Bayh-Dole Flow Down

Every so often I make an effort to show graphically how Bayh-Dole operates.


The Bayh-Dole Act authorizes (arrow 1) the Department of Commerce to create regulations governing the disposition of patentable inventions made in the performance of work supported by the federal government at nonprofits and small businesses, including standard patent rights clauses to be used in all funding agreements. Let’s focus on universities. 

The Department of Commerce regulations apply to federal agencies that sponsor extramural research (arrow 2). Depending on the form of the funding agreement, an agency follows the requirements at 2 CFR 200 (for grants and agreements with institutions of higher education) or FAR 27.303 (for contracts). These regulations require the use of a standard patent rights clause, which may be tailored in various ways. An agency then introduces a patent rights clause into each funding agreement (arrow 3).

The patent rights clause in each funding agreement governs the disposition of patentable inventions made with federal support. If the funding agreement is a grant, then usually the controlling patent rights clause is 37 CFR 401.14(a), incorporated into the funding agreement by 2 CFR 200.315(c). If a contract, then FAR 52.227-11, sometimes with agency tailoring. The clauses are essentially the same, but their reference numbering differs.

Clearly, the operation of Bayh-Dole is directed to federal agencies, not to universities. In compliance with Bayh-Dole, federal agencies require a standard patent rights clause drafted by the Department of Commerce. This clause, which may vary from agreement to agreement, then governs the relationship between the agency and the university.

But we are not done. The standard patent rights clause (37 CFR 401.14(a)(f)(2) or FAR 52.227-11(e)(2)) requires universities to pass through to their technical employees (arrow 4), those who might be expected to invent in work on a federally supported project, three specific obligations to protect the government’s interest: 1) to report timely subject inventions to persons designated for patent matters by the university; 2) to sign papers to permit patent applications to be filed; 3) to sign papers to establish the government’s rights in subject inventions. This flow down takes the form of a written agreement made by each technical employee with the university on behalf of the government (arrow 5).

[This pass through requirement is not in Bayh-Dole. It is entirely a creation of the Department of Commerce and tracks a similar (but also different) requirement that in the Federal Procurement Regulation from 1975–and replaced in 1984 by the Federal Acquisition Regulation, which in turn only implements Bayh-Dole (and which codified the Nixon executive branch patent policy of 1971). In May 2018 NIST added an assignment requirement to the standard patent rights clause, and not only is there no authority in Bayh-Dole for such a requirement but the Supreme Court in Stanford v Roche expressly ruled there was nothing in Bayh-Dole to authorize such a requirement and given the lack of protections in Bayh-Dole for inventors and other “third parties,” the Court indicated that it would find such a requirement “deeply troubling.” Not only that, but NIST screwed with the section labeling in the standard patent rights clause, conflating the standard patent rights clause for small businesses and nonprofits (a) with a standing declaration of exceptional circumstances for certain nuclear power applications in Department of Energy programs (b). Finally, NIST’s assignment requirement is specific to subject inventions. The Supreme Court ruled that a subject invention was an invention already owned by a party to a funding agreement and made in the performance of work under that agreement. Thus, the NIST assignment requires inventors to promise to assign to contractors what the contractors already must own–and contractors cannot own any invention by operation of Bayh-Dole. So, inventors must assign inventions that they have already assigned. Big wow.

Oh, yes–there is another possibility. The requirement at 37 CFR 401.14(a)(f)(2) also makes inventors parties to the funding agreement by way of a compulsory subcontract required by the federal agency and entered into by the contractor. But if that’s the case, then inventors are also contractors, the patent rights clause at 37 CFR 401.9 for inventors kicks in, and inventors are treated as small business contractors with the same right as other contractors to elect to retain title to their subject inventions. The standard patent rights clause, again ranging outside Bayh-Dole’s authority to copy from the Federal Procurement Regulation, forbids contractors from having any interest in any subject invention of a subcontractor as a consideration for the subcontract. In effect, contractors cannot require the assignment of subject inventions owned by other contractors–and therefore, if a contractor complies with the standard patent rights clause here at 37 CFR 401.14(a)(f)(2)–now just (f)(2)–or perhaps just somewhere in (f)–then any assignment requirement it may impose cannot be part of the standard patent rights clause because the standard patent rights clause forbids it. Thus, if contractors comply with the standard patent rights clause, they can require only the assignment of subject inventions that have already been assigned to them. If they don’t comply and don’t require a conforming written agreement from inventors, then those inventors are not parties to the funding agreement and their inventions cannot be subject inventions, and again cannot possibly be the subject of NIST’s stupid assignment requirement.

I know. Even moderately complicated logic is beyond the capacity of federal and university bureaucrats, and if something is too difficult for them, they assume they have the right to make something up that sounds good to them, present it as legal fact, and let their attorneys bully up anyone who disagrees.

Is NIST clueless or malicious? Ah, it doesn’t matter because they won’t be held accountable.

Whatever the case,] this flow down to inventors has nothing to do with protecting a university’s interest in owning inventions. The US Supreme Court in Stanford v Roche was clear that Bayh-Dole does not have anything to do with conveying personal patent rights to universities. Furthermore, volunteers, non-employee collaborators, and clerical employees are expressly not within the scope of this requirement. The standard patent rights clause recognizes that inventors own personally patent rights in the inventions they make. The (f)(2) requirement obligates the university to delegate to technical employees the obligation to “establish the government’s rights in the subject inventions.”

This flow down is crucial to how the standard patent rights clause operates. By requiring this flow down, a university adds its technical employees as parties to the funding agreement, for these three specific obligations, which become effective if there’s a patentable invention. That invention becomes a subject invention because the technical employees who make the written agreement required by the university become “contractors” within the definition of the standard patent rights clause, and therefore their inventions are “of the contractor” and thus within the scope of “subject invention.” The flow down is also a safety provision to cover for any situation in which a university does not have an enforceable patent agreement with each technical employee. In the now distant past, many universities did not have a patent agreement with faculty–the university made no claim on patentable inventions. In such cases, the (f)(2) obligation created the equivalent of a patent agreement within the patent rights clause of the federal funding agreement.

The flow down of (f)(2) is parallel to that of the subcontracting clause, in paragraph (g) of the standard patent rights clause. As required by (g), a university must include the standard patent rights clause in any subcontract. The university is also forbidden from using the subcontract as a condition for obtaining patent rights from the subcontractor. That is, the subcontractor is required to make a written agreement (the standard patent rights clause) on behalf of the government’s interest–just as are technical employees.

University inventors become more than simply “employees” by operation of the (f)(2) clause. The university is required to give them standing as contractors with regard to their patent rights. If the inventors do not assign their patent rights to the university or another approved organization, or to the federal government, then the regulations provide a separate patent rights clause (37 CFR 401.9) to govern the relationship between the federal agency and the inventors, independent of the university. The (f)(2) provision does not merely subcontract–it requires the university to delegate contractor status to its employees and then require the employees to make commitments to protect the federal government’s interest in their inventions.

The diagram above shows how the statute authorizes regulations, which in turn mandate a standard patent rights clause. The patent rights clause is incorporated into funding agreements, accepted by universities. Universities, in turn, flow down certain obligations to potential inventors, who then become parties to the funding agreement.

[Remember, too, that this schema is the default. Bayh-Dole permits federal agencies to alter the standard patent rights clause for such things as “exceptional circumstances.” Thus, the controlling instrument in any given funding situation is the patent rights clause specific to the funding agreement. Not Bayh-Dole’s arbitrary default clause. Not the confused and stupid arbitrary expansion of Bayh-Dole’s default clause by the Department of Commerce. But the clause as actually inserted into a specific funding agreement.

Remember, too too, that if a contractor assigns, substitutes parties, or subcontracts in any way, those on the other end of any such action also become parties to the funding agreement–that is, contractors. For which see the definitions of funding agreement and contractor at 35 USC 201. When a contractor adds parties to a funding agreement, the patent rights clause changes to suit the status of the added party. Thus, if a small business adds a nonprofit, the nonprofit operates under the nonprofit version of the standard patent rights clause, and if a small business adds a large company, the large company does not operate under Bayh-Dole but rather under the Reagan-amended executive branch patent policy, which NIST unhelpfully and ever more stupidly merged into its implementing regulations for Bayh-Dole.

All this *except* for one situation–if a nonprofit assigns a subject invention to a for-profit, the for-profit must accept as well the nonprofit’s patent rights clause–including, in the default version sharing royalties with inventors (in the case of licensing), limiting any other draw off of income earned with respect to the assigned invention to administrative expenses incidental to managing subject inventions, and dedicating any remaining income to the support of scientific research or education. See 35 USC 202(c)(7) That’s a huge exception. Any exclusive license that conveys all substantial rights in an invention–make, use, and sell–and gives the licensee the right to enforce patents on the invention is an assignment of the invention, even if the licensor insists that it retains title to the patents. Just saying. Bayh-Dole only “works as intended” if no one complies. Stupid, sucky law.]

This entry was posted in Bayh-Dole and tagged , , , . Bookmark the permalink.