Boettiger and Bennett look at Bayh-Dole after 25 years and discuss how things ought to change. To set up their discussion, they first characterize Bayh-Dole as having “shifted the incentive structure” for patents. Parts 1 and 2 of this series discuss how Bayh-Dole didn’t so much shift a structure but rather allowed patent brokers to take more federally supported research into an existing structure–and then change that structure to favor institutional profit-seeking from exclusive patent licensing (and ascribe the change to Bayh-Dole).
Imagine a federal law that requires all locks to be removed from homes. That law does not endorse robbery, but what do you think will happen? Bayh-Dole removes the authority of federal agencies to impose restrictions that limited private patent monopolies from interfering with the dissemination of discoveries and inventions made to advance scientific knowledge or the public welfare. Bayh-Dole did not shift the federal agency incentive structure. It allowed incentives that had nothing to do with the federal purposes of the funding in the first place. Bayh-Dole made the implied purpose of all funding–Bayh-Dole is arbitrary–to promote the creation of private patent monopolies. However those monopolies operate is necessarily, by decree, to be in the public interest. Yes, there were vestigial “march-in procedures” included with Bayh-Dole, but they were designed not to work–and in the implementation phase, university patent brokers made sure they would not work, and bragged about it afterward.
Boettiger and Bennett describe the basics of Bayh-Dole:
The Act also requires the institution to establish patent policies for its employees, to actively seek patent protection and to encourage the development of their inventions.
These descriptions are wrong, each in its own way. Nowhere in Bayh-Dole is there any requirement for institutional patent policies. There was such a requirement in the failed IPA program–before an institution could get a master agreement, it had to have a patent policy and practices that indicated an IPA was appropriate. But Bayh-Dole does not do this. Bayh-Dole changes federal patent law to create a new category of invention and then stipulates how federal agencies may contract for the disposition of these “subject” inventions. It is that standard patent rights clause that carries whatever policy requirements are necessary to contracting universities (and other nonprofits, and small businesses). Bayh-Dole in this way is self-implementing. Federal contracts supersede state contracts. The Bayh-Dole standard patent rights clause therefore supersedes any university policy or employment contract or patent agreement on the same subject matter.
One might say, Bayh-Dole’s standard patent rights clause demands that any university decision regarding a subject invention must come after the invention has been made and reported. The IPA worked this way. The difference was that the IPA required that a university have a patent policy that required assignment of inventions to the university only if the university had made the decision to pursue patenting and undertake the obligations for licensing (soon and reasonably and for limited terms of exclusivity) in the IPA. It was review and commit, then acquire and comply. Not acquire and review and maybe commit and maybe comply.
If Bayh-Dole were enforced on its terms, and the standard patent rights clause was enforced on its terms, then inventors would own their subject inventions and would have a choice–(i) they could assign to the university that hosted their research (and only if the university agreed to accept that assignment, complied with the standard patent rights clause, and met whatever other conditions the inventors required–diligence, say, and regular communication, and a right to regain ownership if the university or a licensee failed to perform); or (ii) they could assign to any other invention management agent, provided that agent accepted the standard patent rights clause under which the invention was made (for nonprofits, that means restrictions on assignment and use of royalties); or they could negotiate with the federal funding agency and (iii) assign their rights to the federal government (to go to the public domain or a patent commons where they would have ready access) or (iv) retain ownership and have the least restrictions possible under Bayh-Dole–a better deal than even small business contractors and much better than universities and other nonprofits (who have the most restrictions–why did Bayh-Dole imagine that the nonprofits needed the most regulation?).
In each of these four cases, inventors have the choice. There’s nothing in Bayh-Dole that disturbs the ownership of inventions. The Supreme Court in Stanford v Roche was clear on that point, despite NIST’s idea that it can change the law by “clarifying” a part of the standard patent rights clause so that rather than being within the authority of the statute, the clause will require an action (requiring that inventors assign) that is not authorized by Bayh-Dole. Inventor-loathing university patent administrators have worked for over three decades to suppress this inventor choice set made available by Bayh-Dole. But the choice set is there. In the law. It’s a separate discussion whether it’s a good thing for inventors to have such choices, or whether inventors should be constrained at all. But it’s not much of a discussion whether the choices are there, in the law and the standard patent rights clause. It’s just a matter of university patent administrators who fail to read, or who choose to mislead.
Contrary to our authors’ assertion, Bayh-Dole also does not require universities to “to actively seek patent protection and to encourage the development of their inventions.” Follow close now. First Bayh-Dole doesn’t apply to universities. It applies to federal patent law and to agencies. It changes federal patent law for all subject inventions. It requires federal agencies to do things–to use a standard patent rights clause except in exceptional cases. The standard patent rights clause in turn sets up a conditional: IF a university acquires a subject invention and IF a university chooses to keep that ownership THEN it must file a patent application. The only reason to acquire ownership of a subject invention under the standard patent rights clause, then, is to seek a patent.
But from there, the standard patent rights clause goes dark with regard to what happens next. Our authors claim that Bayh-Dole requires a university “to encourage the development” of subject inventions. Well, not true. But again, let’s work carefully to see how it is not true. There’s nothing in the standard patent rights clause that speaks directly to “development” of inventions. There is a requirement to report on the status of inventions, including date of first commercial sale or use. But there’s nothing to indicate that development is required. There is a procedure of “march-in” if an invention is not made “reasonably available” but march-in, too, does not require “development.” Requirements for preferring small businesses and requiring U.S. manufacturing if a license for exclusive use or sale is granted also don’t speak to “development.” One can license without development (an invention may be simply used, no product needed).
Bayh-Dole’s statement of policy and objective, which establishes the properties of patents on subject inventions, the contracting requirements for federal agencies, and the administration of contracting, specifies “to use the patent system to promote the utilization of inventions arising from federally supported research or development.”
That is: “promote the use” of inventions. “Development” is one of the purposes of the federal funding, not a specific goal for inventions after they have been made. “Development” has to do with technology, not inventions becoming “products” or “commodity products.” Here’s one typical definition of “development”:
Systematic application of knowledge toward the production of useful materials, devices, and systems or methods, including design, development, and improvement of prototypes and new processes to meet specific requirements.
Improve prototypes and develop new methods to meet specific requirements. No mention of “patentable inventions” being the subject of such work. Could be, but could be most anything else, too. The focus in Bayh-Dole is on use, not development. The definition that carries the load in Bayh-Dole is “practical application”–use of subject inventions with public benefits available on reasonable terms. Nothing there about “development.”
It may seem I’m picking on the choice of words. But “development” is a term of art. It has been chosen as a basic characterization of the law. And it’s not in the law. Further, what word would one use otherwise? “Use” (or, “utilization”) perhaps. But if we say “to encourage the use” of subject inventions rather than “to encourage the development” of subject inventions, we head in rather different directions. Development suggests refinements to meet specific requirements. Use means, well, that stuff gets used, regardless of whether it is “developed.” One may use a research tool to do more research, for instance. Or one may apply an inventive soil amendment just as it has been developed, without further development. To argue that Bayh-Dole requires “development” of inventions narrows the law, making it appear that Bayh-Dole endorses university “commercialization” programs. Certainly such programs might meet the expectation of “use” of inventions–a developed invention might also be a used invention. But development stops short of use.
Look at it: development means producing something that is useful. Utilization means that something is used, and necessarily then is useful. Same for practical application–that use can be “established”–with public benefits, reasonable terms. Development is an activity that might produce something useful. A patentable invention is determined to be “useful” as a primary condition. A patented invention has already been judged useful. Development of such an invention then would be an additional step, to meet particular requirements, but not a necessary step. To argue that Bayh-Dole requires development, then, is simply wrong. It’s not a poor choice of words among a bunch of synonyms. It’s a wrong choice of term relative to what the law states. Is it sloppy? Intentional misrepresentation? Perhaps. My sense, though, is that it is just convenient wording–choose wording that then sets up the discussion. In that sense, the wording doesn’t really matter–it’s not meant to inform regarding the law, but to make the discussion to follow appear relevant. In that sense, then, the technical term is akin to “bullshit”–the wording has a performative purpose but not a regard for the truth. Or, one might say, the authors are pretty distant from the law that they claim to have ideas regarding improvement.
One final note on this part of the discussion–Bayh-Dole uses the verb “promote” rather than “encourage.” These surely are similar senses:
promote [origin: forward + move]:
“further the progress of (something, especially a cause, venture, or aim); support or actively encourage”
encourage [origin: to give courage to]:
“help or stimulate (an activity, state, or view) to develop”
“give support and advice to (someone) so that they will do or continue to do something”
There are shared meanings, of course. But “promote” has the sense of moving something–use–forward as a primary action. “Encourage” is remote, to give others “courage” to do something, such as “to use” an invention. Do you see the difference? The law sets a standard: use patents to move use forward. Our authors change this to “encourage others to develop inventions.” Somehow, patent licensing is meant merely to be encouragement. Once a university has “encouraged” private development, its responsibility for invention management is at an end. Anything else has to do with getting money–and that can be had regardless of whether an invention is used or developed or product is sold. That’s how speculative markets work–pay up-front to take a risk, or to acquire an asset so others don’t have it, or for possible use later, or to sell to other speculators.
But Bayh-Dole focuses on use, not on encouragement. The law focuses on more use, available use, established use, not on encouraging others (say, with a monopoly license), to “develop” something. Development is secondary to use. Development is one way to promote use. But it is not the same thing, and not as general. The idea proposed by our authors is that Bayh-Dole requires universities to license inventions for development–that this is enough to satisfy the law, that actual use is not the policy and objective, just encouraging development.
If Bayh-Dole were clearly summarized, the discussion that would follow would be one of how to improve enforcement of the law and of the standard patent rights clause. As it is, Bayh-Dole is a law that’s never been enforced, and a standard patent rights clause that’s never been challenged with regard to practical application. A do WTF you want kind of law. It’s a form of “liberty,” no doubt, but one that has destroyed inventors’ rights and federal agency purposes in favor of institutional control of research results to “encourage the development” of inventions–back to our old friend commodity commercialization before public access or use.
Pingback: Ten Years After 25 Years After Bayh-Dole, Part 2 | Research Enterprise