Anyone who's followed the US v. Microsoft antitrust trial or heard the angry accusations from rival execs at computer conferences knows the basics of the argument against the world's mightiest software company.
It goes something like this: By leveraging its Windows monopoly, Microsoft dominates other software areas, such as browsers, word processors, and spreadsheets. Network effects take over, consumers get the shaft, and Bill Gates' net worth zooms up another 10 billion dollars, give or take a few decimal points.
Read ongoing US v. Microsoft coverage - - - - - -
A new book by two free-market economists, Stan Liebowitz and Stephen Margolis, takes well-argued and careful aim at that idea.
Winners, Losers, and Microsoft (The Independent Institute, $29.95) features what appears to be the first systematic look at the real-world history of the software industry -- and the results are sure to give Department of Justice (DOJ) trial attorneys the heebie-jeebies.
The duo's conclusion? "When Microsoft moved from a low- to a high-market share, its products were always of a higher quality than the market leader's. When its products were not superior, it did not make inroads against the market leader."
This is the book that should, if justice prevails, become as influential in today's antitrust debates as Robert Bork's The Antitrust Paradox did two decades ago.
Here's some free advice to Microsoft's trial attorneys: Introduce all 287 pages as a defense exhibit. If you don't, you're ignoring some of the best evidence that your side has to offer.
Liebowitz, a professor of economics at the University of Texas at Dallas, and Margolis, the head of the economics department at North Carolina State University, devote most of the book to refuting the idea of "network effects."
That theory, which has recently found support among mainstream economists and Justice Department attorneys, predicts that consumers can end up buying the same type of product that everyone else has purchased -- even if it isn't the best -- for compatibility's sake.
The usual example is QWERTY. Supposedly, in 1936, education prof August Dvorak created a keyboard of the same name (with vowels on the home row) that was speedier. But everyone was already using QWERTY, network effects took over, and the unhappy end result is that every one of us wastes countless hours each year typing on a slothful, awkward keyboard.
Many folks in the computer industry seem to agree with this view of history.
Apple's Steve Jobs has mentioned it in speeches. The Wall Street Journal cited it uncritically in a front-page article, saying network effects are "why most people use a keyboard that begins clumsily with the letters QWERTY." So have leading economic textbooks.
Not so fast, say Liebowitz and Margolis, who have emerged as two of economic history's most prominent myth-debunkers. They've taken a hard look at what actually happened, and show in Chapter 2 that "support for the claim that Dvorak is a better keyboard is both scant and suspect."
There was plenty of competition between different keyboards at the time, according to New York Times articles they found, and Dvorak himself was the author of a slipshod Navy study that economists who like "network effects" arguments have cited without reading. In later chapters, the book also debunks the Beta-is-better-than-VHS myth through highlighting articles such as side-by-side comparisons by Consumer Reports.
In its antitrust case, the US government relies on "network effects" theories.
Its top economic expert, Franklin Fisher, claims that the "dominance of Microsoft's Windows 9X operating System ... is protected, among other things ... which are sometimes referred to as network effects."
The Microsoft antitrust suit filed by some state attorneys general makes the identical claim, as have briefs written by Microsoft arch-enemy Gary Reback.
But do network effects really apply here? No, the book argues. Instead, software markets seem to feature "serial monopolies," such as in spreadsheets, where VisiCalc gave way to Lotus 1-2-3 which in turn was succeeded by Microsoft Excel. But that doesn't mean that these pseudo-monopolies hurt consumers -- far from it, since users will switch en masse when a distinctly better product comes along.
Liebowitz and Margolis built a database of product reviews and market share percentages to compare MS products with competitors'. They conclude that MS has been unsuccessful "except when their products are better than the opposition."
They had to resort to an attempt to buy [Intuit]; they have barely dented America Online with the much ballyhooed Microsoft Network; and they only began to erode Netscape's near-monopoly when their own browser came up to snuff."
It's true that the authors are free-market proponents who might reasonably be expected to take a stand against the DOJ's expansive view of antitrust law.
They occasionally get some technical details wrong, such as confusing the magneto-optical drive on the original NeXT computer with a CD-ROM.
But the book remains an invaluable addition to the bookshelf of anyone interested in the Microsoft trial and future high-tech antitrust cases.
To date, much of the bitterness has been over theory. Liebowitz and Margolis inject facts. They may be incomplete, but this book is the best compilation that anyone's offered so far, and now the burden is on Microsoft's adversaries to refute it.
