever that
suggests that Microsoft had any hand in the content of
the book. In this respect, it appears that Liebowitz and
Margolis are the victims, so to speak, of greedy and
manipulative actions by the publisher and Microsoft.
These unfortunate actions -- and the negative media
attention surrounding them -- should not reflect on the
book's content or the authors's reputation.
To recap: In 1999, an open letter, signed by more than 200 respected academics and economists, in support of Microsoft in its antitrust battle with the DoJ appeared in full-page ads in The New York Times and The Washington Post. The ads were ostensibly paid for by the Independent Institute (http://www.independentinstitute.com/). According to a more recent article in The New York Times, however, it has come to light that Microsoft, in fact, paid for the ad and several individuals who signed the open letter have subsequently said that they would not have signed the letter had they known of Microsoft's involvement. Both Microsoft and the Independent Institute have denied wrong doing. Interestingly, press releases on the Institute's web site were altered after the initial posting, as the Intitute continued to change its side of the story. Still, there's no denying that Microsoft and the Independent Institute were in cahoots -- a classic case of dumb and dumber. As for Winners, Losers & Microsoft, judge it, its argument, and the authors on the merits of the book. In all likelihood, Liebowitz and Margolis probably wish they'd signed up with a real book publisher. As Gene Callahan notes in his half of this book review, New York University and Macmillan are jointly publishing a book containing the academic version of the materials presented in Winners, Losers & Microsoft. |
In Winners, Losers & Microsoft, Stan Liebowitz and Stephen Margolis ask the question: Do technology markets, left to themselves, usually produce the best achievable outcome, or can market outcomes be improved by government intervention? This question takes on a topical urgency in light of the government's antitrust case against Microsoft. Note the formulation: We shouldn't ask whether we can imagine some ideal software landscape that we would prefer to the current one, but whether we have any reason to suspect that the government will, in fact, be able to create a preferable arrangement.
After a brief study of the contest between the QWERTY and Dvorak keyboards, Liebowitz and Margolis introduce you to some of the current economic theory behind this debate. While a course or two in economics would hardly put you at a disadvantage in these chapters, the authors do a good job of making the material accessible to those who are not trained economists. A grasp of the basic laws of supply and demand suffices to make this section comprehensible. Relevant theory revolves around whether markets can get "locked-in" to an inferior technology simply because it arrives on the scene first, leading to "market failure," or whether markets are able to change paths when the current course is seen to be inferior. The authors contend that while it is theoretically possible to "get stuck," in practice it should be a rare occurrence.
The evidence for actual market failure in high technology has rested almost entirely on just three cases: QWERTY versus Dvorak, VHS versus Beta, and DOS versus Macintosh. However, in the next section of the book, Liebowitz and Margolis convincingly demonstrate that not one of these instances actually illustrates market failure.
The traditional story on our current keyboard layout is that QWERTY arrived on the scene with little competition, won a typing-speed contest or two, then became "locked-in." The "vastly superior" Dvorak keyboard could not make inroads against QWERTY's established position, and we witnessed market failure. However, Liebowitz and Margolis have found that there were many alternate layouts to QWERTY while it was becoming established, there were many early typing contests, and the major study showing significant superiority for the Dvorak keyboard was done by August Dvorak, the inventor. Subsequent, better-run studies showed only minor advantages for the Dvorak layout, too insignificant to repay retraining typists.
In similar tales, Beta is another superior standard that became locked-out of a market, in this case by VHS. The authors' evidence suggests that in recording length, the only technical difference between the competing standards that mattered to home consumers was that VHS generally had 2:1 advantage over Beta. In the high-end professional market, where Beta's editing advantages do matter, Beta has continued to be the dominant format.
The contest between MS-DOS and the Macintosh operating system might seem to be a clear-cut case of the market having failed us. Isn't the Mac clearly superior? Yet, the authors show that things are not that simple -- due to cheaper hardware, less overhead, and more applications, DOS had many advantages that the Mac was not able to overcome.
All of which leads to Microsoft's current dominance in the software industry. Was it simply the result of a lucky break, where Microsoft was positioned to deliver an operating system to IBM just when it needed one? In the mid '80s, the PC industry cried out for software standards -- was there anyone else who would have been better able to deliver them? Liebowitz and Margolis contend that it is unlikely some group of planners could have improved the current software market. The most convincing point they have in making their case is that it was not on Windows that Excel and Word first became dominant products, but on the Macintosh, where Microsoft had no OS monopoly to exploit. Macintosh Excel had achieved an astounding 90 percent market share well before Windows Excel was even the top PC spreadsheet. For those who contend that Microsoft's application dominance is solely a matter of leveraging their OS dominance, this fact is simply inexplicable.
Do I have criticisms? Yes. Several factors make the authors' case somewhat weaker than they might hope. For instance, reviews, which the authors use to evaluate software, are not nearly as objective a measure of quality as, for instance, typing speed. Liebowitz and Margolis ignore a factor which critics of their argument might list as rendering reviews especially suspect: Do large advertising sales from a magazine to a company make the magazine reluctant to give that company's products negative reviews? Yet, we can't fault them for bias in choosing reviews over some more objective criteria, for it doesn't seem that a better measure of software quality is available.
Also, the authors mention Microsoft's shift from touting OS/2 as the operating system of the future to their elevation of Windows to that role, and point out that Lotus and WordPerfect were both hurt by focusing on OS/2 development while Microsoft focused on Windows. But they don't seem to consider the possibility that this could have been a strategic move by Microsoft to leverage their OS dominance. Those wary of Microsoft's motives might be forgiven for suspecting that Microsoft deliberately lured its competitors down a dead-end while focusing its own efforts on what it suspected would be the eventual winner.
Despite these (minor) misgivings, Liebowitz and Margolis make their case in a convincing fashion. After reading this book, you may not like Microsoft any more than before. (Full disclosure: I am a committed UNIX man myself.) But, however bullying, unresponsive to complaints, and paranoid Microsoft has been, here is a strong argument that the largest factor in their success has simply been arriving in the market with the right software for that moment. What's more, Winners, Losers & Microsoft should be a ray of hope for those who have been told that, short of DoJ action, there is no remedy for Microsoft's dominance. Microsoft will slip, and market opportunities that Microsoft no longer has the agility to pursue successfully will arise.
There has been some controversy around this book lately, including charges that Microsoft was involved in its publication. While it is certainly true that Microsoft contributes a significant amount to The Independent Institute, the nonprofit conservative think-tank which published the book, the authors point out that they have been working in this same area, with essentially the same conclusions, for a decade, well before Microsoft had any interest in antitrust theory. When I queried Stan Liebowitz via e-mail on this matter, he pointed out that several other publishing firms had an interest in publishing the book, and that an academic version of the same material will, in fact, be jointly published by New York University Press and Macmillan. My own feeling is that the facts and conclusions in the book stand on their own, whoever paid to have them published.
-- Gene Callahan
Winners, Losers & Microsoft, by Stan J. Liebowitz and Stephen E. Margolis, is a scholarly book about the economics of broad market high technology, a book which ostensibly attempts to answer the question, "Do bad products backed by the weight of a network of extant distribution relationships drive out technically superior products?" The authors supportably conclude that the answer is a fairly uniform "No." But when there is danger that the reader's attention will turn towards the more iron-fisted techniques of antitrust, the book resorts to metaphysical fireworks as a means of misdirection.
Liebowitz's and Margolis's research is intended to shed light on the current antitrust struggle between the Justice Department and Microsoft. It is also intended to burst populist mythology surrounding certain technological forks in the road, both past and present. The myth they wish most earnestly to puncture is that a large technology vendor (for example, Microsoft, wink, wink, nudge, nudge) can achieve permanent hegemony solely on the momentum of a burgeoning installed base and in defiance of technologically superior competition.
The authors commence their investigation with the QWERTY and Dvorak keyboards and continue with an examination of the video format war between Beta and VHS. They painstakingly and credibly demolish urban legends of the weak technology featuring evil markeeters stabbing in the back the hope of the future.
Studies from the first half of the century ascribing superiority to the Dvorak keyboard are reviewed in the light of modern techniques and proven methodologically suspect. The merits of Beta and VHS are charted and reasonably assessed as a tie, with longer run time being the appeal of the eventual victor VHS.
The conflicting fortunes of various well-known software application wars (Word versus WordPerfect versus AmiPro, Excel versus 1-2-3, Quicken versus Money versus Managing Your Money, and so on) are diagrammed and reasonable argument is made that specific desirable features of the Microsoft products triumphed over well-known flaws of their competitors, and that where technological superiority has been in the competitors' corner, the competitors have held their own.
The authors insert their own anecdotal experiences as consumers (they used AmiPro, tried WordPro, but eventually settled on Microsoft Word) into the discussion of desktop office software. They exhibit "power user" zeal in extrapolating from their own experiences to the experience of consumers in general. Elements of their reasoning in these matters may appear suspect to the computer professional. The apparent intent of this passage in the overall work is to elicit emotional recognition from professional but nontechnical readers.
Liebowitz (Ph.D. in Economics, UCLA) and Margolis (Ph.D., UCLA) are publishing under the aegis of the libertarian-leaning Independent Institute of Oakland, California (not to be confused with the libertarian-leaning Independence Institute of Golden, Colorado.) Then again, the confusion may be warranted, each institute serving as a regional focus group for free-market curmudgeonry.
Winners, Losers & Microsoft is often undermined by the authors' Humpty-Dumpty-like tendentiousness, as when, for example, they write:
There is a world of evidence to support our view. There is not a single clear documented example to support the rival claim.
There was a time when academics, whatever their slant or enthusiasm, recognized the difference between a scholarly conclusion and a blanket assertion. The claim referred to above is the one which opens the final chapter, "The Moral":
Our message is simple: Good products win. ... [B]etter products prevail in the marketplace ... People choose what they want ... Surviving products are imitated and become the norm.
Voltaire put it better in Candide, where his crank philosopher Pangloss teaches that "everything is for the best in this, the best of all possible worlds." One wonders if Liebowitz and Margolis are having us on as they write passages like this one:
Economists learn that when marketplace actions seem like errors at first pass, it is probably the observers, not the actors, who are confused.
which reminds us more of the Upanishads and the Bhagavad-Gita than of Keynes or Friedman.
The authors vaguely acknowledge, then steer the discussion clear of, weighter antitrust matters, the only matters, in fact, which trouble honest observers of Microsoft. John D. Rockefeller helped along the workings of the Invisible Hand by blowing up his competitor's oil wells, thereby assuring that Standard Oil (now Exxon) would remain for over a century one of the two or three biggest oil companies in the world. Some of Microsoft's practices have seemed similarly, if largely virtually, heavy-handed.
I haven't heard that Big Bill in Redmond has ordered any rubouts, but he certainly sent Microsoft's early rival, Digital Research, to sleep with the fishes on terms that left DR CEO Gary Kildall bitter for life and convinced DDJ contributor Andrew Shulman (see "Examining the Windows AARD Detection Code," DDJ, Sept. 1993 at http://www.ddj.com/articles/1993/9309/9309d/9309d.htm) that some developer betas of Windows were rigged intentionally to make DR DOS look buggy. None of this is adequately addressed in Liebowitz's and Margolis's work, which states, for instance:
Digital Research had produced CP/M, the leading operating system prior to DOS. According to legend, IBM was planning to use CP/M when it produced its first PC but the behavior of the president of Digital Research was not to IBM's liking. They went looking for another operating system and wound up with Microsoft and DOS.
All true, but is the apparently incestuous relationship of DOS and CP/M, are the later, well-documented techno-shenanigans, entirely extraneous to a discussion of the perception of antitrust which dogs Microsoft and which the authors have set themselves in the title of the book to refute?
Winners, Losers & Microsoft is, in sum, an unabashed religious tractate for capital-L Libertarian Free Marketism in its most undiluted, most Randian form. Accepted as such, and in view of the interesting data gathered by the authors affording a numerical retrospective of two decades of commercial PC software, it's a very entertaining and highly thought-provoking read.
--Jack J. Woehr