Bifurcation of Liability and Damages

Professor Siebrasse’s comment to my post of May 14 raises the topic of bifurcation, which in my view deserves its own post, so here goes.

As discussed in my book, bifurcation of liability and damages is common in many countries, including Germany, the U.K., Canada, and Japan.  It’s less common in the United States, although my understanding is that it has become common practice in some venues such as the District of Delaware.  On the merits, bifurcation has its advantages and disadvantages.  The principal advantage is that if you try liability first and the result is a judgment for the defendant, you don’t have to waste time trying damages.  One disadvantage, at least in the United States, is that you may have to convene a second jury to consider damages.  This was one drawback raised by one of the federal judges at a conference on patent remedies at the University of San Diego Law School that I attended earlier this year.  Of course, in other countries where there are no jury trials in patent cases (i.e., pretty much all other countries), this wouldn’t be an issue.

Another disadvantage, I suppose, of bifurcating is that if the patentee has a weak or nonexistent case on damages, it may be better to flesh that out at an early stage of the proceedings.   Judge Posner’s decision to dismiss both parties’ claims in Apple v. Motorola last June was premised on precisely this rationale:  if the patentee isn’t entitled to an injunction under the U.S. eBay decision, but it doesn’t come forward with admissible evidence quantifying its damages, there is no right to proceed to trial for a judgment that can result in only nominal damages.  (Of course, it remains to be seen whether the Federal Circuit will agree with Judge Posner’s evidentiary and other rulings in the case).   In a system in which bifurcation is the norm, a dismissal of the liability case on the ground that the plaintiff cannot quantify its damages almost certainly wouldn’t happen.  (Indeed, in countries where bifurcation is the norm, if the patentee wins on liability the parties typically settle before a damages trial ever begins—which is probably one reason there tends to be less case law on damages than in the U.S.) 

There are several other reasons why the Posner outcome wouldn’t happen anywhere else (which is not—I hasten to add—to imply that it was wrongly decided under U.S. law).  As I have pointed out previously, most other countries do not follow eBay and thus still routinely award permanent injunctions (though perhaps this will change).  Also, courts in some countries may have greater leeway to estimate damages on their own (e.g., the German concept of nach freier Überzeugung; see Markus Schönknecht, Determination of Patent Damages in Germany, 43 IIC 309, 311–12 (2012)); and plaintiffs often have the option of recovering the defendant’s profits attributable to the infringement as an alternative to lost profits or a reasonable royalty.  The U.S. eliminated this remedy for utility patent infringement in 1946, retaining it only with respect to design patents (where it came up in last year’s Apple v. Samsung judgment).  The inability to quantify one’s own damages wouldn’t matter so much in a system that allows you to recover the defendant’s profits attributable to the infringement, at least where the defendant bears the burden of proving what costs should be deducted from its overall profit.        

            

Anyway, to get to Professor Siebrasse’s comment: 

“In Power Integrations, Inc. v. Fairchild Semiconductor Intern., Inc. (Fed Cir 26 Mar 2013), the district court bifurcated infringement and damages issues from those relating to validity of the asserted patents. The trial relating to damages was held first (see 585 F.Supp.2d 568). In that case, should damages be assessed on the basis that the patent is valid? The district court indicated that this kind of bifurcation is “routine” 2006 WL 2435089.”

Personally, I think it’s odd to hold the damages trial first; but if this is what the court decides to do, I suppose it should still estimate damages based on the assumption that the patent is valid and infringed, since a damages judgment ultimately will only be entered only if that assumption turns out to be correct.  As for the routineness of bifurcation in the U.S., see above; it may be routine in some places but not generally.

Power Integrations raises some other interesting issues that I will address in future posts.  One of them relates to the U.S. rules on patent marking and how they relate to damages awards.  On this issue (surprise, surprise) U.S. law tends to be much more complicated than the law in any other country.