3.24.00
What Hath Ward Wrought?

3.24.00
Leave E-commerce Alone

3.24.00
E-mailgate

3.23.00
Senator, You're No Ronald Reagan

3.23.00
Against Microsoft: A Primer For Conservatives

3.22.00
Kessler Control

f the hundreds of thousands of words that have come tumbling out of word processors since the trial court’s decision in the Microsoft case, most are hostile, often bitterly so, and almost all of those are written by people who would generally be classified as conservatives. Since modern conservatism professes devotion to free markets and to the rule of law, this virulence suggests an inadequate understanding of the case. That, sad to say, seems to be true of National Review and the writers it chose to bolster the pro-Microsoft view.

The facts need to be restated. Microsoft’s monopoly is protected by the “applications barrier”: Consumers want an operating system that has the most applications (word processing, etc.), and applications writers go to the system with the most consumers. This is the “chicken-and-egg” problem that makes it impossible for competitive operating systems to thrive. But Netscape’s browser, which first made access to the Internet easy, also has the capacity to receive applications, so that, if it were used widely enough, neither consumers nor applications writers would care what operating system underlay the browser, and Microsoft’s monopoly would gradually erode. The core of the case was Microsoft’s attack on Netscape’s browser by, among other tactics, incorporating its own browser into its operating system, thus making it inconvenient and expensive to use Netscape’s browser as well. Computer manufacturers had no choice but to take Microsoft’s browser; Netscape, having no operating system, could not effectively respond. Microsoft’s internal communications made clear that the object of incorporating its own browser in the operating system was entirely predatory, to “cut off Netscape’s air supply,” and not, as claimed at trial, to increase efficiency or respond to consumer desires. Indeed, Microsoft executives were clear that on the merits, their browser was unlikely to win against Netscape’s.

NR’s editorial comment opens with the least promising line of attack — that Microsoft does not really have a monopoly of personal-computer operating systems. That will come as news to computer manufacturers, who have repeatedly lamented, often in anger, that there is no viable substitute for Windows. Manufacturers must have it or shut down. No more realistic is the contention that because most consumers already own some version of Windows, Microsoft is its own competition, which accounts for prices “declining in every market in which Microsoft has a product.” Judge Learned Hand rejected this argument in his 1945 Alcoa opinion, but it is irrelevant anyway, because consumers automatically get the latest version of Windows whenever they buy a computer. Microsoft does not lower its price for its operating system; any price decline is owing to machine and chip manufacturers.

That Netscape, as a company, survived Microsoft’s assault is also irrelevant. The issue is the destruction of Netscape’s threat to the applications barrier that protects Microsoft’s monopoly. NR’s prediction that America Online and Netscape, after the acquisition by AOL, would control 58 percent of the browser market has proved considerably wide of the mark. AOL has prudently decided not to challenge Microsoft’s browser with Netscape’s, with the result that Microsoft’s current market share is about 70 percent and rising. The browser war is over.

The failure of the editorial’s arguments means that NR’s position must be vindicated, if at all, by the two articles published in the same issue. That is unfortunate. One article, by Prof. Richard Epstein, addresses the merits of the controversy, and the other, by Holman W. Jenkins Jr., attributes discreditable motives to just about everyone who supports the court’s decision. Neither is remotely adequate. Mr. Epstein displays a thorough incomprehension of antitrust theory and hence of the case against Microsoft. He starts with the mistaken notion that “it is virtually impossible, historically, to identify any successful case of predation” because the predator must lower its prices below cost and suffer greater losses than the prey. As I explained over 20 years ago, that argument holds only when the predator must expand output and sell below marginal cost. There are numerous successful instances of predation when those constraints were not present. Microsoft, having a monopoly of operating systems, did not have to increase its rate of output, and its marginal costs were negligible. Microsoft continued to make profits on the joint product of operating system and browser, while Netscape, lacking an operating system, had to offer its only product, the browser, free, which is certainly below any cost you care to name. It is nonsense to say that Netscape was not hurt because it was free to offer its browser for nothing. That is freedom to lose money forever. But Epstein loses all credibility with the following fantastic statement: “The blunt truth is that this tie-in is worthless.” Then why did Microsoft insist upon it? If you want a blunt truth, it is that the tie-in drove Netscape’s browser from the market, and to Microsoft that was worth a great deal.

The fact that Microsoft attacked Netscape by incorporating its own browser in its operating system leads Epstein to accuse me of invoking the “shopworn theory of tie-ins that [my] own earlier work . . . largely discredited.” Wrong. What I and others discredited was the notion that a monopolist could gain a second monopoly profit by tying the sale of a complementary product to the sale of the monopoly product. When products are complements — razors and blades, for example — the entire monopoly profit can be extracted on the sale of the monopoly product. The razor purchaser will pay a monopoly price for a shave but cannot be made to pay two monopoly prices for razor and blade.

Epstein then gets the argument backwards by saying that I think the Microsoft tie-in is different because it was “used not to enhance the monopoly over operating systems, but to prevent Netscape from competing in the browser market.” To the contrary, I have written repeatedly that the incorporation of the browser in the operating system was not designed to obtain a monopoly profit for the browser but to protect the Microsoft operating system by removing Netscape’s browser as a receiver of applications, i.e., to enhance the operating-system monopoly by removing a potential substitute.

He also brings up the old complaint that there was no antitrust injury because consumers were not harmed. Clearly, they were. When competitors fix prices, consumer harm is conclusively presumed. There is no reason for the action other than to alter the competitive price. It is so here. Microsoft’s use of predatory tactics to defend its monopoly indicates that, absent such tactics, competition might more rapidly erode that monopoly. Consumers would have been offered lower prices and competitive innovation sooner.

Holman Jenkins attempts no analysis of the case’s merits but contents himself with offering a false narrative of the industry and insulting those who think Microsoft might be guilty of anything. He describes Microsoft as primarily responsible for cheap, powerful desktop computers as well as the Internet’s information cornucopia. Those are not the facts. The powerful and inexpensive desktops were created by computer and chip manufacturers. Microsoft did not lower the price of its operating system; and it had as much to do with the creation of the Internet as Al Gore. Microsoft bought rather than invented its most important technologies and deserves credit not as an innovator but as a marketer.

But the misinformation about Microsoft’s achievements is as close as Jenkins comes to analysis. Instead, he reverts to the ad hominem abuse that has characterized his writings on this case from the beginning. In his latest version, he sees the case as the government “cutting an object of envy down to size,” in which task it is supported by a “substantial part of the public” that wants Bill Gates punished for daring to “put himself forward.” He supports this remarkable antitrust analysis with references to Sigmund Freud. Jenkins would do well to master the economics relevant to antitrust before turning to psychoanalytic explanations. The best that can be said for his fantasy is that it seems to accord with Bill Gates’s view of the case. Jenkins would do well to leave persecution complexes to the executives of Microsoft, who are masters of the paranoid style. A journalism of personal destruction is no more attractive than a politics of personal destruction. Both are forms of McCarthyism, which is today almost exclusively the preserve of the liberal Left, and should be left to them.

The Microsoft case will soon move on to the remedy phase. That will present more complex problems than the conclusion that Microsoft violated the Sherman Act. The Department of Justice will have to decide whether to seek a remedy addressing only Microsoft’s behavior, or urge structural relief. Either way, there are complicated issues to be addressed — but conservatives should not reflexively oppose consideration of the dissolution of Microsoft. There are advantages to a structural remedy that should not be overlooked by free-market advocates. A structural remedy would avoid detailed regulation by court decree. Regulation of this industry is probably the worst of all possible courses; certainly no one should want to repeat the court supervision of the telephone industry following the AT&T breakup. Not only might the effects be deadening, but a detailed decree would be vulnerable to Microsoft’s demonstrated capacity to maneuver around prohibitions. The AT&T experience does, however, provide one hopeful aspect of a dissolution of Microsoft: The parts of the company after dissolution might have far greater value to shareholders than the intact company does.

What we need now is less slogan-mongering and more thoughtful analysis of a complex topic. A position that boils down to “Private sector good, government bad” is less a philosophy than a tantrum. Aaron Director, the founder of the law-and-economics movement at the University of Chicago, was fond of saying that laissez faire never meant more than that proposals for government intervention should be examined under a presumption of error. Indeed, that is all a sensible, rather than a knee-jerk, conservatism can mean. The facts of the Microsoft case are easily sufficient to overcome that presumption.

Richard A. Epstein
Judge Bork’s pointed defense of Netscape again confuses the welfare of Netscape with the welfare of consumers. His limp conclusion is that antitrust injury to consumers should be presumed from Microsoft’s aggressive entry into the browser market, just as it is in horizontal price-fixing cases. But why? Rigged prices exclude some consumers altogether from the market, and raise the costs to others. Huff and puff as he will, Bork cannot deny that over the short run, Microsoft’s entry lowered, not raised, consumer prices, and undercut the Netscape monopoly in the browser market.

So what long-term consequences should lead us to reject these gifts? Bork claims that the case involves a potent mix of predation and tie-ins that insidiously undercuts innovation by imposing barriers to applications. He writes as though giving away the browser for free condemns Netscape to perpetual losses. Wrong. As to predation, the entire prospect is far-fetched when the marginal cost of additional copies of Netscape or Explorer is virtually zero. How is it possible then to distribute additional copies at below cost? Indeed, Netscape now offers its browser for free — because, like Microsoft, it hopes to earn revenue from third-party advertisers and vendors. Bork never explains why the absence of an operating system prevents Netscape from carrying out the competitive counterstrategy it has in fact employed. Clearly, there’s no predation here; if Bork were right, Netscape would have just abandoned the market entirely.

Nor is Bork right on the tie-in question, because there is an obvious alternative explanation. Breaking Explorer apart from Windows ignores the efficiency benefits that integrated programs give to those consumers who prefer Microsoft Explorer, while prolonging Netscape’s monopoly in the browser market. It is wrong to look at only the undesirable consequences of a tie-in; its upside for consumers counts as well.

Anyone who finds Bork’s consumer-injury claim convincing need only read the final pages of Judge Jackson’s decision. Jackson starts with a huge concession: “The inclusion of Internet Explorer with Windows at no separate charge increased general familiarity with the Internet and reduced the cost to the public of gaining access to it, at least in part because it compelled Netscape to stop charging for Navigator.” That is a monumental boon to consumers, and in his attempt to minimize it, Jackson can only come up with a list of minor and fragmentary inconveniences: clutter on the computer top, in an age when computer capacity is measured in gigabits; a slower opening of applications, when the speed of computers is increasing exponentially; the forcing of consumers to take Explorer when they would rather have no browser at all, when there must be just a hundred such people in existence. Where’s the beef? All of these obstacles put together were not enormous enough to prevent AOL from gobbling up Netscape for a fancy price. None of them remotely justifies breaking up Microsoft, which will now have its hands full in trying to compete with the newer AOL-Time Warner. In a sense, Netscape has won: The present government suit condemns Microsoft to market immobility in an age of mergers. Is this the antitrust case of the millennium? Let’s break up today’s Chicago Bulls instead.

Holman W. Jenkins Jr.
Judge Bork challenges my antitrust analysis, but I wasn’t offering an antitrust analysis. I was musing on cultural aspects of the Microsoft case. It was the public, I said, that saw the lawsuit either as “government shamelessly attacking a success story, or necessarily cutting an object of envy down to size.” I don’t happen to agree with Bork on the merits of the lawsuit, but surely there is more to the world than antitrust arguments. The opponents of Microsoft did not just file legal briefs. They worked to shape public opinion. Come to think of it, didn’t Netscape hire Bork to influence public opinion? I guess that’s because Netscape thought public opinion was important.

Bork also says I attribute “discreditable motives” to Microsoft’s opponents, but it would be more true to say I attribute “business motives” to them. Even if I agreed with the merits of their case, I would suspect that their motives were based more on market imperatives than on legal reasoning.

For that matter, I might be tempted to agree with Bork that Microsoft risks some exposure under the antitrust laws and yet still believe that antitrust is a poor use of public policy. Intel has announced a new line of PC devices that don’t use Microsoft software. AOL just bought up Time Warner. With each day that passes the kinds of concerns that were used to peddle the lawsuit to the larger public seem more ludicrous and far-fetched.

Columns / Current Issue / Goldberg File / Nota Bene
Washington Bulletin / Subscribe / Ad Info / Home