Third layer: Patent rights clauses, continued
We are still working through Bayh-Dole’s third layer, the patent rights clauses. We finish up the discussion of the (f)(2) written agreement requirement with another look at subject invention reporting and assignment, the parallels to the (f)(2) requirement in the subcontracting requirement at (g), and the result: the inventor patent rights clause at 37 CFR 401.9 is in fact the primary patent rights clause required by Bayh-Dole. This is true even with the 2018 NIST rule change that requires organizations hosting federally supported research to require inventors to assign their inventions to the host organization. We finish with a look at the special requirements of the nonprofit patent rights clause.
The apparatus of the contracting section of Bayh-Dole is directed entirely at subject inventions. The US Supreme Court made it clear that an invention must be a subject invention to be within the scope of the law:
But because the Bayh-Dole Act, including §210(a), applies only to “subject inventions”—“inventions of the contractor”—it does not displace an inventor’s antecedent title to his invention.
As the Supreme Court also ruled, an invention is not a subject invention until it is owned by the contractor. The result: inventors have no obligation to disclose their inventions under Bayh-Dole, unless a contractor complies with the (f)(2) written agreement requirement.
There’s absolutely no mention of such an inventor disclosure obligation in the statute. Here is about as close as we get:
(1) That the contractor disclose each subject invention to the Federal agency within a reasonable time after it becomes known to contractor personnel responsible for the administration of patent matters, and that the Federal Government may receive title to any subject invention not disclosed to it within such time.
If Bayh-Dole was going to require federal agencies to require reporting of inventions by inventors, this would be the place to have done it. And this provision is one that was amended in 1984. Here’s the original text:
(1) A requirement that the contractor disclose each subject
invention to the Federal agency within a reasonable time after it is made . . .
In the original Bayh-Dole, there might be an argument that a contractor have some mechanism in place to require reports of inventions made in projects receiving federal support and that the contractor claims ownership of in order to be able to report such inventions to the federal agency.
The 1984 amendment walks back the timing for such disclosures–inventors can take an indefinite time to report inventions–the clock starts only when the proper people designated for patent matters receive an invention disclosure. It’s a strange amendment, to be sure. If anything, it weakens any authority in Bayh-Dole for a patent rights clause to obligate inventors to disclose subject inventions. In any event, clearly, Bayh-Dole has nothing to say about the disclosure of inventions that are not subject inventions, not even “just in case” an invention might be a subject invention. Again, even inventions made with federal support are not subject inventions until a contractor acquires ownership of the invention. If a contractor does not bother with invention ownership, Bayh-Dole’s requirements for patent rights clauses with nonprofits and small businesses do not apply.
This is where law professor Sean O’Connor says, “See, there must have been a mistaken assumption that universities had patent assignment agreements.” But that cannot be. University patent brokers helped to write the law, just as they helped to create the IPA program that the law was loosely based upon. They knew there was no consistent university practice of invention assignment agreements, and not even patent policies that necessarily required assignment–and especially at the University of Wisconsin, where Howard Bremer, a key contributor to both the IPA program and Bayh-Dole, operated at WARF. Something else must have prevented Bayh-Dole from including a requirement that inventors disclose their inventions to their employers soon after those inventions had been made. It was no mistaken assumption.
Bayh-Dole does not require reporting of inventions by inventors, nor even the reporting of subject inventions by inventors. Reporting by inventors is avoided–“after it becomes known” is about as far from requiring inventors to report inventions to their employers as the law can get and still set a time line for the employer to report inventions to the federal agency.
The Supreme Court confirmed that Bayh-Dole’s contracting requirements apply only after an invention becomes a subject invention. There’s not even a matter of “non-compliance” for a small business or nonprofit if it doesn’t have any “backup” mechanism by which inventors report inventions anyway, and the organization takes ownership of them to turn them into subject inventions.
Without the (f)(2) written agreement requirement, an organization could ignore Bayh-Dole’s patent rights clause altogether–simply take no ownership position in inventions made with federal support. Presumably, then, statutes and executive branch regulations preempted by Bayh-Dole for subject inventions would continue to operate–but the organization would have no compliance obligations. The organization could choose not to participate in the exploitation of patents on inventions made in projects receiving federal support. Alternatively, a clever organization could implement a policy that requires inventors to assign their inventions to a holding company or invention management agent, and use that holding company or agent as the base to manage the inventions and any patents. There would be no violation of Bayh-Dole. There would be no frustration of Bayh-Dole’s federal policy. It’s a dirt easy workaround, but for (f)(2), which of 2018 requires each contractor to require inventors to assign subject inventions–inventions owned by the contractor–to the contractor.
Since Bayh-Dole’s preemption clause (35 USC 210) applies only to subject inventions, as the Supreme Court ruled, for all those inventions arising in federally supported projects that are not subject inventions, the prior regime of statutes and executive branch patent policies still apply. If an invention is not made a subject invention, then the federal government still has the authority to take title to that invention that the government had before Bayh-Dole. Section 210 does not repeal executive branch patent policy or statutes that make special provisions for the disposition of inventions–it merely preempts those policies and statutes for subject inventions.
We can see, then, why the (f)(2) requirement is as it is. When a contractor complies with (f)(2), it turns inventors into contractors, and because the inventors own their inventions, those inventions become subject inventions, and so come within the scope of Bayh-Dole’s definition of subject invention–but now not under the inventor patent rights clause authorized by 35 USC 202(d) rather than the standard contractor patent rights clause required by 35 USC 202(c) and implemented at 37 CFR 401.14(a). The (f)(2) requirement gives federal agencies a direct relationship with each inventor–subject to a patent rights clause specific to inventors. Within that relationship, it is reasonable to direct inventors to report subject inventions through their employers, so that employers have an opportunity to identify those inventions to which they may have an ownership claim independent of any federal funding agreements.
But here’s the strangeness. No-one complies with this (f)(2) requirement. I’ve never seen any university or company comply. Instead, organizations create policies that demand assignment of all inventions, and claim that this demand renders (f)(2)’s written agreement requirement moot. They are wrong, of course, but that’s how the “law” has operated for three decades–Bayh-Dole, the law regarding a contract that everyone must use but no-one complies with. The new assignment requirement introduced into (f)(2) by NIST in 2018 means that any prior contractor requirements to assign are displaced by the federal requirement presented by (f)(2)’s assignment provision. An organization cannot rely on its own requirements to assign–it has to comply with the new assignment rule, which is a matter of federal contracting, not state-enforced contracting.
Section (f) has a parallel in section (g), which involves subcontracts. A contractor in subcontracting must “flow down” the patent rights clause to subcontractors, who must do the same if they subcontract. The three sections of (g) deal with three different subcontracting situations:
(1) nonprofit or small business subcontractor
(2) any other subcontractor
(3) when the prime contract was a contract
Here are the differences. In the case of (1), the contractor cannot have an interest in the inventions of any subcontractor, as a condition of the subcontract:
The subcontractor will retain all rights provided for the contractor in this clause, and the contractor will not, as part of the consideration for awarding the subcontract, obtain rights in the subcontractor’s subject inventions.
In case (2), where the subcontractor is a non-small company or foreign company, the contractor can require an interest in any subcontractor invention but the subcontract must follow the proper form of the patent rights clause for such a subcontractor. In case (3), if the prime contract is a contract rather than, say, a grant, then the subcontract forms a new contract between the subcontractor and the federal government involving the standard patent rights clause.
The first case, involving nonprofit or small business contractors, is important for establishing the fourth standard patent rights clause, at 37 CFR 401.9, involving employee-inventors. The inventor patent rights clause instructs federal agencies to treat inventors as small business contractors:
Agencies which allow an employee/inventor of the contractor to retain rights to a subject invention made under a funding agreement with a small business firm or nonprofit organization contractor, as authorized by 35 U.S.C. 202(d), will impose upon the inventor at least those conditions that would apply to a small business firm contractor under paragraphs (d)(1) and (3); (f)(4); (h); (i); and (j) of the clause at § 401.14(a).
If you have spent your time reading standard university-generated accounts of the Bayh-Dole Act (whether from patent administrators, law professors, or faceless university-supported lobbying groups), this result may be shocking. According to the Supreme Court ruling, Bayh-Dole applies here only to subject inventions. Federal agencies, then, according to the Supreme Court have no authority under Bayh-Dole to refuse to “allow an employee/inventor to retain rights to a subject invention”–so long as the employee/inventor discloses the invention and elects to retain title. Under (d)(1) and (d)(3), a federal agency can request title to a subject invention if the contractor (now the employee/inventor) fails to disclose the invention (either to the federal government or in compliance with (f)(2) to his or her employer).
Paragraph (d)(2) and all of section (c) are absent from the inventor patent rights clause–an inventor is not required to file patent applications, and the federal government’s right to obtain title is restricted to rights in foreign patent applications. Rather more strange is the absence in the list of section (b)–there’s no requirement for the federal agency to receive a non-exclusive license to the subject invention. The other requirements specified are these:
(f)(4) notice of federal funding in patent applications
(h) reporting on invention use
(i) preference for US manufacturing
The (f)(2) requirement to require a written agreement then comes into some clarity, once we have seen the parallel (g)(1) requirement that small business and nonprofit contractors cannot have an interest in subject inventions acquired by small business or nonprofit subcontractors. Inventors are declared to be treated as small business contractors. The requirement to require a written agreement operates as a compulsory subcontract to select employee-inventors, where the consideration for the subcontract cannot have anything to do with the contractor’s interest in inventions. Any interest that a contractor might claim in inventions made with federal support in the subcontracted work with a small business or nonprofit contractor has to be entirely outside the subcontract.
The effect of (f)(2) and (g)(1) is that the inventor patent rights clause at 37 CFR 401.9 is the primary patent rights clause under Bayh-Dole. 37 CFR 401.9 is the only patent rights clause that pertains to inventions that become subject inventions simply because an inventor owns his or her own inventions made in projects receiving at least some federal support. The (f)(2) assignment requirement introduced in 2018 applies to subject inventions owned by the research host contractor, not to subject inventions owned by the inventor contractor. That’s the only way to make sense of 37 CFR 401.9 and the (f)(2) requirement that inventors agree to sign papers to establish the government’s interest in subject inventions. If the inventors had to assign their subject inventions to the research host contractor, then they would have no rights remaining to establish any government rights in their inventions. If the 2018 (f)(2) assignment requirement had intended to cover all such inventions, then the requirement to sign papers to establish the government’s rights would have been deleted. But it was not deleted.
Although Bayh-Dole is organized to focus on contracting with small businesses and nonprofits, as a matter of the implementation of the patent rights clauses, the inventor patent rights clause becomes the fundamental patent rights clause in Bayh-Dole. This outcome is particularly important for university faculty, who when they invent with federal funding, are not required to file patent applications in the U.S.–they can publish their invention without worrying about patents, so long as they disclose their inventions and elect to retain title. In short, the implementation of Bayh-Dole respects academic freedom: faculty cannot be forced to publish in the patent literature by either their employer (via (f)(2) and (g)(1)) nor by the federal agency funding their work (via 37 CFR 401.9’s elimination of the requirement to file a patent application).
If we then re-present Bayh-Dole based on its implementation, we would describe things as follows:
Bayh-Dole requires federal agencies to require organizational contractors to allow inventors working in projects at small businesses and nonprofits supported at least in part by federal funding to retain title to their inventions, provided they timely disclose their inventions and elect to retain title. These inventions become inventor subject inventions.
If an inventor’s actions create a claim to equitable ownership by a research host, an inventor must assign his or her entire right, title, and interest to the inventor subject invention to the business or nonprofit contractor, and the invention becomes subject to the business or nonprofit patent rights clause in the federal funding agreement.
If, however, the employing organization makes the assignment of inventions a condition of participation in the federally funded project or the use of resources committed to that project and the like, such an agreement is preempted by the (f)(2) written agreement requirement.
Look at 35 USC 202(a) again, given this context:
Each nonprofit organization or small business firm may, within a reasonable time after disclosure as required by paragraph (c)(1) of this section, elect to retain title to any subject invention
Inventors are “small business firms” for the purpose of the inventor patent rights clause. The (f)(2) compulsory subcontract makes inventors also contractors. The (f)(2) agreement requires inventor contractors to disclose inventions so that the employer-contractor can comply with their disclosure requirement at (c)(1).
Only when inventors assign subject inventions to their employers do the inventions come under the employers’ patent rights clauses at 37 CFR 401.14(a). No one reading only Bayh-Dole would ever reach this understanding of the implementation of the law, and yet the regulations make this result clear, even with NIST’s effort to muddy the regulations still further by inserting an assignment requirement for subject inventions.
We can turn, finally, to the differences between the business patent rights clause and the nonprofit patent rights clause. Bayh-Dole requires that the nonprofit patent rights clause carry four additional requirements. These are set forth in section (k) of the patent rights clause at 37 CFR 401.14(a):
(k)(1) requirements on assignment of subject inventions
(k)(2) sharing royalties with inventors
(k)(3) use of royalties and other income
(k)(4) preference for small business licensees
We will mention a few aspects of each of these requirements. (k)(1) requires federal agency approval for nonprofit assignment of subject inventions, unless the assignment is to an invention management organization. The key thing is that (k)(1) requires the assignee to accept the nonprofit’s patent right clause. Once a nonprofit has acquired ownership of a subject invention, it cannot assign that invention to anyone else without passing through the nonprofit patent rights clause–including, especially, the requirements in this section (k). This assignment restriction is directed at subject inventions, not at patents on subject inventions. That means that a nonprofit cannot grant an exclusive license to make, use, and sell a subject invention (which assigns the invention) unless it requires the licensee (now an assignee) to manage the subject invention under the nonprofit clause. The implications of this restriction become clear when we turn to the remaining paragraphs in (k).
Businesses have no restrictions on assignment of subject inventions. Subject inventions when assigned by businesses do not carry with them the business patent rights clause. Thus, an easy way for a business to circumvent restrictions such as the requirement for exclusive licenses to use or sell in the U.S. to carry a requirement to use product manufactured in the U.S. is for the business to assign the subject invention. Upon assignment, the U.S. manufacturing requirement no longer operates except in the case of further exclusive licensing by the assignee.
Paragraph (k)(2) requires nonprofits (and nonprofit assignees) to share royalties with inventors. A royalty is any consideration for a license, so this provision is specific to royalties. If a nonprofit sells product based on an invention (such as a biomaterial or piece of software) or sells research services tied to an invention, that income is generally not a royalty and does not have to be shared under (k)(2)–although the nonprofit may have its own policy requirements for sharing such other income. An implication of (k)(1)’s chain of nonprofit patent rights clauses, then, is that each assignee in a nonprofit chain of subject invention owners must share royalties with the inventors.
For example, if a university acquires a subject invention from its inventors, the university must share royalties. If the university grants an exclusive license that has the effect of assigning the subject invention, then the assignee also has an obligation to share royalties with the inventor on transactions that would be described in the exclusive license as sublicensing. Such transactions may take place when a university moves a subject invention to an affiliated research foundation or commercialization corporation. If the assignee grants an exclusive sublicense (and thus assigns the invention again), then the sublicensee also has an obligation to share royalties with the inventors. Although this is not how universities do things, it is what the patent rights clause requires, based on what Bayh-Dole requires. Inventors must share in royalties from each assignee in the chain of assignments, not just the first assignee. Otherwise, one can see that as assignments pile up, each assignee taking its cut, the amount that finally comes back to the university to share with inventors is a small percentage of the royalties earned further down the chain of owners. In this way, (k)(2) is a non-dilution requirement to protect nonprofit inventors’ financial interest in their inventions.
Businesses have no similar obligation to share royalties with their inventors.
Paragraph (k)(3) restricts how nonprofits can use any income from subject inventions, whether by licensing or otherwise (“royalties or other income earned by the contractor with respect to subject inventions”). A nonprofit is limited in the expenses it can charge to a revenue stream from a subject invention–it may recover only “expenses incidental to the administration of subject inventions,” those expenses include payments to inventors. Any money left over must be used “to support scientific research or education.”
One can see the effect of (k)(3) on any contemplated assignment by a nonprofit of a subject invention to a for-profit company, such as by means of an exclusive patent license to make, use, and sell the subject invention and to enforce the patent on the invention. Such an exclusive license is an assignment of the subject invention. The for-profit company then operates under the nonprofit patent rights clause, including (k).
The upshot: the company cannot book income with respect to the subject invention any way it wants. Income earned with respect to the subject invention must be dedicated to public purposes–scientific research or education. For-profits might be expected to refuse such exclusive licenses, choosing instead a non-exclusive license to most rights (perhaps an exclusive license only to sell). A reduced license amounts to the nonprofit breaking up the patent monopoly right when it licenses. Breaking up the patent monopoly right opens up the subject invention to “free competition and enterprise” as other companies have an opportunity to make and use the invention, even if they cannot sell it. At least they do not have purchase it from the exclusive licensee for their own use–if the exclusive licensee ever develops a product to sell, which turns out to be rare.
Finally, there’s (k)(4). Nonprofits must have a preference in licensing for small companies. The preference includes a requirement that the nonprofit “will make efforts that are reasonable under the circumstances to attract licensees of subject invention that are small business firms.” This requirement is distinct from simply licensing inventions to small businesses in general–it is specific to subject inventions, and it means there must be a preference in licensing, and it means that the nonprofit must make efforts to “attract” small business firms. This same requirement then extends to any assignee of a nonprofit, including those taking exclusive licenses to make, use, and sell. While they may wish to make their own product, (k)(4) requires them to make efforts to attract small business licensees as well–something that would promote free competition and enterprise, but which might not sit well with the typical for-profit company used to licensing patents exclusively as justification to develop new products.