The angel is in the big picture: a response to Lemley.

• Author: Vermont, Samson
• Position: Mark A. Lemley, Michigan Law Review, vol. 105, p. 1525

TABLE OF CONTENTS I. LEMLEY'S THREE RESERVATIONS II. LEMLEY'S FOUR ALTERNATIVE REFORMS A. Exempt Reinventors from Willfulness Damages B. Treat Reinvention as Evidence of Obviousness C. Exempt Reinventors from Injunction D. Adopt Prior User Rights CONCLUSION An invention within close reach of multiple inventors differs from an invention within distant reach of a lone inventor. The differences between these two archetypes of invention--"reinventables" and "singletons"--remain unexploited under current U.S. law. Should we reform the law to exploit the differences? Mark Lemley (1) and I agree that we should. To date, those economists who have closely examined the issue concur. (2)

What are the differences between reinventables and singletons? First, reinventables can be brought into existence with incentives of lower magnitude. This suggests that we can obtain reinventables at a lower price than we currently pay--i.e., with less monopoly loss than we incur today. Second, reinventables generate disproportionately more haste and redundancy, as the rival inventors race and duplicate each other's efforts. This suggests that we already pay more, in rent dissipation and lost opportunity, for reinventables than for singletons (holding all other things equal). Third, reinventables generate disproportionately more litigation as the race winners, or the "trolls" to whom the winners transfer patents, eat up time and resources suing the inventors who finished a close second or third. (3) This suggests that we already pay more in administrative costs for reinventables than for singletons.

The angel is in the big picture in that there is consensus among those who have closely examined the issue that we should reform the law to exploit these differences. The devil is in the details of just how to reform it. Naturally, professional economists have elided the law-related details, focusing instead on their models--models that show an increase in social welfare if the law is reformed so that reinventables hold out the prospect of shared duopoly. (4)

Lemley and I, in contrast, take a stab at some of the details of how legal reform could take shape. My proposal is that we regard an independent inventor ("reinventor") as exempt from the first inventor's patent, provided that the reinventor completed the invention before receiving notice that the first inventor had already completed it. (5) Lemley expresses three reservations about my proposed reinvention defense, and then offers four alternative proposals.

1. LEMLEY'S THREE RESERVATIONS

   Lemley's first reservation is a general one: we are playing with fire. History shows that the most important inventions are often invented by multiple inventors at roughly the same time. This implies that the reinvention defense will have a disproportionately greater effect on the most important inventions, which implies that the reinvention defense might turn out to be penny wise and pound foolish.

   Point well taken. We are indeed tinkering near the bull's-eye of invention. The expected social cost of delaying an important invention--one as important as, say, the polio vaccine--may be so high as to justify the insurance premium we pay in the form of the social costs attributable to treating reinventables and singletons as if they were the same. I concede, therefore, that neither the courts nor Congress should adopt the reinvention defense tomorrow. It needs to be vetted for a few years. Ideally, the vetting process would provide some answers to the following questions about magnitudes: How much monopoly loss is attributable to reinventables under current law? How much hasty and redundant R&D is there? What percentage of important inventions are invented by multiple inventors at about the same time? How often would reinvention lead to Cournot duopoly (under which prices moderately exceed those under free competition) rather than Bertrand duopoly (under which prices are driven down to those under free competition)? Most importantly, to what extent are reinventables more a function of forces exogenous to the patent system than singletons are?

   Lemley's second reservation addresses my claim that the fact of reinvention--that multiple inventors converged on the same invention at about the same time--is evidence that a moderately smaller incentive would have sufficed to bring forth the invention in a timely manner. He argues that the truth of my claim depends on the type of invention in question. Some inventions (e.g., pharmaceuticals) are cheap to invent but expensive to test for safety and efficacy. These inventions may require the extra incentive provided by our current winner-take-all patent system.

   Again, point well taken. One would expect underproduction of drugs if inventors could invent them cheaply and then free-fide on the costly efforts of other inventors to test the drugs for safety and efficacy. This problem, however, is fairly confined to drugs, and the FDA already deals with it by granting five years of market exclusivity to a new drug...