John Jarosz and Michael Chapman have published an article titled The Hypothetical Negotiation and Reasonable Royalty Damages in 16 Stanford Technology Law Review 769 (2013), available here. I read and commented on an earlier version in draft. Here is the abstract:
Reasonable royalty damages are the predominant form of relief awarded in patent infringement cases and, of late, have been a lightning rod for assertions that the patent protection system is out of control. The primary tool used to assess reasonable royalty damages is the hypothetical negotiation construct arising from the seminal Georgia-Pacific Corp. v. United States Plywood Corp. decision in 1970. The construct provides that a reasonable royalty should be determined by hypothesizing an imaginary negotiation between a patent holder and an infringer over use of a patented invention at the time of first infringement. This Article examines the wisdom of the historically heavy reliance upon the construct. We question whether this construct is likely to achieve the ultimate goal of reasonable royalty damages–namely, to provide the patent holder with fair and adequate compensation for the unauthorized use of a patented invention. We find that the foundation for the construct is tenuous and that the use of the hypothetical negotiation construct introduces unnecessary and unproductive questions and conflict into the determination of reasonable royalty damages. We propose that the determination of reasonable royalty damages be based on a direct and objective assessment of a patent’s (1) incremental benefits, (2) licensing comparables, and (3) design-around costs. We propose a balancing and weighing of the results of these different approaches without the introduction of artificial bargaining drama, guided by the objective of ensuring fair patent holder compensation in light of the infringement at issue.
Personally, I agree with portions of the authors' analysis, disagree with other portions. I still favor use of the hypothetical ex ante negotiation construct, for example, and would tend to disagree with what I interpret as the authors' preference for using ex post data (see e.g., article pp. 812 et seq,, 814-15 n.212). But it's a very stimulating article, with a good deal of interesting history and analysis, and I certainly recommend taking a look at it.