I've always been confused by the "natural monopoly" defense of Microsoft. The traditional natural monopolies were the regulated utilities, a status I'd assume Microsoft doesn't aspire to. Though it may fit there, under the "essential facilities" theory, too. The new wave "natural monopoly" theory is Brian Arthur, network effects, increasing returns, etc.; that analysis also tends not to favor Microsoft.
The argument is simple: the structure of the software market is characterized by high sunk costs, decreasing marginal costs of production and network effects, all characteristics of a natural monopoly.
The most efficient form of production in such a market is a single producer. Any effort to break up the single producer will lead to less efficient production, less output, and an unstable competitive environment, at least until a new monopolist emerges.
As for W. Brian Arthur, he's been accused of rediscovering Austrian economics. I will grant that he did apply a mathematical patina to the field, but I leave it to others, specifically the disciples of L. Von Mises, to say whether that represents genuine progress.
Another thing: Arthur makes a convincing case for the existence of multiple equilibria in the economy. What he does not show so convincingly is that the equilibrium that will result from the kinds of government intervention he advocates will make people better off than they were pre-intervention.
In the middle, we have the Chicago School, I guess, which says let'em be, more or less. I'm not sure exactly how much legal standing the Chicago School has, it certainly has adherents on the bench, but I don't think that interpretation is enshrined in case law as the one true view of antitrust.
Bork's book basically says, don't break up companies to less than their efficient size in the name of antitrust, because efficiency is all that antitrust is about.
The alternative is conduct remedies. However, in "Natural Monopoly and Its Regulation," Posner makes the case that regulation of natural monopoly is inefficient. He states in the article that the article presupposes a strong antitrust enforcement effort, but the article was pre-Bork.
Also, the kinds of regulations that regulators impose on natural monopolies in the regulated sector of the economy are strikingly similar to the "conduct remedies" proposed by latter day interventionist adherents of the Chicago School approach (including the various proposals to regulate Microsoft). In fact, I'd say they are conceptually idential and that the arguments Posner makes against them in the regulated arena apply with equal force in the arena governed by antitrust law. At any rate, I think that's what Microsoft will probably argue. |
