Thure, I was wondering if, while I'm waiting for my copies of Arthur's books to arrive in the mail, you could you do me a favor and straighten me out on some things.
Please explain whether network effects lead to the creation of other than efficiency-based barriers to market entry. In other words, what I would like you (or someone) to explain is your perception of the literature's view on why network effects are a barrier to entry. And, is the barrier created by network effects Stiglerian or Bainian?
(FWIIW, I don't think there's any way on Earth it's Stiglerian -- but the point of my question is to encourage you to show me why I'm wrong.)
I get that the greater number of users enhances the utility of a network and that this draws more users and applications developers (in the case of Windows) in a "posiitve feedback loop." I understand that those users will (usually) make sunk cost investments in acquiring the ability to use the dominant network.
("Ability to use" includes both the cost of the software and the cost of learning how to use it. By giving away its browser, the network effects literature would suggest that, by eliminating the consumer's sunk cost in purchasing the software, Microsoft has actually lowered the barriers to entry in the PC browser market whose domination it has wrested from Netscape).
But, to me, network effects sounds like a type of efficiency, and not the sort of thing that would permit a monopolist to reduce output or raise prices beyond what is most efficient for a particular market (or at least the closest point that is achievable in that market to optimal efficiency). As Bork says:
Either there is no restriction of output, and hence no greater than competitive profit, to induce the entry or the growth of smaller firms; or the large firms are so superior in efficiency that, even if there is a restriction of output, no new firm would find entry profitable and no smaller firm is able to grow.
(This, of course, undermines the argument that breaking up Microsoft will enhance competition [which Bork equates with "efficiency" or "consumer welfare"].)
And, if this is true, how would more firms survive in the market for PC operating systems?
Isn't network effects literature as applied here basically saying that Microsoft is a natural monopoly?
BTW, Stigler is an economist at the University of Chicago; Bain hails from Harvard, I think. Bain was influential in the '50's and '60's, and takes a very broad view of the kinds of things that can constitute a "barrier to entry" that a firm can maintain to exclude competition. Stigler is influential in the Chicago school antitrust analysis. For example, Bork takes the view that the only barriers to entry are Stiglerian:
"What must be proved to exist, therefore, is a class of barriers that do not reflect superior efficiency and can be erected by firms to inhibit rivals. I think is clear that no such class of artificial barriers exists. . . . Stigler states: 'Barriers to entry arise because of economies of scale, or differences in productive factors, or legal control of entry.' None of these, of course, is properly subject to attack by antitrust. There is in the list no 'artificial barrier' other than legal control of entry by government,and since we want the more efficient firm to have the extra share of the market its efficiency commands, there is no reason for concern over the other barriers."
What I would like to know is: how do network effects allow Microsoft to reduce output below, or raise prices above, competitive levels, or, if it is a "natural monopoly," the most efficient levels possible in its markets? And, if network effects don't allow Microsoft to do either of these things, how is government intervention going to improve things from an efficiency standpoint?
And, if government intervention is intended to facilitate the replacement of Microsoft by a new natural monopolist (which is exactly what will happen if either of the two conditions outlined in the first Bork quote above is true), why is such facilitation a concern of antitrust law? In other words, why does antitrust law care who is the monopolist in a natural monopoly, or whether replacement of one natural monopolist by another in the same market is easy?
If the purpose of antitrust intervention is to prevent Microsoft from using predatory tactics to preclude a new technology from supplanting or commoditizing Windows, how do we know that the new technology will lead to a more efficient market structure than that dominated by Microsoft?
And, if Microsoft is a natural monopolist, how is Microsoft's having to give either present or future monopoly profits to consumers the form of a free browser anticompetitive in the sense that Bork defines the term? Would not the government's blocking such actions by Microsoft actually be more anticompetitive?
Maybe reading Bork is making me go nuts, but after thinking about this, I'm not even sure predatory conduct by a natural monopolist should be illegal. |
