Before I begin this post, I'd like to take the opportunity to thank Professor Michael Grünberger and the University of Bayreuth for their hospitality in hosting the conference this past weekend on Intellectual Property and the Public Domain. All of the presentations were outstanding, and it was great to meet and talk with some of the leading European IP scholars, judges, and students. I believe that at least some of the presentations will form the basis for papers to be published in a future issue of Zeitschrift für Geistiges Eigentum (ZGE) / Intellectual Property Journal (IPJ).
On the first day of the conference, Professor Alexander Peukert of Goethe University gave a very interesting presentation on the public domain, toward the end of which he critiqued the default rule under which IP licensees have no right to recover royalties they paid prior to the invalidation of the IP right. (This is the default rule, by the way, in the U.S., as well as in Germany, and Japan (see my book pp. 269, 331), and I would guess in many other countries as well.) Professor Peukert critiqued the rule because in effect the licensee may wind up paying for something that was in the public domain. In response, I suggested that his proposed elimination of the rule might make licensees worse off, because if licensees had an inalienable right to recover back those royalties they would in effect be shifting the risk of invalidity to licensors, and this in turn would provide licensors with an incentive to demand higher royalties to compensate for the added risk they (the licensors) would be taking on. Professor Peukert then clarified that he had in mind only a switch of the default rule, so that the presumption would be that licensees get their royalties back, but the parties could alter this presumption by contract. In his view, this would make licensees more conscious of the possibility that the IP might be invalid and, perhaps, less likely to license a right that may be invalid.
Perhaps he is right, but the Coase Theorem would suggest that the default rule is irrelevant if transaction costs are zero and there are no other bargaining obstacles. Of course, transaction costs are rarely zero, so often the default rule does matter, but in this instance is there any reason to think that it would be more difficult to contract for a right to recover back royalties under the current default rule, than it would be to contract away from this right under the proposed default rule? I'm not sure it is. I certainly wouldn't presume that licensees typically are the parties with weaker bargaining power, to the extent that matters; and I would think that before licensing an IP right one would give some thought as to whether it is likely valid and infringed. (There may be certain predictable exceptions to this expectation, e.g., when PAEs approach small businesses and threaten to sue; but we can deal with that problem in other ways, I should think.)
Another possible qualification of the Coase Theorem would arise when a party values having an entitlement more than she values acquiring it, that is, when there is an endowment effect present. (An endowment effect is present when the amount one would be willing to accept to transfer an asset is greater than the amount one would be willing to pay to acquire it. There's a large literature in behavioral law and economics on the topic.) If licensees value having a right to get back their royalties more than they would pay to acquire such a right, the efficiency consequences of one default rule over the other are ambiguous. But again it's not clear to me why we should assume that licensees (at least licensees that are corporate entities) would suffer from endowment effects; and even if they did, again at best I think all we can say is that the choice of default rule is unclear (though then I suppose one could use Professor Peukert's championing of the public domain as a tie-breaker). Finally, if there are few bargaining obstacles and parties normally don't depart from the current default rule, that might suggest that by and large they prefer for licensees to bear the risk of noninfringement and invalidity (and to be compensated for this in the form of lower royalties) than the other way around, in which case a switch of default rules itself would lead to needless transaction costs to bargain around it.
For these reasons, I'm not convinced by the proposal that the default rule should change, though I'm open to argument that the default rule at issue is "stickier" than I imagine. In any event, I thank Professor Peukert for having stimulated thought on an interesting topic--one that is reminiscent in some ways of the current debate in the U.S. over whether a contract to pay post-termination royalties should be unenforceable per se (see most recent blog post here). Another related topic, which I discuss briefly in my book, is whether licensees should be able to contract away their right not to challenge patent validity and infringement. I used to think the answer was yes, but I am somewhat more skeptical now of my earlier position because of the potential effect on third parties if the party with the greatest incentive to challenge validity bargains away that right. Finally, to the extent that a real-world licensee bears a risk that an infringer does not--that she will have to pay for the prejudgment use of an IP right that is subsequently invalidated--reasonable royalty damages should, in this regard, be appropriately higher than real-world licensing fees (as is sometimes the case in Germany and France, for this very reason; see also my paper with Norman Siebrasse, where we discuss the relevance of risk shifting to the calculation of reasonable royalties.)
Finally, since I mentioned the endowment effect above, I should note the following quote from (arguably) Bayreuth's third-most famous one-time resident (after Wagner and Liszt), the nineteenth century writer Jean Paul, which I noticed in the hotel restaurant (which had a string of his quotes above the breakfast bar):
Der Besitz macht uns nicht halb so glücklich, wie uns der Verlust unglücklich macht.
I'd translate this as "Ownership makes us not half so happy, as loss makes us unhappy." A better description of the endowment effect would be hard to come by. So it appears that Jean Paul anticipated the behavioral law and economists by at least a century and a half!