Customer Reviews

Winners, Losers & Microsoft

5 star:		(14)
4 star:		(1)
3 star:		(0)
2 star:		(2)
1 star:		(1)

 

 

Forget Microsoft. This book will make you doubt everything you've ever been taught about lock-in, bundling, and the ease with which market dominance can be abused.

But don't forget about Microsoft entirely--the book makes clear that the conventional wisdom about how the company achieved dominance could lead to "remedies" that would have a terrible effect on the software market as a whole--inevitably dooming winners to constant hamstringing or worse in order to enforce an artificial measure of competition.

The movie puts the lie to much that has been said about Microsoft by its critics, competitors, and even its advocates. Certainly no one connected to Microsoft has made such a devastating rebuttal of the charges against it. This book is a must-read for those interested in the outcome of that case.

Liebowitz and Margolis have done research on intellectual property issues for over two decades. Their work on new technologies began over a decade and a half ago. Their piece on the Fable of the Keys, was the lead article in the Journal of Law and Economics in 1990. Unfortunately, their other work on path dependency has languished in less well-known journals.

It is a wonderful asset finally to have all their work on technological lock-ins put together in this one volume, and accessibly written for anyone who is interested in the topic.

I've been a long-time Microsoft user/hater, but I find most of the arguments put forward by the authors to be very compelling: When MS products receive top ratings, those products tend to dominate the market and prices fall. I'm not sure I agree with most software reviewers that MS IE is better than Netscape, but perhaps that's a personal preference.

But as I said, this book is a superb and accessible summary of the authors' work on path dependency and is a telling condemnation of BOTH sides in the antitrust suit agains Microsoft.

There are a lot of myths about why products succeed or fail. Many claim that customers get "locked into" products simply because they arrive on the scene first even when much better products are available. Some claim that Microsoft has taken over different markets despite inferior software simply because of the monopoly that it has in operating systems. Liebowitz and Margolis provide straightforward, convincing, and imaginative evidence that these claims are false.

It is amazing how many stories like the superiority of the DVORAK keyboard hang around for years with no supporting evidence. They make for great stories, but as these authors point out they are false.

If you want to learn about how markets work, read this book.

Finally, the previous commentor's remarks about these authors being bought off is offensive and false. Liebowitz and Margolis wrote about these issues a decade before Microsoft became involved in its current legal problems. Anyone who reads this book will realize that Microsoft would have been a lot better off if they had hired them rather than the lame effort they got from the MIT business school dean.

This book is unfortunately titled as it is really primarily about bringing real data and rigor to bear on many of the conventional "stories" about the economics of the new economy, rather than dwelling on the Microsoft antitrust situation. Clearly, the new economy tends to be characterized by more network effects, increasing returns to scale, and general "winner-take-all" effects than the historical economy. However, certain stories about early lock-in effects of technologies that are inferior, but that by luck got the early lead, have been uncritically passed from author to author. Examples are the QWERTY keyboard, VHS vs Betamax, Windows vs Apple, etc. Liebowitz and Margolis show that most of these stories do not hold up under close examination -- that in fact, these are not examples of the market failing to take the "right path". Actually, the market generally seems to get it right. This book is also the best we have seen in its treatment of the overall economics of information technology standards.

A great book. Must reading for anyone interested in the economics and politics of high technology markets. Explodes a number of myths. Eg. If Microsoft software dominates the PC market because they control the Windows operating system why are Excel and Word the number one products in the Macintosh market? Liebowitz and Margolis dug up hundreds of software reviews to provide evidence for their views. By the way, as was pointed out in the review of the book in the WSJ, Liebowitz and Margolis are academic economists who have been studying and writing about this issue for almost 10 years - long before the Microsoft anti-trust trial started. The book is about a lot more than Microsoft in anycase.

Leibowitz and Margolis tackle one of the central problems in modern antitrust economics: Can inferior products prevail in competition? If so, can the producer of such products achieve a monopoly? Standard neoclassical economics denies that either outcome is possible. Economic evolution tends to favor efficient equilibria, while inefficient local equilibria tend to become extinct. Two fashionable economic theories have been advanced, however, which claim that inferior products can not only prevail in competition but also achieve monopolistic status.

Path dependence claims that inefficient local equilibria can persist over time. Initial conditions, which may be determined by chance or other non-economic forces (such as political interests), direct the system down a particular path. Subsequent deviations from that path may be precluded as too costly, even if there are more desirable or efficient alternatives available. Where the cost of reversing the initially chosen starting point or conditions is especially high, an inefficient equilibrium may result that resists correction by market forces.

Network externalities is the other key element of the inferior product lock-in story. Some products become more valuable as the number of persons using them increases. Each person who adopts the product thus confers positive externalities on other users. Personal computers (PCs) are a commonly cited example of this phenomenon. The value of Microsoft's Windows operating system, for example, depends in part on how many people use it and thus create markets for compatible hardware and software. The more people who use Windows, the larger the market for ancillary products becomes, thus encouraging the development of more ancillary products of higher quality and lower price.

The claim made on behalf of the combined effects of path dependence and network externalities is that one cannot depend on markets to get it right. Arbitrary initial starting conditions favor an inferior product. That product generates network externalities. As the number of users of the product rises, the size of those externalities increases, which gives the inferior product a huge competitive advantage. Eventually the network product achieves a "lock in" effect. At that point, path dependence kicks in. It becomes too expensive to switch to a new product even though that product is superior. Hence, the inferior product prevails.

In WINNERS, Stan Leibowitz and Stephen Margolis popularize their well-known (among economists) work debunking these fashionable theories. An important theoretical contribution, albeit one which is likely to be of interest only to economists and economically-minded lawyers, is the distinction they draw between "second-degree" and "third-degree" path dependence, which is driven by the availability of a feasible alternative to the suboptimal equilibrium. Only if such an alternative is available, does the phenomenon in question demonstrate true path dependence. Only if such an alternative nevertheless remains not taken, can the phenomenon in question even potentially be deemed inefficient in either the Pareto superior or Kaldor-Hicks sense.

Of greater interest to general readers will be WINNERS' debunking of several urban legends about path dependence and lock-in. Several now famous lock-in stories are reviewed, most notably the QWERTY keyboard and the VHS videotape format. Leibowitz and Margolis thoroughly debunk both stories. As for the supposed superiority of the Dvorak keyboard over the prevailing QWERTY model, for example, they show that the studies on which the claim of Dvorak's superiority was based were flawed. Drawing on evidence from ergonomic studies, computer simulations, and training experiments, they find that the Dvorak keyboard offers no significant advantage over the QWERTY model. Hence, the QWERTY fable (as they call it) does not even demonstrate second-degree path dependence, let alone the real thing.

Of primary interest to the general reader, however, will be Leibowitz and Margolis' extension of their argument to the antitrust case against Microsoft. The success of Microsoft's Windows operating system over alternatives such as Mac or OS/2 is frequently attributed to a network externalities/lock in story. Assume for a moment that the lock in story is true. It is not clear what the antitrust implications ought to be. Antitrust historically was justified by concerns that monopolies are harmful: they result in lower output and higher prices than competitive markets. It is not at all clear that this is true of software markets. Even if it were true, the positive externalities that result from the network effect might well outweigh those costs. Hence, a monopoly created by network effects arguably ought to be deemed a natural monopoly and, as such, tolerated by the antitrust laws. (The use of a natural monopoly in one market as leverage to monopolize a second one, as Microsoft is alleged to have done, of course, cannot be so justified.)

If Leibowitz and Margolis had limited their arguments to these points, the book would not have been controversial. (It also wouldn't have sold as well or gotten the attention it so richly deserves.) Instead, however, they try to debunk the claim that Microsoft's products are inferior ones that prevail only because of their network effects and path dependence. Here, I think they are on weaker ground than in either the QWERTY or VHS examples. Their primary empirical evidence on the quality of Microsoft products is software reviews in leading computer magazines with respect to a comparative handful of product markets. This is one plausible measure of quality, but surely it is not the only one. (I have often wondered whether the volume of advertising in those magazines effects their editorial content, for one thing.) Hence, it seems to me, that they overstate the case when they assert their data prove that Microsoft prevails in market competition because it makes high quality products that it sells at low prices. Having said that, however, this is an incredibly important and remarkably accessible introduction to the key economic principles that underlie the software markets.

For those whose memories are short, Liebowitz and Margolis provide poignant reminders that Microsoft's success -- like that of the QWERTY typewriter and VHS video format before them -- was by no means pre-ordained or simply the work of imposing sounding "monopolistic practices." Where Microsoft has provided the best product at the most reasonable cost, it has succeeded. Where it has not -- i.e. Microsoft Money and MSN -- it has failed.

This book should be required reading of those government bureaucrats who now seek to "remedy" that which was never broken.

No economist (or team of economists) has produced during the 1990s research that is as compelling, as exciting, as important, and as well-grounded in both theory and data as that of Stan Liebowitz and Stephen Margolis. These guys' work is truly and deeply impressive. And their writing is a joy to read.

They tackle some of the most challenging issues at play in modern economics -- and they succeed brilliantly! Because of the research these two scholars produced earlier in this decade, it is inexcusable for anyone to trot out the success of the QWERTY keyboard or that of the VHS tape as examples of "inefficient lock-in." Because of the research they have done, and report clearly, in this new book, it is now inexcusable for anyone to claim that Microsoft's market success is due exclusively -- or even principally -- to "path-dependency" or "inefficient lock-in."

Liebowitz and Margolis show beyond any reasonable doubt that in those markets where Microsoft achieves and maintains a large market share, it does so by offering consumers deals that consumers fine attractive. Period.

WINNERS, LOSERS, AND MICROSOFT is economics at its finest and most relevant.

This book presents a compelling discussion of economic theory and offers scrutiny of several flawed theories of market failure with specific regard to Microsoft and the government's antitrust case against the software giant. Winners, Losers, and Microsoft gives noteworthy insight into the underlying assumptions of these deficient economic theories as well as into the perceptions used as their foundation of evidence. Leibowitz and Margolis point out crucial flaws in these theories which are widely used today by some economists. They also seek to destroy the government's evidence for Microsoft exhibiting a damaging monopoly role in the software industry. The authors' assertion is that markets cannot be oversimplified through questionable assumptions and faulty evidence in order to come to a convincing theory of market failure. Their main claim is simply that certain technologies win out over others in time because they are of superior quality.
The text begins by offering up the economic theories that are later attempted to be disproved. The main topic of contention is the fact that some economists, especially those involved in the antitrust case against Microsoft, claim that market failure is possible in new technology industries because consumers are forced into buying inferior technologies because of network effects, the notion that consumers' buying habits are heavily influenced by the number of other consumers using that product. Further detail is explained concerning a strong kind of network effect which is also claimed to lead to market failure, a standard. Another confused claim by some economists is that Microsoft's elevated position in the software industry is detrimental to competition and a case of monopolistic harm being done, disregarding the fact that Microsoft could have had the better product.
Another issue that the authors question is that economic theories of path dependence are strong support for the idea of market failure. The authors argue that there is a very real possibility that such models do not translate well to real-world situations and market processes. They go on to point out that there would be many more signs and proofs of market failure were there any validity to that theory. They explain that path dependence theories rely on evidence from past technologies that isn't reliable or even accurately reported. It is also mentioned that economics literature gives multiple inconsistent meanings to path dependence and that a system to classify path dependence at three levels is possibly appropriate. Liebowitz and Margolis define first degree path dependence as making the right predictions about the future but those decisions seeming erroneous at points along the way, but essentially no harm occurring. Second degree path dependence is described as making the wrong prediction about the future but you did so from the standpoint in which you made the right decision given the available information. Finally, third degree path dependence involves knowing that you are making the wrong or inefficient decision but still making that choice anyway. It is this third and most extreme view of path dependence that supports claims for market failure.
Past technologies that are said to prove the existence of this market failure are then discussed, including the QWERTY keyboard and the VHS video system. The QWERTY keyboard was said to be an inferior technology that was harder on the user and led to slower typing speeds. The main competitor to the QWERTY, or Universal, keyboard was the Dvorak Simplified Keyboard (DSK) that was introduced in 1936. August Dvorak claimed to have designed a keyboard that he supposedly proved was easier to learn and that reduced finger movement. The QWERTY keyboard, having been patented in 1868, was designed with the keys arranged in such a way that the typewriter for which it was designed operated the best. The main piece of evidence that economists and proponents of market failure (lock in) use to support this case as being market failure is a 1944 Navy study conducted to compare typing speeds between the two types of keyboards. This study, once carefully examined, reported flawed and incomplete data yet came to a decisive conclusion. This study was also believed to have been done by Dvorak himself, an important piece of evidence. Yet somehow this data is used consistently to this day as evidence that the allegedly inferior QWERTY keyboard should have never prevailed over the DSK. Also, most of the other body of evidence for the supposed preeminence of the DSK has proven to be based on anecdotal research and inconclusive studies.
The evidence for the superiority of Beta over VHS is another example of faulty reporting of research leading to misconceptions about the technology that is higher in quality. This text points out that the quality differences between Beta and VHS weren't as great as may have been reported and that, despite the slightly better picture quality, Beta lacked another feature (longer tape length) that most consumers wanted at the time of their heated competition and possessed attributes that weren't necessary to the average consumer (better editing and special effects functionality). Their conclusions on this matter were that market failure couldn't be said to have happened because the better product won the battle (VHS) and that maybe Beta was better suited for broadcasters and not the average home user.
The next section of the book addresses the accusations that Microsoft demonstrated injurious monopoly actions through elevating prices and impeding superior technologies from achieving market share because of lock-in network effects. The authors show evidence through software industry price data that in the markets where Microsoft achieved control of a majority of the market share that prices actually tended to decline in every case. This evidence overrules the monopolistic accusations of Microsoft elevating prices due to control of the market by network effects. It is stated from the assembled data that, in markets where Microsoft competes, consumers have seen price declines of approximately 60%. Also taken into account is the notion that tipping, the idea that market share should grow at an ever increasing rate due to its increased attractiveness to consumers, can occur in markets where there are significant network effects. In fact, where Microsoft has dominated the market it has been shown to steadily grow to that dominance and not the sudden jumps and accelerations expected from tipping.
The analysis of the spreadsheet, word processor, desktop publishing, and online services markets was intriguing because all of the criticism and accusations of Microsoft in each of these markets seems to lack any shred of evidence. The authors' contentions are based on credible economic patterns and widely accepted principles. Through examining market share (based on revenue) and product quality (based on a multitude of reviews), it is shown that there is convincing substantiation for the claims that Microsoft dominated the markets in which it had higher quality products and did not achieve significant market share when reviews showed Microsoft's product quality lacking. These conclusions contradict the accusations of lock-in due to network effects because Microsoft's product quality was consistently rated the best when it achieved a large market share, not the anecdote that Microsoft had inferior products but succeeded due to blocking competition. The authors make convincing arguments repeatedly for the fact that the best product will come out on top in a laissez-faire economy and that this is exactly what happened time and time again with Microsoft.
Following this whole discussion is the appendices of the text which offer analysis of the charges and the rulings against Microsoft by the government. There are repeated instances of the Justice Department using economists Franklin Fisher, Brian Arthur, and Garth Saloner to instill the belief of network effects, lock-in, and path dependence into the judge and jury. Disproving this evidence might have been Microsoft's best chance at saving their case considering that these theories and economic conditions are completely theoretical and have little basis in fact. Also mentioned is the fact that the Department of Justice discredited many of Microsoft's witnesses and managers, not helping the defense' dismal situation. The eventual ruling by Judge Jackson reflected the nature of Microsoft's legal defense - extremely ineffective. Jackson ruled ultimately that Microsoft was guilty of all antitrust charges brought against it but one, almost completely favoring the Department of Justice' opinion. What was the most disturbing and surprising to me was the fact that Judge Jackson implied in later interviews that he made his decision based not on the content of what the defense had to argue but on the fact that many had been discredited by their untruthfulness. Jackson also made very few changes to the government's proposed solution to the Microsoft problem because he didn't feel competent to recommend any solution himself. I think this only highlighted the fact that Judge Jackson didn't have enough knowledge or understanding of economics to rule effectively or fairly on the case, simply going by whatever the government had to recommend.

This book is unfortunately titled as it is really primarily about bringing real data and rigor to bear on many of the conventional "stories" about the economics of the new economy, rather than dwelling on the Microsoft antitrust situation. Clearly, the new economy tends to be characterized by more network effects, increasing returns to scale, and general "winner-take-all" effects than the historical economy. However, certain stories about early lock-in effects of technologies that are inferior, but that by luck got the early lead, have been uncritically passed from author to author. Examples are the QWERTY keyboard, VHS vs Betamax, Windows vs Apple, etc. Liebowitz and Margolis show that most of these stories do not hold up under close examination -- that in fact, these are not examples of the market failing to take the "right path". Actually, the market generally seems to get it right. This book is also the best we have seen in its treatment of the overall economics of information technology standards.