Submission Number: 00024
Received: 6/16/2011 2:06:56 PM
Commenter: Valerio Torti
Organization: University of Wisconsin
Agency: Federal Trade Commission
Initiative: Request for Comments and Announcement of Workshop on Standard-Setting Issues, Project No. P111204
Attachments: No Attachments
LICENSING POLICY RULES AT THE CORE OF THE HOLD-UP PROBLEM _________________________________________________________ The concrete risk of patent hold-up should be carefully appraised and linked directly to the effectiveness of the policy rules of standard setting organizations. Indeed, the analysis of these provisions may lead to the conclusion that hold-up is facilitated (rather than contrasted) by those licensing rules widely adopted so far by SSOs, i.e. FRAND terms. The divergent interpretations of FRAND given by the literature have not clarified when licensing royalties may be deemed excessive or fair. The indefiniteness of the meaning of these conditions usually allows the owners of IPRs to defend any price finally charged as a fair and reasonable rate. Therefore, it is clear that FRAND terms are not an efficient solution mainly because they leave potential implementers of a technology uncertain as to the economic terms on which essential patents will be licensed to them. Such uncertainty, increased by the divergent economic theories developed by the literature, may lead key market players to avoid SSOs processes, or to hesitate in developing technologies which may still be the subject of litigation among interested parties. These considerations seem to justify the view that a FRAND licensing model implies more questions than it gives answers. In order to avoid the risks of subscribing undefined FRAND terms and litigating their meaning before a court, SSOs’ members could negotiate ex ante the specific price terms under which licensing any rights. This model would avoid uncertainty on the level of royalties patent holders may finally charge, and might thus function as an incentive for implementers to take part to the standard setting process. However, under these circumstances, there could be the risk that discussions on licensing terms may ultimately lead to exhausting policy battles between SSOs’ participants, which may finally compromise the whole standardization process. Perhaps, the most important reason for excluding the ex ante policy model lies on the fact that negotiations might be interpreted by innovators as dangerous and inconvenient. In other words, IPRs owners may well fear that negotiating licensing terms with implementers (representing usually the vast majority of participants) may finally lead the price down. Indeed, industry participants have usually all the interest to keep the royalty rate low. In light of these arguments, ex ante negotiation of licensing terms would not probably represent the most efficient solution. A proper licensing policy framework, I believe, may be based on a rule requiring IPRs owners to disclose unilaterally the maximum price they would charge for any patents they own covering the standard. Such a model has already been mentioned by the European Commission in its recent Guidelines on Horizontal Cooperation Agreements, and has also been considered by a few organizations. Under this framework, innovators would have to set ex ante and unilaterally the highest rate to be applied in case their standards were selected. The acceptance of a maximum cap regime may drastically reduce the concerns for patent hold-up and the related inefficiencies. Indeed, this option would avoid the risk of lengthy litigation on the meaning of unclear licensing policy, and would probably discourage IPRs owners from charging fees higher than those established. This is because a maximum royalty cap may have more chances to be enforced before a court than a FRAND model may ever have, and may hence work as a deterrent from implementing misleading conduct. A maximum cap regime, furthermore, may also prove to work better than the ex ante negotiation model. Absent the risk to be imposed low prices by implementers, IPRs owners would be more encouraged to participate to standard setting. In brief, this would explain why it should be preferred to any other model in the setting of SSOs IPRs policies.