Friday, September 21, 2007
Posted by Bill Page
It is probably a sign of progress that all sides in the Microsoft litigation, both here and in Europe, seem to agree that the relevant goal of antitrust enforcement is to benefit consumers. The CFI’s Microsoft decision, for example, mentions consumers well over 100 times, usually in the context of considering whether practices affect “consumer welfare,” “consumer choice,” or the like. But the simple fact that the opposing sides with such fundamental differences can invoke the same interest in support of their positions suggest that we need further clarification of what the consumer interest means and how best to advance it.
I suggest that an appropriate starting point is to ask: what is the immediate and obvious effect of the challenged practice on consumers? If the immediate effect is to benefit consumers, then the plaintiff should face a heavier burden to show (by a coherent theory and supporting evidence) that the long-run or net effect will actually be to harm consumers. If the immediate effect is to harm consumers, then the defense should face a heavier burden to show that the long-run effects will be positive. This approach recognizes that future market events, especially innovations, are difficult to predict, and courts have a dubious track record in their efforts to do so. Something like this idea may explain the many hurdles the Supreme Court (in Matsushita and Brooke Group) has placed in the path of predatory pricing claims, because price cutting undoubtedly benefits consumers immediately. The idea also may explain why the Court took so long to overrule Dr. Miles: resale price maintenance undoubtedly raises retail prices; the Court decided (in Leegin, just last term) that per se illegality was inappropriate only after the Court was convinced that consumers may nevertheless benefit from valuable services that the higher prices elicit.
In the U.S. Microsoft case, the D.C. Circuit properly took this approach to reverse several of Judge Jackson’s holdings that had condemned Microsoft for offering better products (e.g., the Windows-specific version of Java) or free services. It wrote “The rare case of price predation aside, the antitrust laws do not condemn even a monopolist for offering its product at an attractive price, and we therefore have no warrant to condemn Microsoft for offering either [Internet Explorer] or the [Internet Explorer Access Kit] free of charge or even at a negative price. Likewise, as we said above, a monopolist does not violate the Sherman Act simply by developing an attractive product.” The D.C. Circuit also, however, upheld liability for various forms of integration of Microsoft’s browser with the Windows operating system. In these parts of the opinion, we have argued, the court gave insufficient weight to the immediate benefits of integration to consumers, and undue weight to speculation about possible long-term harms.
The CFI has made the same error in its analysis of Microsoft’s tie of Windows and Windows Media Player. The Commission’s decision rested on the theory that Microsoft harmed consumers by not offering them a choice of Windows without Media Player. Thus, it was OK for Microsoft to combine the media player and Windows, so long as it gave OEMs and ultimately consumers the option of buying a version of Windows without Media Player. The EC’s theory, of course, has been decisively refuted by the failure in the market of “Windows XP N,” the stripped version of Windows that Microsoft developed to satisfy the Commission’s order. Consumers didn’t want the stripped version, even though OEMs could have installed a non-Microsoft media player in place of WMP—that was evidently not an adequate substitute, despite the EC’s contentions to the contrary (¶ 923). The CFI insisted (¶ 943) that it could not consider the failure of XP N because is required to decide the case based upon the facts and law at the time of the commission decision. But it should have been obvious to the EC from the outset that no one would want a version of Windows with less functionality at the same price. Consumers (and OEMs) understandably view Microsoft’s provision of a media player along with Windows an immediate and unalloyed benefit. The fact that other, competitive OS vendors like Apple also provide media players along with the OS confirms this obvious point.
Under the approach I am suggesting, recognition of this immediate consumer benefit should have triggered a substantially higher burden to show that the action would harm competition. The fact that Microsoft gains an “advantage” by providing an obvious benefit should not make its actions unlawful, in the absence of a coherent theory, supported by evidence, that the practice will harm consumers in the long run. I suggest that the EC did not meet this heightened burden. Some ties can reduce competition. But, contrary to the CFI’s findings (¶¶ 967-70), I submit that bundling a product that is free in every channel of distribution and that does not otherwise prevent installation of rival products is not seriously exclusionary. OEMs and users can and do freely install other media players and content providers can and do encode their video in multiple formats.
The CFI accepted the EC’s finding that Microsoft’s bundling gave it an “unparalleled advantage” over rival producers that altered the incentives of OEMs and content providers in ways that and “inevitably had significant consequences for the structure of competition” (¶ 1054). These effects raised a “reasonable likelihood” that “an effective competition structure would not be ensured in the foreseeable future” (¶ 1089). But any harm to consumers, as opposed to rivals, from these predicted developments are speculative.
The EC emphasized, of course, that it did not want to prevent Microsoft from bundling, only to require it to sell an unbundled version. This aspect of the remedy certainly limits the costs of the EC’s order, but at the expense of making it ineffective in achieving any competitive benefit.
September 25 Update: For more confusion about how to help consumers by antitrust enforcement, see Unbundling Microsoft Windows, published by a Brussels think tank. The authors argue that it would benefit consumers to require that computers and operating systems (except Macs) be sold separately. Thus, consumers would have to buy the hardware and software separately and install the software themselves. This "would have a significant effect on the market share of Windows, providing the competitive marketplace that Ms Kroes has called for." For an expert refutation of this logic, see http://arstechnica.com/news.ars/post/20070924-eu-group-calls-for-end-to-windows-tax.html