H. Holden Thorp, editor in chief at Science magazine and formerly chancellor at Washington University and before that the University of North Carolina, published an editorial in Science, “An opportunity to improve innovation” that provides insight on the UW FAST start scheme. The thrust of the editorial is that Bayh-Dole has been successful, won’t be repealed or amended, and so the new administration should change policy in some unspecified way so that university research does not depend “on the marketplace to spur the kind of research that benefits society–a stated rationale for passing the act.” Thorp’s editorial is garbled in places, but it might best be viewed as a plea for help from a former university leader. We will look first at Thorp’s main argument about technology transfer and then look at what Thorp has to say about FAST start schemes.
I will be critical of Dr. Thorp’s premises but I am sympathetic to the core point that Thorp makes–that there should be some public alternative to the patent-licensing-for-commercial-development-profit-seeking model.
My point in response is that universities don’t need any change in federal law or policy to pursue public alternatives. They need administrators with brains and spines, however, and that’s proven to be more difficult to come by. Instead, administrators compensate for a lack of ability by changing organizational structures, rebranding tired old licensing operations with sparkly new names, and creating dopey things like “express” patent licenses for startups so that administrators don’t have to think and don’t have to have any capability to be responsive to the particulars of any given startup.
Dr. Thorp treats as fact that Bayh-Dole has been “successful.” But Bayh-Dole has not produced such “economic success and progress”–there’s no evidence for that. Bayh-Dole keeps any such evidence secret, and universities report bogus model output as fact and metrics that do not track either economic success or innovation. Universities think so little of Bayh-Dole that they don’t even report their outcomes with Bayh-Dole subject inventions separately from other inventions–even though Bayh-Dole’s standard patent rights clause requires separate accounting.
Dr. Thorp claims that “spurring” research that benefited the public was a “stated rationale for passing” Bayh-Dole. Dr. Thorp doesn’t bother with who did all that stating, but it’s not in Bayh-Dole. Benefiting the public might be said to be a necessary rationale for all federal legislation, but Dr. Thorp’s implication is that Bayh-Dole was intended (by Congress, or by someone else who at this point is secret and doesn’t matter) to alter the direction of university research rather than to disrupt federal open access to inventions made in research or development judged worthy of public subvention.
There’s nothing to support Dr. Thorp’s assertions. 35 USC 200 lays out Congress’s policy and objectives with regard to Bayh-Dole. There’s nothing there about changing the course of research so that it benefits the public. Bayh-Dole limits patent property rights so that patents on inventions arising in federally supported research are used to promote the utilization of those inventions. Research may be beneficial to the public in all sorts of ways that don’t involve patenting–for instance, discoveries of laws of nature and algorithms (not patentable); collection of specimens and data; testing and validation (or disconfirmation) of theories, hypotheses, and research claims; development of software and databases for public access; demonstrations of prototypes and applications for past discoveries; training of students in research methods, instrumentation, and strategies for discovery.
Even if patenting is involved, that patenting doesn’t require using patents to suppress the use of a claimed invention. A patent may be used to document an advance, announce it formally in the national repository of the patent system, raise the standard of the “prior art” for future patenting, preclude others obtaining patents on the same subject matter through failure to be the first to file a patent application, contribute to a standard, and counter the enforcement of patents that might be used to block the practice of the invention or its development by a broader community of users. Even if someone is determined to suppress use by others of a claimed invention with respect to a specific area of application and a specific product or method within that area, that doesn’t require the suppression of all other areas of application and anything that might be made in those areas. This cascade of benefits of using a patent non-exclusively is baked into 35 USC 200–use the patent system to promote free competition–it’s darned hard to use the patent system to promote free competition if all competing (and non-competing) uses of a claimed invention are suppressed by the threat of infringement lawsuits.
In 35 USC 200, the law aims to reserve rights in patents in order to protect the public from nonuse and unreasonable use of inventions arising in federally supported work. The concern here is protecting the public, not benefiting the public (other than in the sense that the public benefits from protection).
In 35 USC 201(f), the definition of “practical application” involves use of inventions with benefits available to the public on reasonable terms. But that’s about how patented inventions get deployed after the research has been done. There’s nothing in Bayh-Dole about university faculty choice of research topics or the direction of that research. And all that the definition of practical application gets used for in Bayh-Dole is to permit on condition for march-in (35 USC 203(a)(1))–where the government’s remedy for nonuse or failure to timely develop, or deployment with unreasonable terms is to compel licensing to others, restoring as it were “free competition.” How federally supported research benefits the public and how the use of the patent system to provide benefits to the public on reasonable terms are two distinct things.
Thorp tries to differentiate between research that benefits the public and research that enters the “marketplace” to be deployed. It’s an odd distinction. There can be, obviously, both–a finding that the public can use directly (including other researchers, companies, and technical professionals) and which can also be incorporated into various mass produced products for use by industry or government or “consumers.”
We could, as a matter of technology transfer, discuss how research inventions get used without first having to be sequestered by patents in the hope that they will become profitable, mass-produced commercial products. Preventing access to inventions may be the single greatest way to disrupt technology transfer activities and chase people off to other alternatives.
Internet communications protocols, for instance, became standards rather than products exclusively controlled by any one company. Bayh-Dole’s stated goal is not “commercialization” before all else, but rather use of the patent system only where no use at all of an invention is possible without substantial further effort at “development,” and there is no other prospect for engaging anyone–foundations, government agencies, consortia of companies, wealthy donors, crowd-sourced efforts–to do that work. This “if all else fails” mindset is behind the use of the word “necessary” in Bayh-Dole’s federal licensing requirements in 35 USC 209, and those “necessary” stipulations arise from 35 USC 200’s restriction of the use of the patent system to promote utilization of inventions, not the suppression of their use.
For any given invention, there are often multiple transfer pathways–involving lab to lab interactions; from a university lab to industry labs, and from labs migration into company work (research, development, production). If this lab-based transfer activity grows, then perhaps one might see specialty manufacture and distribution start up to support access by both labs and companies that don’t have the resources or capability to DIY their own implementations. Beyond that, one might see commercial mass production of product based on whatever platform of technology that emerges from the lab to lab and lab to company and company to labs activities. Of course, one can also license patents exclusivel to speculative investors who then set out to raise the money to create their own labs into which the invention is transferred and developed, or more often developed a bit so that the whole operation can be sold to new speculative investors or to a big company.
This account of multiple pathways for IP-based technology transfer, however, is alien to the thinking of university patent administrators, who imagine a world in which research takes place in isolation, companies exist only to mass produce products based on inventions in which they hold exclusive patent rights, and the purpose of technology transfer is to find one such company for each invention, and do an exclusive license that cuts out everyone else, with the expectation that doing so is the only way inventions will ever benefit the public. The purpose of such exclusive licenses is not to find effective pathways by which research inventions might get used, but rather to make money for the university–first by recovering patenting costs, second by trading on the speculative value of a given patent in suppressing (or taxing) company use of the claimed invention (in any form, not just the form an invention might take as part of any given single product), and third by gaining income from other company assets acquired in connection with granting a patent license, such as equity in the licensee or a big upfront fee–neither of which has anything to do with the licensee’s actual use of an invention to benefit the public.
Perhaps you start to see how a one-license-for-all-startups works against the discovery of technology transfer pathways for inventions and other research-generated findings. The business of extracting payments in exchange for patent licenses is almost but not entirely disconnected from how new technology moves from one group that knows it to other groups that don’t know it but would use it if they did know it. For a university spinout started by the same people who have done the inventing, there isn’t any technology transfer–except as a university intrudes, takes rights, and then fumbles around trying to figure out how to return those rights (to the startup company, which is just the inventors again) and make as much money as it can while disclaiming any liability for being involved. The bureaucratic idea is to preserve everything–taking ownership, paying for patenting even if the company doesn’t want it, and then offering the rights back with lots of conditions and demands for payment. The bureaucratic problem isn’t that this whole process is unnecessary, cumbersome, and strained (which it is)–the problem rather becomes how to make the process appear to be somehow standard and therefore “automatically” accepted.
Bureaucrats never admit that something they have done isn’t working. As Clark Kerr put it, for universities, change comes from the outside. For a bureaucrat, everything is great and could get greater. Thus, when Dr. Thorp pleads for help from the White House while claiming that everything is going grand, we may conclude that things are terrible, university administrators are helpless to make changes, and whatever improvements that will happen will have to come from the outside.