What Happens?

I’ve spent some time in previous posts mapping out ways title to inventions is managed under Bayh-Dole. I’ve argued that Bayh-Dole sets up an apparatus of disclosure and obligation that permits a range of practices by universities while protecting government interests in support of the objects of the law–which start with using the patent system to promote the use of federally supported inventions. I’ve traced through the language of the law to show that this reading is at least permitted, and accounts for the language of the law throughout, as well as long-standing university practice. There is no need for a reading supporting federal preemption of inventors’ personal rights. Or, if someone is going to go there, there is a need to displace this reading, not simply assert that preemption is right because one wants to be right and it would be bad, or at least inconvenient, if one were wrong.

But so what? Why does it matter? Let’s look in two directions for the answer. First, generally, how does title to inventions move under Bayh-Dole? Second, next post, what might this mean in the case of Stanford v. Roche?

(Note: I use here university for “contractor” and for all the others that might be included in Bayh-Dole; it’s a shorthand, but it’s also the university folks I’m most concerned with at the moment. )

There are three ways for a contractor to deal in title under Bayh-Dole: assign, subcontract, and jointly invent. There’s a fourth way, in which the government steps in. We deal with these in turn: one, two, four, and three.

Assignment is managed at 37 CFR 401.14(a)(k)(1) [Now 37 CFR 401.14(k)(1)]. Here, universities are allowed to assign their “rights” in subject inventions to other organizations having a primary purpose in managing inventions, and to any other organization with agency approval. The language is broader than that of simply assigning title to inventions. The language occurs here in a contract clause–401.14(a)–to be including in each funding agreement (but for those with exceptions). The rights that may be assigned are those available to the contractor under Bayh-Dole—the right to elect to retain title, the right under written agreements to cause inventors to execute papers necessary to file patents and to obtain patent protection, and the like. That is, again, the invention rights in Bayh-Dole are treated as rights under contract to obtain title to invention, combined with contractual obligations to seek patent protection, and use those patents to promote the utilization of the underlying inventions.

Bayh-Dole is implemented in funding arrangements. That’s the point of 401.3(a): put this set of terms in every funding arrangement, as a contractual condition. If the contractor (or an assignee of the contractor) does not exercise those contractual rights, then the federal agency has them through the required written agreements. Another way of putting this: Bayh-Dole does not require special performance of the contractor. The contractor can assign or subcontract away some or all of the obligation, so long as whomever receives the rights under contract to manage the rights to inventions accepts the obligations under Bayh-Dole that come with those contracted rights.

Subcontracting is managed at 37 CFR 401.14(a)(g). Universities may subcontract research to other organizations, without the restrictions for assignment, provided that the Bayh-Dole obligations flow down to subcontractors who are qualified to receive funds under Bayh-Dole, and other agency procurement clauses are imposed, as the agency directs, on any other organizations. The subcontracting requirements provide that when the prime funding agreement is in the form of a contract, each subcontract sets up as an independent contract between the funding agency and that subcontractor. The prime does not have “privity” in that award pertaining to patent matters. The relationships over patents are governed directly between the agency and the subcontractor. When the funding agreement is in the form of a grant or cooperative agreement, however, the flowdown effect of the clause is to implement Bayh-Dole for the subcontractor in the place of the prime contractor. In essence, the prime assigns performance under Bayh-Dole to the qualified subcontractor.

In short, if the prime funding agreement with the agency is in the form of contract, then there is also a contract between the agency and a subcontractor with regard to patent matters (which may be Bayh-Dole or not). If the prime funding agreement was a grant or cooperative agreement, then there is still a contract between the prime and the sub allocating Bayh-Dole responsibilities, but there is no contract formed between the sub and the agency.

This level of detail helps us to see that a federal grant to a university is not construed as a federal contract. That is, the award operates under Circular A-110 (where one finds Bayh-Dole at __.36(b)) [now 2 CFR 200.315(c)] rather than under the Federal Acquisition Regulations. The relationship is one of compliance with the terms of the award, not a federal procurement contract. As such, the deliverables are undertaking the proposed work in compliance with the award plus this Bayh-Dole rider.

It should be emphasized how Bayh-Dole is implemented in the language and style of contracts. In an award, Bayh-Dole is a condition of the award. In a contract, it forms the patent clause along with other matters regarding the interests of the parties. If Bayh-Dole applied to contractors directly, as law, wherever federal money was being spent on research, there would be no need for this apparatus of contract formation in subcontracts. But Bayh-Dole is a law attached to potential procurement. What it manages is the disposition of contractual rights to inventions made with federal funding, with the government deferring its interest in procurement in favor of local choice in the matter of inventions. It harmonizes that interest in inventions across agencies and normalizes the ways in which an agency may vary from these defaults. It does not dictate to universities that they manage inventions or patents in a uniform way, but rather how the agencies uniformly manage their end of the bargain.

If a university subcontracts a portion of research under Bayh-Dole, the subcontractor ends up with either a duty to, or a contract with, the federal government. Either way Bayh-Dole prevents the prime contractor from having a claim in the subcontractor’s inventions. For flow down of Bayh-Dole from a grant, such an interest is forbidden as a condition of the subcontract. For contracts to organizations not qualified under Bayh-Dole, the university must impose the FAR clause specified by the agency–which will pertain to agency interests in inventions arising in the subcontract, not the university’s interests. The contractor and subcontractor can deal in inventions between themselves, but either way it has to be for some other consideration than the granting of a subcontract. That is, it may be voluntary (the subcontractor could choose to assign its Bayh-Dole rights to the contractor, for instance, as it might in an inter-institutional agreement following on joint invention), or it may bargain with contractor to assign or license patent rights to the contractor, or to a contractor’s designee, as one might in a joint venture).

Before turning to the third way, we will note the fourth. The government may obtain title to inventions in certain circumstances, as set out in 37 CFR 401.14(a)(d): if the university fails to disclose an invention or elect title to an invention as required; elects not to retain title; or fails to file patent applications, or decides not to continue prosecution of an application or maintenance of a patent, or defend a patent in a reexamination or opposition proceeding. In these circumstances, the university is obligated to “convey” title. It is worth marking the use of “convey”. Why not “assign” (if the university already has the rights, vesting with it outright?) or “transfer or re-assign” (if it could obtain the rights either way)? One could say, “it’s just a synonym used randomly for effect, or a drafting artifact, an accidental.” Or one could say, “convey is used because if the university has not done those things necessary to obtain title from the inventors, then its obligation under Bayh-Dole is to cause the title to be assigned by its inventors as obligated by written agreements to protect the government’s interest, as the government requests in writing.” That is somewhat longer, and therefore inconvenient, but is just as clear as the short cut, and makes better sense of the use of “convey” which now is a general term to cover all possible routes by which the university may cause the title to be established with the government–transfer or re-assign or cause to be transferred, assigned, or re-assigned, depending on what has gone on before the default. The university has the obligation to enforce those written agreements with its research personnel to protect the government’s interest in obtaining title when the university defaults or chooses not to exercise its own options under Bayh-Dole.

[These issues were resolved by the Supreme Court’s decision in Stanford v Roche in 2011. Title does not vest in a contractor. A subject invention is one acquired by a contractor in a conventional way. What was left unaddressed was the effect of the patent rights clause’s (f)(2) written agreement requirement, under which contractors are required to require certain of their employees (but not non-employees) to make a written agreement to protect the government’s interest. That requirement is not in Bayh-Dole. To put it there would appear to require a determination of exceptional circumstances under Bayh-Dole. But nothing of the like has been done. Is that written agreement requirement therefore even enforceable? Well, no one complies with it, so that’s something? But if it is a valid requirement, and no one complies, does that undermine a contractor’s claims on any invention?

What’s clear is that if the (f)(2) requirement is met, each employed inventor is made a party to the funding agreement, and thus is by Bayh-Dole’s definition a contractor, and any invention thus made is necessarily a subject invention under the Supreme Court’s reading of the law, and thus the inventor is subject to the patent rights clause at 37 CFR 401.9 not the patent rights clause at 37 CFR 401.14. Inventors under 37 CFR 401.9 are treated as small business contractors but with fewer obligations–for instance, they have no obligation to file a patent application.]

The third way of managing title is through joint invention. The contractor and its research personnel have their choice of collaborators. Bayh-Dole does not preclude the use of funding from other sources, or the involvement of other personnel in the research. The policy expectations are laid out in 37 CFR 401.1. There, the regulations are at some pain to consider the impact of funding or involvement of “other sources” in “closely related” projects supplying funding or aid “for the purpose of expediting or more comprehensively accomplishing the research objectives of the government sponsored project.” For these, what happens? First, the regulations say that the “ownership provisions” apply to any invention made in the performance of the federally supported project. What isn’t said is whether this means that the ownership provisions preempt other arrangements pertaining to title to investment (as some might wish for) or whether this means that the university is required to include these inventions in the management of its share of any such invention, to the extent that it comes with the scope of the federally supported project, that is, whether the invention falls outside the “planned and committed activities” of the project. Here’s the example given by the regulations:

An example of such related but separate projects would be a government sponsored project having research objectives to expand scientific understanding in a field and a closely related industry sponsored project having as its objectives the application of such new knowledge to develop usable new technology.

The scope of a project is given by its planned and committed activities, which would in turn be set forth in the proposal for research submitted to the agency (or the solicitation specification, in the case of some contracts), along with the budget that allocates resources to do the work. Separate accounting isn’t determinative. Nor is the timing of the collaboration. What matters is whether the invention comes within the scope. If the invention is in a separate but related project, and does not “diminish or distract” from the federally supported work, then it is not a subject invention.

One might think at this point that it may be an open issue then whether research to investigate whether, say, a particular way of assessing the effectiveness of a treatment for a disease might be the planned and committed federal activity, and the use of this information applied to the development of a particular assay might be a “closely related” project that does not diminish or distract but may indeed expedite or more comprehensively accomplish the federally funded project. Certainly careful attention to scoping should attend a determination of an invention as a subject invention when collaboration with other sources is involved.

In this, one might even ask whether it is a violation of Bayh-Dole to label as a subject invention something that is not, whether through inattention, to make federally supported research appear more productive, or to gain control of an asset that otherwise may be beyond one’s reach. It is one thing to manage an invention in a manner similar to Bayh-Dole, and another to assert that it is a subject invention when a reasonable investigation into the matter would show that it is not. One might wonder what would happen if a government audit of reported subject inventions found a number were in fact not, and that election to retain title had no bearing on such inventions, and that representations to inventors that an invention was “subject” and subsequent reliance on written agreements required under Bayh-Dole (for those who did not find notice of election of title adequate to establish actual title) cast into doubt the premises on which assignments were made.

Universities operate, for the most part, open research facilities. This openness includes exchanges among laboratories in the same institution, as well as across institutions, and between non-profit and for-profit organizations. Science is science, wherever one finds it, regardless of the tax standing of a given employer. Bayh-Dole obligations on disclosure, written agreements, and education on timely disclosure are directed at the university’s employees, not all research personnel. The written agreements required at 37 CFR 401.14(a)(f)(2) are potentially distinct from those of employment, however, as suggested by the reference to employment agreements later in the same paragraph. In particular, the university policy most closely aligned with these agreements is generally that of compliance by university personnel with the requirements of extramural research contracts.

With a joint invention, collaborators that are not employees of the university and therefore not within the scope of written agreements directed to the university’s management of its Bayh-Dole obligations (to protect the government’s interest) also have an ownership interest in the inventions. How does this square with 37 CFR 401.1, which provides that the ownership provisions apply if the invention comes within the scope of the federal project? First, it’s clear that the university must follow Bayh-Dole. What about the “other source’s” interest? Is that interest preempted by Bayh-Dole? If so, then what does this mean for, say, a company collaborator in a federal project? Does it mean that the company must also follow Bayh-Dole, even if Bayh-Dole might not otherwise be indicated? That is, the company must decide whether or not to elect to retain title? Or does Bayh-Dole stipulate that title to the invention is wholly with the university and the company personnel who co-invent have no rights in their work to assign to their compay? Or does it mean that only the university and its personnel must follow Bayh-Dole with regard to their undivided ownership interest in the invention, but Bayh-Dole does not apply to the company collaborator?

To resolve these questions, we can point to 37 CFR 401.10, which makes clear that when the collaborator is a federal employee, Bayh-Dole does not apply to the federal employee or his/her agency. That is, it is not law of the land, as it were, but only law of the university contractor under a funding agreement. In the case of a federal co-inventor, Bayh-Dole only applies to the university interest. We can expect that this also provides guidance for the situation in which the co-inventor is from a company, or even another university (but not funded there with federal funds to be associated with the federal project). The collaborator is not subject to Bayh-Dole unless its work falls under the “planned and committed activities” of the federal award–that is, its involvement is in the scope of work and/or budget of the award, so that it has accepted Bayh-Dole provisions through its own application for funds (and subsequent funding agreement), or through a permitted subcontract under 37 CFR 401.14(a)(g), or by assignment of Bayh-Dole rights in inventions as authorized by 37 CFR 40.14(a)(k)(1).

Collaborators’ interest in patents is conventional: that is, it is treated as personal property owned by the inventors, subject to any obligations the collaborators may have, such as to assign patent rights to their employers, which may take the form of a promise to assign or a present assignment to future inventions. For joint inventions, there is no flow down of Bayh-Dole to the collaborators or their organizations absent a mechanism in contracting to do so. [For which see Bayh-Dole’s definition of a “funding agreement.”] Bayh-Dole applies to the university’s prospects for an undivided interest in such joint inventions, but does not affect the disposition of the undivided interest of collaborators. There is no “taking of title” by federal fiat from collaborating joint inventors simply because they contribute privately to a federally funded invention. Nor do they have an obligation to assign based on Bayh-Dole. Nor does the university need either a preemption or assignment of their rights to comply with its obligations under Bayh-Dole. [And this is what the Supreme Court ruled in Stanford v Roche.]

The situation with joint inventions is this: the university contractor (should it elect to retain title) is obligated to follow Bayh-Dole in the management of its interest in subject inventions and any patents that issue on these. The university is obligated to file patent applications (or cause these to be filed), and to do those things with its licensing to accomplish the objectives of the law (use the patent system to promote the use of the inventions). The collaborator (co-inventors or their assignees) have no such obligations. The collaborator stands outside of Bayh-Dole requirements. Bayh-Dole does not force the collaborator’s title to vest with the university. Nor does it take any rights from the collaborator other than by operation of patent law regarding joint inventions. This is a reason why eminent domain issues do not arise in Bayh-Dole. There is no taking. If one argued that Bayh-Dole places ownership of inventions exclusively with the university, then one might expect there would be an eminent domain problem for collaborators who were not subcontractors or assignees.

Let’s return to the special case of federal employees as collaborators addressed at 37 CFR 401.10. In such instances, the government, should it acquire rights to the invention through its employee inventors, may do with those rights what it will. The government is not self-obligated under Bayh-Dole, nor does it gain the benefit of Bayh-Dole (if by that one means, it owns Bayh-Dole inventions outright, without obtaining title to invention by assignment and without reference to the employment agreements or IP policies it otherwise uses to obligate invention rights). In fact, if the government chooses to transfer or re-assign rights to the contractor, the requirement is that the contractor is to treat those rights as if they are Bayh-Dole: the contractor will share royalties with the federal inventor as if that inventor was one of the contractor’s own—as well as the other obligations under Bayh-Dole. One can see here, even with federal employees who co-invent, the structure of title under Bayh-Dole in joint inventions. The collaborator interest (which is undivided in the invention) is not subject to Bayh-Dole.

We can extend this understanding by looking at what happens if the agency allows the university’s inventor to “retain title.” This is addressed in 37 CFR 401.9, building on 35 USC 202(d):

(d) If a contractor does not elect to retain title to a subject invention in cases subject to this section, the Federal agency may consider and after consultation with the contractor grant requests for retention of rights by the inventor subject to the provisions of this Act and regulations promulgated hereunder.

This is about as clear as it gets. If the university already held title, and then elects not to retain it, how odd is it that the title turns out to be retained by the inventors, pending an action by the agency? One would think that if the university held title outright under Bayh-Dole, then it would be required under 35 USC 261 to transfer or assign that title by means of a writing, either to the agency (under 37 CFR 401.14(a)(d)) or to the inventors (with the agency’s approval, as one might expect for a non-qualified organization under 37 CFR 401.14(a)(k)(1)). But here we find that the inventors can simply “retain title.” It is as if the title train had never left the station. Either it’s magic, or Bayh-Dole suspends silly writing requirements for transfer and passes title around like hotcakes without formalities, or it has left 35 USC 261 unmessed with and rights to inventions are personal.

[Here’s the Supreme Court’s treatment:

Under the Act, a federal agency may “grant requests for retention of rights by the inventor . . . [i]f a contractor does not elect to retain title to a subject invention.” §202(d). If an employee inventor never had title to his invention because title vested in the contractor by operation of law—as Stanford submits—it would be odd to allow the Government to grant “requests for retention of rights by the inventor.” By using the word “retention,” §202(d) assumes that the inventor had rights in the subject invention at some point, undermining the notion that the Act automatically vests title to federally funded inventions in federal contractors.

I like how the Court also used “odd” in its account of 202(d). ]

What makes the best sense? Do you believe in magic? or same as it ever was? It is worth noting that the minimum obligations on an inventor in retaining rights to a subject invention are a substantially reduced subset of the obligations on a university as title holder. The government retains rights to obtain title in default under (d)(1) and (3), imposes a requirement to include a federal funding statement with any patent under (f)(4), and requires reporting of utilization, preference for US industry in exclusive licensing, and march-in rights if things don’t go well with utilization or licensing. That’s it. You may ask yourself, why is this? The answer might be: if inventors hold title personally, a great deal of the government’s objectives under Bayh-Dole are already met. It’s the behavior of universities to get manage title and license rights that is of more concern under the law. Inventors have it easy because they are one of the named beneficiaries of the law. They don’t even have to allocate any remaining balance after invention administration to scientific research or education, because they have, shall we say, a very generous share, entirely consistent with Bayh-Dole objectives.

What is the result of all this? Why have you read this far? The collaborator in a joint invention may dispose of its interest in the patent as it will. The collaborator’s personal interest as a co-inventor is largely aligned with Bayh-Dole’s objective of using patent rights to get subject inventions out of the cold clammy hands of federal agencies and university patent licensing offices alike and into the warm hands of anyone else, with a preference for American hands. Personal ownership outside of Bayh-Dole would appear to work at worst indifferently to government objectives under Bayh-Dole. That those persons assign their interest to companies would be in line with what the law desires—companies stimulated by patent rights to use of federally supported inventions. In a joint invention situation, there is no monopoly right, so there can be competition (another stated objective of Bayh-Dole), the rights are already with a company (preferably the collaborator is American), and there are no exclusive licenses possible without the university and the collaborator forming a relationship (yet another objective of Bayh-Dole). The government’s interest in Bayh-Dole contracting is not to secure exclusive rights to patents to universities, but to charge universities in the management of those rights in inventions they do take, to meet the obligations of the law. The university arriving late to this arrangement, stumbling through buses of policy and self-interest on the way, is something of a Spanish Inquisition, in the Monty Python skit, serving no particular purpose as to government innovation policy, but understandably out of sorts when it realizes the situation.

The shorthand for university convenience and desire to make money with monopoly patent rights is not there in Bayh-Dole. The purpose of the law is not to mandate that universities have exclusive control over patents so they can accumulate a heaping portfolio and get fat on licensing fees in some few of them. The purpose is clear: to use the patent system to promote use in American industry. The named beneficiaries are American industry, small business, scientists, educators, inventors, and the public, and of the public those particularly named in a federal award. Universities are charged with being a servant to these goals, not the primary eater at the table with both thumbs in the pie. The rewards for doing the things asked under Bayh-Dole may include royalty streams when the promotion of inventions leads to commercial sales under a license, but patents also may lead to industry standards, to development of expertise, to commons engagement to build platforms, testbeds, and libraries of tools, and other collaborations that do not result in royalty streams. These outcomes, too, have their rewards, if universities cared to recognize them.