Liivak on Nominal Damages

Oskar Liivak has posted a new paper on ssrn, titled When Nominal Is Reasonable:  Damages for the Unpracticed Patent.  Here is a link to the paper, and here is the abstract:

To obtain a substantial patent damage award via reasonable royalties, a patentee need not commercialize the patented invention; infringement is all that is needed. This surely incentivizes patenting but it disincentivizes innovation. Why commercialize yourself? The law allows you to wait for others to take the risks, and then you emerge later to lay claim to “in no event less than a reasonable” fraction of other people’s successes. Today, it is rational to be a patent troll rather than an innovator.

This troll enabling interpretation of reasonable royalties is wrong as a matter of patent policy and, surprisingly, it is also wrong as a matter of patent history and statutory interpretation. The creation of reasonable royalties by the courts in the nineteenth century did mark a significant change to patent damages but it was nowhere near as sweeping as today’s interpretation would suggest. Up to the mid-1800s, the existing routes to patent damages were stringent, available only to patentees who had already commercialized their patented invention. Courts developed reasonable royalties for budding innovators who were laying the groundwork for innovation but who could not yet satisfy the existing strict routes to patent damages. Those cases never extended reasonable royalties to those who simply sat on their patents waiting to extract payment from others. Starting in the 1970s, reasonable royalties came unmoored from that foundation. Infringement alone, without any commercialization efforts, now creates a presumption of compensable harm and the near guarantee of a substantial payout. Today’s view of reasonable royalties is unsupported by patent history and it in fact sits in tension if not outright conflict with binding Supreme Court cases. Properly understood, some efforts to commercialize are still a necessary element for substantial reasonable royalties. As a result, nominal damages are reasonable for infringement of an unpracticed patent.

It's an interesting paper, and I was glad to see that this draft was available after having spoken with Professor Liivak about his thesis at a conference this past spring.  Professor Liivak's principal argument, as I understand it, is that NPEs that seek only to license their patents after having found someone who is allegedly infringing them, rather than engaging in ex ante technology transfer, should be awarded only nominal (that is, zero or near-zero) damages against unintentional infringers.  He makes both doctrinal and policy arguments in support of his thesis.  Stated in my own terms, the policy argument is that patents are intended to promote invention, disclosure, and innovation.  Where the alleged infringer is a more efficient make or user of the patented invention, in comparison with the patent owner--which must be the case where the patent owner is not a practicing entity--static efficiency would improve if the infringer went ahead and made and used the invention.  The only reason to provide the patent owner with any remedy is that doing so promotes the policy of promoting invention, disclosure, and innovation, that is, it promotes dynamic efficiency.  In Professor Liivak's view, however, the patent system's role in promoting invention is uncertain, and where the infringer is the one doing the innovating--that is, where it's the one that is commercializing the technology, without any help from the patent owner--affording the patent owner a remedy doesn't promote innovation either.  (Perhaps one could extend the argument to incorporate Kitch's prospect theory as well, by arguing that where the patent owner is not seeking out prospects ex ante but rather only seeking to license anyone and everyone ex post, awarding a remedy would not serve the prospecting function either.  I suspect that prospect theorists wouldn't approve of Professor Liivak's argument nonetheless, on the ground that it inhibits the patent owner's ability to decide whom to license and on what terms.)  


That said, I'm not altogether persuaded, for several reasons.

First, as a doctrinal matter, I'm not sure I'd give as much weight as Professor Liivak does to the century-old precedents that he argues are consistent with his thesis but which never quite come out and state that thesis it directly.  I think it's fair to assume that notions of what constitutes harm to the patent owner can evolve over time and are not necessarily bound by the courts' understanding as of the year 1900 or earlier.  Moreover, prior to 1946 U.S. law permitted the patent owner the option of recovering the defendant's profits (a remedy that remains in the U.S. for design patent infringement, and that remains available in most other countries today for the infringement of all types of patents).  Perhaps the assumption was that if you couldn't prove some harm other than the loss of hypothetical licensing revenue, you at least had the option of recovering the defendant's profit.  Once that premise has gone away, however, how confident can we be that the old cases  continue to provide the applicable rule of decision for calculating damages?

Second, another rationale for awarding (non-nominal) reasonable royalties based on the willing licensor-willing licensee framework is that the defendant benefits from the unauthorized use of the patent and thus would reap a certain unjust reward if it could do so with impunity.  This unjust enrichment theory is not quite consistent with the characterization (in section 284 of the Patent Act) of reasonable royalties as compensation (as opposed to restitution), but perhaps it's not entirely irrelevant either.  (For further discussion of the restitution concept as applied to reasonable royalties in U.S. law, see my paper Reining in Patent Remedies:  Three (Increasingly Immodest) Proposals, 30 Santa Clara High Tech. L.J. 1, 25-26 & n.94 (2013).  Note also that the idea that reasonable royalties are based on some sort of unjust enrichment principle is present in some other countries' laws, including the U.K. where as I understand it the term "user principle" refers to the concept of rewarding the patentee based on the value of a license to the infringer.)  It would provide a basis for the current practice even if a stricter definition of "harm" as a prerequisite to awarding compensatory damages would not.  (It might be interesting, though, for Professor Liivak to consider the somewhat parallel discussion of whether copyright owners suffer harm if all they are faced with is a loss of potential licensing revenue. This comes up in some of the literature on fair use by, for example, Christina Bohannan.) 

Third, although I understand Professor Liivak's theoretical argument that allowing the nonmanufacturing, nonlicensing patent owner to recover reasonable royalties penalizes inadvertent infringers who benefit society by innovating (commercializing), it seems to me that his proposal would suffer some practical problems, even if it were desirable in theory.  Could patent owners avoid the rule by engaging in token use or licensing?  What would be the cost of determining whether a defendant was an inadvertent infringer, whether the patentee engaged in good faith efforts to license its patent ex ante, and so on?  Notwithstanding Professor Liivak's skepticism on the point, might there be a reduction in the incentive to invent, even if there were a corresponding increase in the incentive to innovate?   Would there be a reduction in the incentive to try to find and license patents prior to launching a product? 

Relatedly, although Professor Liivak argues that his proposal does not amount to a working requirement (see p.5), it seems functionally very similar to one, if the patentee cannot recover damages (or, I would assume, an injunction) absent an effort to commercialize its technology.  (My recollection is that working requirements, where they exist, arguably can be satisfied by manufacturing, importing, or licensing.)  True, the patent is not invalidated as it would be under a working requirement, but a patent that is valid but unenforceable seems functionally no different from one that is invalid.  (An aside--if the patentee couldn't get damages, could it get an injunction on the theory that the infringement threatens it with harm for which there is no adequate remedy at law?  That would turn things on their head, it seems to me, since the nonpracticing patent owner armed with an injunction could demand very substantial postjudgment royalties as a condition to continuing the use of the patent.  In the old cases Professor Liivak discusses, was there any question whether a court of equity would grant the noncommercializing patentee an injunction?  See, e.g., the old Continental Paper Bag case, which the Supreme Court rather unpersuasively distinguishes in eBay.)  It also might seem to import an "intent" requirement into the definition of infringement as well, to the extent an inadvertent infringer would escape liability altogether in cases in which the patentee was making no effort to commercialize its technology itself.  Whether a working requirement or an intent requirement would be a good idea or not is debatable (I tend to think not, but it's debatable), but it seems pretty clear to me that U.S. law historically has eschewed these requirements.

In any event, though I disagree with Professor Liivak's to some degree for the reasons above, his paper provides an interesting and thought-provoking read, and should engender much discussion.