Are IP and Essential Facilities “fundamentally at odds”?
Among the stock arguments in the debate whether IP rights can be essential facilities and if so whether and under what conditions mandatory RAND licenses can be pursued under Section 2 is the claim that a duty to license (e.g., API specifications) is fundamentally at odds with the grant of the IP right itself. That is because patents and copyrights explicitly involve the power to exclude others from infringing those rights. A consequence of this argument has been the call for stronger protection against antitrust duties to share for IP rights than is afforded to tangible property. The CSU v. Xerox case is an illustration of this line of reasoning. Irrespective of one’s policy position with respect to essential facilities claims, the argument that duties to license are fundamentally at odds with the IP grand is unconvincing, because both IP rights and essential facilities are expressions of very similar tradeoffs between losses in short term static efficiency in order to promote long term dynamic efficiencies. Copyrights and patents are granted to promote the progress of science and useful arts in the long run. The right to exclude, and the costs that come with it, is a means to that end. Essential facilities recalibrate the exclusion/incentive tradeoff where the expected gains from maintaining the upstream incentives to innovate and invest are outweighed by the expected gains from increased downstream competition. Given the structural similarity of the tradreoffs involved, essential facility tweaks to the IP exclusion default are not “fundamentally at odds” with the IP grant. Rather, both are expressions of the same policy concerns. That doesn’t mean, of course, that such tweaks — or exceptions to the IP rules — are always appropriate, as some facilities are far more likely to induce significant downstream welfare gains from being opened than others. It does mean, however, that one cannot simply avoid the issue by pointing to an existing IP right.
RSS Feed
December 2nd, 2009 at 4:53 pm
Well said. In a paper posted to SSRN, I have argued that the essential facilities doctrine should apply to natural monopolies, whether obtained through IP or otherwise. The problem with applying EF doctrine to IP seems to be that most IP looks like a natural monopoly when measured in terms of fixed costs versus revenue over time whereas only IP that protects platforms or standards is really natural-monopoly like. One alternative is to put doctrines from IP to work productively in helping identify natural monopolies more easily — doctrines like the originality doctrine of copyright or the reverse DOE in patents.