NIST doesn’t understand Bayh-Dole. Let’s take another look at their mind-numbing FAQ.
Here’s the question NIST asks:
Does an invention need to be reported if it was conceived before an award but reduced to practice as part of the award?
This is an important question, but NIST casts it with a fundamental ambiguity. The point of the FAQ ought to be to call out the ambiguity and show how to better understand circumstances and anticipate outcomes. But no. Instead, NIST gives an answer that is not helpful.
NIST’s question is important because it drives to the heart of the ways in which the federal government is involved in funding research activities. We may distinguish three such purposes: (1) to engage talent to solve problems faced by the federal government–that’s procurement of research services, and the government specifies the problem and receives proposals and statements of capability from contractors with an interest in attempting to solve the problem; (2) to attract talent to work in areas judged by a federal agency to be important for the advancement of public welfare–whether science, technology, agriculture, health, safety, or transportation; (3) to seek out talent and provide funding because what that talent proposes to do is judged to be worth doing, whether that doing is learning how to conduct research or investigating nature or society or producing interesting and useful technology.
In each of these arrangements, one can anticipate that a contractor may well go–we can compete for the offered funding because we have developed some great ideas that will give us an advantage in doing the research. The thrust of the FAQ is to what extent the government may force those great ideas into the scope of Bayh-Dole when those ideas were developed outside of federal funding, even developed to the point of forming a conception, a patentable invention imagined in all of its parts and functions and recognized by its inventor as inventive.
At stake is this bit in Bayh-Dole (35 USC 200(f)):
No funding agreement with a small business firm or nonprofit organization shall contain a provision allowing a Federal agency to require the licensing to third parties of inventions owned by the contractor that are not subject inventions unless such provision has been approved by the head of the agency and a written justification has been signed by the head of the agency.
That is, the standard patent rights clause does not permit a federal agency to appropriate inventions that are not subject inventions–at least the federal agency has no authority to require licensing of those inventions. And since Bayh-Dole deals only with subject inventions, a federal agency has no other right to ownership or a license to inventions that are not subject inventions. A federal agency is constrained to rely upon 28 USC 1498 for its use of a contractor’s subject invention that is not a subject invention, and exposure to the contractor’s claim for compensation in the Court of Federal Claims.
In other words, the NIST FAQ asks–to what extent should a contractor expect that federal agencies are working a harvester scheme to bring as much of the contractor’s technology within the control of the federal agency without fully specifying and funding the formation of that technology? The answer, given NIST’s FAQ answer, is always.
It is in the interest of federal agencies (including NIST, which does its own share of research contracting) to construe Bayh-Dole to bring as much as possible within its scope. Doing so gives the agency a royalty-free license to practice and opens up the threat of march-in to encourage contractor submissive behaviors. In short, federal agencies under Bayh-Dole gain tools to leash up innovation. It comes as a silliness then for NIST to run conferences about unleashing what NIST has been responsible for leashing in the first place.
Oddly, university administrators also share this interest in expanding the scope of Bayh-Dole. They do so for a similar reason–they use Bayh-Dole to claim that federal law preempts policies on faculty freedom of research and publication and (at a point in the past) policy on patent ownership and requires the university to take ownership of inventions. An expanded scope for Bayh-Dole means that university administrators can take control of more inventions than their policies would otherwise permit, and can do it with the threat that any resistance is illegal.
To answer their question, NIST first defines “subject invention.”
The Bayh-Dole Act applies to a “subject invention,” defined as “any invention of the contractor conceived or first actually reduced to practice in the performance of work under a funding agreement.” 35 USC 201(e).
The Supreme Court in Stanford v Roche ruled that “of the contractor” means “owned by the contractor.” Bayh-Dole applies only after an invention has been acquired by the contractor. Let that sink in. If a contractor has not acquired an invention, then no matter whether it has been conceived or first actually reduced to practice under a federal contract, there is no required disclosure to the government.
Bayh-Dole also defines “invention,” which NIST doesn’t mention–an invention is one that “is or may be patentable” or (through a total ineptness or cleverness in drafting–your choice) protectable under the Plant Variety Protection Act (which is not part of federal patent law, except for this bit in Bayh-Dole). If an invention is not patentable, then regardless of ownership, the invention does not have to be disclosed. Any invention that has not been fully conceived is not patentable. That means any invention for which the inventor has not recognized its inventiveness is not patentable because conception requires the inventor’s recognition of inventiveness. Any invention that has not been reduced to practice is not patentable. Only when there has been both conception and reduction to practice is an invention possibly patentable. One knows an invention is patentable when the Patent Office allows claims. Thus, for Bayh-Dole, the definition “is or may be patentable” in practice becomes “may be patentable”–meaning, there is full conception and full reduction to practice. The FAQ is whether the federal government has a claim when full conception takes place before federal funding but first actual reduction to practice takes place after that funding, “as part of the award.” To parse these things, we have to get at “first” and “actual reduction to practice” as well as the sloppy thinking of “part of the award.”
We are well past NIST’s comprehension of Bayh-Dole at this point. If an inventor is not a contractor, and the inventor owns the invention (which is the legal default), then there is no obligation for the contractor to disclose the invention under Bayh-Dole. NIST cannot comprehend how to deal with such a thing, and so has thrown an assignment obligation into the standard patent rights clause–but that assignment obligation is directed to subject inventions, which a contractor already owns. So that is just needless additional administrative kerfuffling. Nothing like more paperwork to unleash American innovation! Perhaps we shouldn’t be too unkind. If all someone has got to work with is administrative kerfuffling, then we can expect more of it as the way to fix a decrepit, broken policy.
Elsewhere we have worked through what Bayh-Dole does do, how it protects inventor’s rights (when those inventors are not hired to invent), and the parallel law that applies when Bayh-Dole doesn’t. But no one complies with Bayh-Dole, federal agencies don’t enforce the standard patent rights clause, NIST is clueless with regard to how things should operate under the law, and so there’s not much point in revisiting the law itself–it’s not followed and its administration is hopeless. What we have to do is learn how to avoid Bayh-Dole’s corruption.
The key thing in the definition of subject invention is that a contractor has to own an invention that is or may be patentable (only possible if the invention has been both conceived and reduced to practice) before Bayh-Dole applies. But watch what NIST does:
An invention conceived and actually reduced to practice by a contractor prior to commencement of a funding agreement would not be a “subject invention” that would require reporting.
That’s true, but it is not the whole truth. An invention conceived and reduced to practice–no first, no actually (notice how NIST drops the first) is not a subject invention. Further, any invention not owned by a contractor is not a subject invention, regardless of when the inventing happens.
NIST blunders on:
However, an invention which had been conceived but not actually reduced to practice by a contractor prior to commencement of a funding agreement, and which was first actually reduced to practice under that agreement, would be a “subject invention,” even if a provisional application had been filed prior to commencement.
Apart from the last clause (“even if . . . ) the answer here merely repeats the definition. It’s no help at all. “If we own an invention that we have first actually reduced to practice under a federal contract, and the definition of subject invention says that if such an invention is first actually reduced to practice under a federal contract, it is a subject invention, then is it really a subject invention?”
Oh wow. It’s sort of like the FAQ answers for the WTFPL software license. Can I really do anything? Aaargh! But that’s not the question that contractors ask. Their question is, “If I have ideas toward an invention prior to receiving federal funding, how do I keep any inventive activity involving those ideas clearly outside the scope of Bayh-Dole?” You can see why NIST wouldn’t want to be helpful with a meaningful answer here.
Again, if the invention is not owned by the contractor, then it is not subject to Bayh-Dole’s disclosure requirement because it is not a subject invention. If an invention has not been made–conceived and reduced to practice–then it is not a subject invention. If an invention has been conceived before federal funding but with a patent application filed after the start of the federal funding, but without using any test or demonstration data within the scope of that funding, then it is not a subject invention. In what follows, we will get at why this is, and how it comes about in Bayh-Dole’s regulatory framework, and how to avoid federal agency harvesting schemes.