Why is Bayh-Dole a part of federal patent law? It concerns how federal agencies contract for research, not (apparently) how patents are awarded. It doesn’t make any sense for Bayh-Dole to be in patent law. Instead, why wasn’t Bayh-Dole made a part of the Federal Grants and Cooperative Agreements Act of 1977? But since Bayh-Dole is part of patent law, then we might expect that Bayh-Dole makes patents on subject inventions different from patents on ordinary inventions.
Let’s have a look.
Here’s the original statement of purpose for the Federal Grants and Cooperative Agreements Act (things got revised in 1982, two years after Bayh-Dole):
(b) The purposes of this Act are—
(1) to characterize the relationship between the Federal Government and contractors, State and local governments, and other recipients in the acquisition of property and services and in the furnishing of assistance by the Federal Government so as to promote a better understanding of Federal spending and help eliminate unnecessary administrative requirements on recipients of Federal awards;
(2) to establish Government-wide criteria for selection of appropriate legal instruments to achieve uniformity in the use by the executive agencies of such instruments, a clear definition of the relationships they reflect, and a better understanding of the responsibilities of the parties;
(3) to promote increased discipline in the selection and use of types of contract, grant agreement, and cooperative agreements, and to maximize competition in the award of contracts and encourage competition, where deemed appropriate, in the award of grants and cooperative agreements; and
(4) to require a study of the relationship between the Federal Government and grantees and other recipients in Federal assistance programs and the feasibility of developing a comprehensive system of guideline for the use of grant and cooperative agreements, and other forms of Federal assistance in carrying out such programs.
The FGCAA (“fig-caw”) established (overtly) simple rules to determine whether the government should use a grant, contract, or collaborative agreement. A contract was to be used for procurement, a grant for providing support for a public purpose, and a collaborative agreement when the work involved substantial involvement by the federal government.
In the FGCAA, we have elements that we find in Bayh-Dole: matters of sponsored research, especially at universities; a need for uniform policy that’s government wide; and rules on acquisition of property and services by a contractor relative to the interests of government. The definition of “funding agreement” in Bayh-Dole is essentially that of the “contracts, grants, and cooperative agreements” set out by the FGCAA, with the Tennessee Valley Authority excepted. We find one element, as well, that is not present in Bayh-Dole: a distinction between federal procurement and assistance (or subvention). We will come back around to this distinction.
Here is what the FGCCAA has to say about tangible property:
The authority to make contracts, grants, and cooperative agreements for the conduct of basic or applied scientific research at nonprofit institutions of higher education, or at nonprofit organizations whose primary purpose is the conduct of scientific research shall include discretionary authority, when it is deemed by the head of the executive agency to be in furtherance of the objectives of the agency, to vest in such institutions or organizations, without further obligation to the Government, or on such other terms and conditions as deemed appropriate, title to equipment or other tangible personal property purchased with such funds.
Here we have a focus on nonprofit organizations, just as there is in Bayh-Dole (but for the added element of small businesses in anticipation of the SBIR program–but even the addition of small businesses in Bayh-Dole is powerfully strange). Here in the FGCAA, the authority to allow tangible personal property purchased with federal funds to be owned by a nonprofit is a matter of agency “discretionary authority.” The FGCAA delegates the decision to the agency head, consistent with “the objectives of the agency.” An agency head may require government ownership, allow contractor ownership, or set out some other arrangement that provides the government with whatever it needs–access and use, say, or stipulating how others may use the equipment (or whatnot), or what happens when the equipment is no longer wanted by the contractor.
For federal grants with nonprofits, 2 CFR 200 implements uniform regulations. Here’s what 2 CFR 200.313 has to say about equipment:
Subject to the obligations and conditions set forth in this section, title to equipment acquired under a Federal award will vest upon acquisition in the non-Federal entity. Unless a statute specifically authorizes the Federal agency to vest title in the non-Federal entity without further obligation to the Federal Government, and the Federal agency elects to do so, the title must be a conditional title. Title must vest in the non-Federal entity subject to the following conditions:
(1) Use the equipment for the authorized purposes of the project during the period of performance, or until the property is no longer needed for the purposes of the project.
(2) Not encumber the property without approval of the Federal awarding agency or pass-through entity.
(3) Use and dispose of the property in accordance with paragraphs (b), (c) and (e) of this section.
The “conditions” of the section focus primarily on the priority of use of equipment by the recipient, proper inventory accounting for equipment, and what happens when equipment is no longer needed by the recipient. For instance, a recipient cannot charge less for private use of the equipment than private sector rates. Another condition, interestingly enough, is that the “non-Federal entity” recipient of a grant (in Bayh-Dole’s terms, “contractor”) must allow the project to continue to use the equipment:
Equipment must be used by the non-Federal entity in the program or project for which it was acquired as long as needed, whether or not the project or program continues to be supported by the Federal award, and the non-Federal entity must not encumber the property without prior approval of the Federal awarding agency.
This is a regulation that protects projects. A federally funded project acquires equipment. That project has priority use of the equipment, even after the federal grant has ended, for as long as the project has need of it. A grant recipient cannot “encumber” the equipment for some other purpose without federal approval. That’s a nice check on recipient behaviors–and suggests that recipients may have had a problem with how they were managing equipment acquired with federal funds.
Thus, in the FGCAA, we see a similar dichotomy in ownership of tangible property to that of Bayh-Dole. The federal government may own tangible property, or if the recipient owns the property, then the federal government by default must impose conditions on that ownership, including how the property is used in the project, made available to others, and disposed of when no longer needed. There are uniform procedures, but federal agencies have discretion to direct how valuable equipment (worth $5,000 or more) is surplussed.
When equipment is purchased with federal funds, a basic question is Who owns the equipment? If I provide my money so you can buy a car, it’s my money that you used, so it’s my car unless I say otherwise. If I *give* you my money, so that it’s then *your* money, then when you buy the car, it’s *your* car, because you used *your* money. And here we have one of the odd things in federal contracting–that when a university receives federal money to support a project, the money is still the federal government’s money. The money may be used by the university to pay salaries, cover overhead, and acquire stuff–but the money itself is not the university’s money. If it were, then there would be no issue on where title to equipment vests upon purchase.
Even here, though, the idea of “title” is a bit odd. A recipient might own “title” to equipment, but only so long as the recipient wants the equipment and the equipment has a value greater than $5,000. After that, it’s up to the federal government how to dispose of the equipment–and if the equipment is sold, the federal government is to get a share of that value back. Here “title” means something along the lines of “conditional quiet possession” rather than outright ownership. It’s more like a no-cost lease. The federal government still has an “interest” in the value of the equipment is used, and a “right” to specify how the equipment is disposed of when no longer needed. As the implementing regulations make clear, unless a statute provides otherwise, title is always “conditional.”
In Bayh-Dole, we find something similar. “Title” to inventions made with federal support may be acquired by a contractor, but the federal government requires through a patent clause in its funding agreements rights in any such subject invention unless the agency finds “exceptional circumstances,” in which case rights can be most anywhere the agency decides, based on its purposes. The standard patent rights clause used by federal agencies stipulates how the patent property right in a subject invention may be used by a contractor as well as how money arising from the use of the subject invention is used:
the balance of any royalties or income earned by the contractor with respect to subject inventions, after payment of expenses (including payments to inventors) incidental to the administration of subject inventions, be utilized for the support of scientific research or education
That’s a broad statement–not just licensing income (“royalties”) but also any other “income earned.” And not just from patents on subject inventions, but any income earned “with respect to subject inventions.” So Bayh-Dole requires that federal agencies must contract to control not only the use of subject inventions and patent property rights in those inventions but also the use of any money arising from subject inventions or patent property rights in those inventions. “Title” may be with the contractor, but certainly not the exclusive right or interest. And even title is conditional–if the contractor fails to report an invention or does not file patent applications or does not maintain a patent, then the government contracts to have the right to acquire title. Again, it’s all very similar to the FGCAA for tangible personal property. Subject inventions under Bayh-Dole are not ordinary inventions–conditions apply as set out by Bayh-Dole’s statement of policy as well as that policy is interpreted by the implementing regulations and standard patent rights clause.
But inventions are not like equipment or even other intangible assets, such as copyrights. An employer has no equitable interest in an invention made by an employee, even within the scope of the employee’s employment, as a matter of employment. There must be some other agreement regarding inventions. Here’s the Supreme Court in Stanford v Roche:
But, as noted, patent law has always been different: We have rejected the idea that mere employment is sufficient to vest title to an employee’s invention in the employer.
Similarly, merely contracting with another company for research services does not give the commissioning party title to inventions made by the contractor’s employees–absent an agreement regarding their inventions, they don’t have any obligation to assign title to either the contractor or the commissioning party, and title certainly does not “vest” with the commissioning party–even when that party is the federal government. The federal government, like any other party funding research, must contract for rights to inventions. That’s what the (f)(2) agreement requirement does in the standard patent rights clause. It is not a requirement that inventors assign to their university employer, but rather is indifferent to the relationship between a university and its employees–especially its faculty “employees.” Instead, the (f)(2) agreement establishes an agreement between the federal government and any potential inventor working within the scope of a funding agreement with regard to patentable inventions. When a patentable invention is made, the (f)(2) agreement conditions go true, and the federal government has conditional rights in the invention, pending whatever it is that the inventors choose to do with their rights.
Bayh-Dole in a nutshell:
Federal agencies must use a standard patent rights clause that requires research hosts to pass through to potential inventors a requirement that they make a commitment to establish the rights of the United States in patentable inventions made in the performance of work under a funding agreement.
That’s the fundamental agreement, from statute to regulations to funding agreement to standard patent rights clause to (f)(2) written agreement. Everything else that matters arises as a result of agency and inventor choices. Federal agencies may declare exceptional circumstances, and then things are different. Inventors may choose to assign their inventions to their employer, and if so then conditions of the patent rights clause kick in depending on whether the employer is a small business or a nonprofit. Or inventors may choose to retain their rights, in which case the federal agency may decide whether to request assignment of title or allow the inventors to retain their rights, subject to the fewest conditions of all possible owners of a subject invention.
Thus, for inventions, “vesting” follows common law–ownership is with the inventors. Only if a university acquires title to a subject invention does title “vest” in the university, and then only conditionally. It matters, further, how a university acquires title. If it acquires title as a condition of inventors using the resources provided by the federal award, then it is acquiring or improving the invention “with a Federal award” (2 CFR 200.316)–and the intangible property trust relationship requirements of the federal funding agreement kick in. In a federal funding situation, it takes a great deal of law-bending to argue that the university has an equitable interest in inventions. The federal government provides the funding, pays the salaries, pays the expenses, pays the overhead–the university is not out anything for the effort, but for the loss of its personnel for some portion of their “official duties”–and even then, the university releases its personnel to work on the federally funded project.
It’s not even so clear (despite the insistence of university attorneys) that a university policy statement asserting that inventors have agreed to assign their inventions to the university actually operates as claimed, or even if it does, supersedes the federal obligations that a university agrees to require inventors to undertake as a matter of the standard patent rights clause. A university can’t release inventors to make an agreement with the federal government regarding the establishment of the government’s rights and simultaneously insist that it still has an agreement with inventors that requires inventors to assign their rights to the university. It doesn’t work. The federal obligations take precedence. No wonder universities routinely refuse to comply with the (f)(2) requirement and are working to turn (f)(2) into a university assignment clause. It’s just that there is no authority in Bayh-Dole for doing so. As the Supreme Court put it,
The Act’s disposition of rights—like much of the rest of the Bayh-Dole Act—serves to clarify the order of priority of rights between the Federal Government and a federal contractor in a federally funded invention that already belongs to the contractor. Nothing more.
If Bayh-Dole clarifies rights *after* an invention has been acquired by a contractor, then there’s nothing in the law that vests title with the contractor, gives the contractor the right to take title, or requires the contractor to acquire title. Nothing more means something–there is no authority in Bayh-Dole to require federal agencies to require that universities acquire title to inventions made with federal support. Agencies might make this requirement by determining “exceptional circumstances” of course, but that requires a determination procedure to establish just why it is that inventors should be stripped of their ownership of inventions as a matter of federal contracting to support research that they have proposed in the public interest. Why should inventors be forced by the federal government to assign their inventions to a research host–which may have no capability to manage inventions, may have interests that conflict with those of the investigators and their project, and which may not even want to own inventions (though this apparently is much less likely these days–all the more reason for Bayh-Dole to build in contracting protections for inventors and projects relative to the claims of a research host)? There’s no reason.
All the federal government needs is an agreement to acquire rights in such inventions for its purposes–either a non-exclusive license or assignment of title. (And one wonders, in research supported by subvention rather than procurement, why the government needs even a non-exclusive license so broad as set out–“to practice and have practiced”–why not just restrict the right of patent owners of subject inventions to bring claims of infringement against the “United States”? That would be simpler, involve a whole lot less administrative paperwork, and provide the government with most everything it needs–and it still has “march-in” rights for most anything else. Regardless, Bayh-Dole deals with what happens after an invention becomes a “subject invention.” It does not stipulate how any given contractor acquires title to such an invention. But once a contractor does acquire title, the patent rights clauses created under Bayh-Dole’s authority stipulate the conditions on which a contractor may retain title.
It’s not clear, then, why Bayh-Dole is in patent law and not in the FGCAA. In the FGCAA, Bayh-Dole would not define a new category of patentable invention but would rather do what the IPA program did–define subject invention by contract as a conditional research deliverable, just as the FGCAA does for equipment and other tangible inventions. If the contractor acquired title to a subject invention (which would now be defined in terms of the scope of the agreement and not be concerned with whether a contractor owned the invention), then the various conditions on that ownership would apply as a matter of the federal contract, along the lines of 2 CFR 200.315(c)–where the implementing regulations for Bayh-Dole are incorporated by reference. So why isn’t Bayh-Dole in FGCAA to start with?
Bayh-Dole is part of federal patent law, not part of federal law regarding contracts. Bayh-Dole changes patent law. The definition of “subject invention” changes patent law. The statement of “policy and objective” (not merely “purposes”) changes the scope of patent rights available to a patent owner. The conditions federal agencies must place on subject inventions further limits what a patent owner of a subject invention may do, not merely as a matter of federal contracting but also as a matter of federal patent law. Something is going on. It’s clear that subject inventions are not ordinary inventions, and patents on subject inventions are not ordinary patents, and title to patents on subject inventions is not ordinary title, and a patent owner’s rights to assign, license, enforce, and abandon a patent on a subject invention are not ordinary patent rights.
If Bayh-Dole were in the FGCAA, then patents on subject inventions would be ordinary patents. Patent law would not change. What would matter then would be the invention deliverables defined by the funding agreement, and it would make sense, as it does for acquired equipment, that the federal government could set out the conditions upon which the government required delivery of title to inventions, a non-exclusive license, or no right to practice, or no conditions whatsoever on how the contractor used, inventoried, or disposed of a subject invention. Bayh-Dole would then purport to make those conditions “uniform” (or arbitrary, depending on your value system).
But Bayh-Dole isn’t in the FGCAA. It’s in patent law, and that makes a great difference in what it does, how it does it, and what patent owners of subject inventions can do. It’s just that most such patent owners are in denial about it all. They want the law to confirm that whatever they do is proper and aren’t willing to commit to follow the law where it limits their actions. They refuse, even, to read the law for what it says rather than what they want it to say, demand that it says. And that’s the nub of the problem of Bayh-Dole. Folks refuse to read it. Folks refuse to enforce it. And what we get is wholesale looting of invention rights, misrepresentation of the law, and misrepresentation of the effects of the law. One might say “it is what it is”–and that’s well and true, but it’s not Bayh-Dole.
[Postscript. Here’s Ashley Stevens’s account of why Bayh-Dole ends up in federal patent law and not where it should have been:
Then Bruce Lehman, who was on Kastenmeier’s staff and who would one day become Commissioner of the U.S. Patent and Trademark Office, called Allen with a deal. The House Judiciary Committee, which Kastenmeier chaired, had passed out an Omnibus Patent Bill. Kastenmeier would add the provisions of Bayh–Dole to his bill in the House if Bayh would agree to accept
the other parts of the House bill affecting the operations of the Patent and Trademark Office. Bayh had competing bills in the Senate on these provisions but Allen accepted the deal.
Allen here is Joseph Allen, a staffer for Senator Bayh. Representative Kastenmeier (D-Wisc) is from WARF’s home state. Howard Bremer of WARF claimed to have worked closely with all involved to get Bayh-Dole passed. So Bayh-Dole gets pinned to federal patent law because it couldn’t get passed on its own. But in becoming part of federal patent law, Bayh-Dole’s definitions and policy also become patent law. That’s a huge change, really.]