Monday, September 30, 2024

Patent Damages and Pessimism

Dennis Crouch’s recent blog post on the Federal Circuit’s decision to rehear EcoFactor v. Google en banc states that “I begin this post with pessimism about the fundamental inquiry into patent damages.  In most cases, I see patent damages as inherently speculative.  The law’s hypothetical negotiation to reach a reasonable royalty is a counter-factual endeavor set so far apart from reality that is results have little meaning.”  I’m not quite as pessimistic as Dennis, but he certainly has a point.  As I see it, from an economic standpoint an award of reasonable royalties ought to result in the patentee recovering some portion of the value the defendant derives from the use of the patented technology in comparison with the next-best alternative.  That’s a pretty solid theory of what reasonable royalties should do, in my opinion, but carrying out the theory in the real world can be extremely difficult.  What is the portion the patentee should recover—is half a reasonable heuristic, or should we inquire into the actual parties’ bargaining power?  How do we know what the next-best alternative is? What if the next-best alternative patented?   As a factual matter, how do we quantify the benefit the implementer derives?  Must we rely on indirect evidence such as comparable licenses, despite their many problems?  And even if we can resolve these issues, there remains the fundamental problem that the more we invest in achieving an accurate result (one that, we believe, conforms closely to the theoretical ideal), the greater the adjudication and administrative costs.  So what is the optimal tradeoff between administrability and accuracy?  How would we know when we achieve it?

One thought that occurred to me in reading Dennis’ post is an argument presented by Eric Posner in his 2002 book Law and Social Norms (and in his earlier article A Theory of Contract Law Under Radical Judicial Error), to the effect that contract law formalities serve a signaling function.  The underlying premise is that courts aren’t very good at determining contract damages.  Accordingly, people enter into formal contracts not so much to preserve the option of having a court accurately determine damages in the event of a dispute, but rather as something of a commitment strategy:  complying with contract formalities is a way of signaling that the parties will work things out for themselves, with knowledge that if they fail—if one party takes opportunistic advantage of the other, over the course of the relationship—the other party has the option of invoking the judicial system, and will do so (at cost to itself) to preserve its reputation for not being a pushover.  The point, in other words, is to discourage opportunism, in recognition of the courts’ ability to discern opportunism when they see it, but maybe not to do much else very well. 

It's a clever theory, and I’ve wondered whether it might have some applicability to patent law too; but, ultimately,  I’m not sure it does.  In patent infringement disputes, sometimes the parties have the option of negotiating a license before the implementer starts using the technology, but other times they don’t; evidence suggests that inadvertent infringement is very common.  When there is such an opportunity, however, one could imagine that the uncertainty of going to court might encourage both parties to act reasonably, since either could wind up losing big:  implementers face the risk of huge damages judgments, and patentees of invalidation, among other things.  As for which side has the advantage, though, it really depends a lot on the specifics.  (Patentees will say that, in general, implementers have most of the advantages, and implementers will say the opposite.)  Still, one might think that the law of patent remedies should be designed, where possible, to induce the parties to work things out for themselves as much as possible (which is, by the way, the perspective of the German courts, as I understand it, in their resistance to recognizing many exceptions to awarding the prevailing patentee injunctive relief).  One problem, though, is that where the implementer only becomes aware of the patent ex post, or has no choice but to use the technology because it is standard-essential, the patentee may be in a position to extract a royalty that exceeds the ex ante value of the invention to the implementer (the hold-up problem).  Another is that the invalidation of invalid patents is a public good, and so licenses (settlements) that avoid invalidation or delay the implementer’s entry into the market are not necessarily socially optimal.  (I’ve been thinking about this a lot recently in connection with some writing I’m doing on pay-for-delay settlements of pharmaceutical patent disputes, and in reading a recent paper on this topic by Michael Carrier and Edward Bank.)  Relatedly, there is a social interest in ensuring that royalties, whether negotiated or adjudicated, do reflect the value of the technology and not other, irrelevant factors.  Preserving the patent incentive scheme means not shortchanging patent owners, but also not requiring implementers (and, by extension, the public) to pay more than the technology contributes to the state of the art.

So, maybe there is room for pessimism.  An ideal system of patent remedies may be unattainable—and even if we did attain it, we wouldn’t know it when we saw it.  Perhaps, then, we’re back to square one:  there is a social interest in accuracy, and a social interest in administrability, and we have to figure out some way to accommodate the two, but without ever really being sure we’ve achieved the right balance.  For my part, I’m inclined to think that, when in doubt, accuracy should be the more important consideration; and I’m hopeful that the EcoFactor case will move us a bit more in that direction.  But there will always be considerable, perhaps insurmountable, degree of uncertainty in trying to craft the optimal rules for patent remedies.  It might be best if we first acknowledged that.  

Pictured above:  the great pessimist, Arthur Schopenhauer.

Postscript:  For another insightful take on the inherent difficulties courts face in awarding patent damages, see John Golden's 2010 article Principle for Patent Remedies  (and, for what it's worth, my short response here).

Wednesday, September 25, 2024

Federal Circuit to Rehear En Banc EcoFactor v. Google

Here is the order, and here is a link to a just-posted notice about it on Law360.  From the order:

The parties are requested to file new briefs, which shall be limited to addressing the district court’s adherence to Federal Rule of Evidence 702 and Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), in its allowance of testimony from Eco-Factor’s damages expert assigning a per-unit royalty rate to the three licenses in evidence in this case.

The original panel decision came down in June, and I blogged about it here.  Various amici filed briefs urging the rehearing, as reported by Dennis Crouch a few weeks ago in an informative post on Patently-O, with links to the briefs (see also his earlier post on Google's petition for rehearing en banc).  I confess that I have no clear idea where this is headed, but if the court succeeds in providing some greater clarity on the standards for admissible testimony on damages, and on the use of comparables, that could be helpful.  I may have more to say about this as the case progresses.   

Update:  here is a link to a follow-up Law360 article that quotes, among others, me, and here is a link to an article on Bloomberg.