This paper – which you can find here – provides a short critical introduction to the intersection between IP and antitrust laws. It begins by pointing out the obvious: in exchange for public disclosure, the patent system gives a successful patent applicant the right to exclude others from using the patented technology without her/his permission for a period of time. The first instinct is to consider that this exclusivity amounts to a legal monopoly, which can make antitrust antennas start to tingle. However, it is now a truism that having an IP right does not mean one also has market power. Nonetheless, in the US a number of (older) decisions refer to “patent monopolies.” This piece ask whether this “patent monopoly” language is useful, misleading, or something in between. Without surprise, the authors seek to argue that it is exceedingly misleading: simplistic assertions that patent rights constitute “monopolies” are not particularly informative or helpful.
The argument starts by looking at the relationship between the patented product and market definition. It notes, correctly, that “It is widely recognized that, in certain contexts, other technology (whether patented or unpatented) may be a close substitute for the particular patented technology at issue.” It further notes (I’m not sure that as accurately) that: “In most situations, a patent holder does not hold a “monopoly” over a relevant technology market. The relevant technology market also includes the other (patented or unpatented) technologies that compete with the particular patented technology at issue.”
So far, so standard. But at this point the discussion gets more interesting. Patent rights are not self-enforcing; patent holders cannot physically withhold their technology from others because the technology is disclosed in the patent itself, and the patent is published. In particular, patent holders have to enlist the assistance of courts in order to enjoin others from infringing patent rights. And even if a patent is recognised, courts may decide not to grant injunctive relief, but grant damages instead; in such situations, the patent holder will have lost the ability to control the use of its patented technology. Furthermore, a patent grant need not give the patent holder the right to actually practice the patented invention. Other parties may have patent rights as well, and the patent holder may need permission from others to practice its patent. This can be the case of “improvement patents” and, importantly, of formal compatibility or interoperability standards. In the case of standard essential patents there may be hundreds if not thousands of patents, held by dozens or hundreds of firms, which are claimed to be “essential” to make and sell products compliant with a given standard.
In such scenarios, it is (the authors argue) unlikely that a single patent will have market power, particularly if it is subject to FRAND licensing conditions. Looking in more detail at FRAND, the authors acknowledge that there is the possibility of “hold up” (a situation in which the patent holder is able to charge excessive royalties that exceed the “inherent value” of its technology), as well as the possibility of “royalty stacking” (where the implementer must pay royalties to multiple patent holders). However, market power is less likely in the context of FRAND than in its absence, since FRAND commitments act as a significant, contractually-enforceable constraint on the patent holder’s ability to exercise any “market power” over its patented technology. A last point the authors make, which is sometimes forgotten but that I think is quite important, is that patents are probabilistic: “There is only some probability that the patent, if asserted against a particular product, will be found both valid and infringed. Empirical studies of “win rates” (including one we wrote) show that only about half of litigated patents are found valid and infringed”. Consequently, in many IP-intensive industries, accused infringers can and often do practice the claimed technology without paying for it (though they may ultimately have to pay patent infringement damages should they be sued and lose). In this context, equating a patent with monopoly power is clearly an error.”
I am not really onside about how limited power an IP right grants, and I’m not sure I follow the authors’ argument on FRAND. While I would obviously not consider a SEP to inherently grant market power, the holder of an essential patent may nonetheless hold market power in the context of the relevant standard (as is recognised by both case law and doctrine). The fact that there may be hundreds of patents does not make it less so. After all, the market power may arise at least partially as a result of standard-setting (and, in particular, of the relevance for final products of the relevant standard). This is why FRAND commitments are required, and this is why one can say there may be competition issues (and abuse of market power) in this area. One can hold that such an issue can be addressed through contract instead of competition law, and there is undoubtedly force in such an argument ; but that is different from holding that one of many SEP patents cannot have market power. That is, ultimately, an empirical question.
Otherwise, I like the article. It could have been a more straightforward primer on why IP protection does not amount to market power for antitrust purposes; but it covers most bases succinctly and to the point, and gets the point across quite well.