There are three categories of invention in Bayh-Dole–inventions arising in federally supported research or development, subject inventions, and inventions owned by the federal government.
The general scope of Bayh-Dole given in 35 USC 200 is that of “inventions arising in federally supported research or development.” Bayh-Dole imposes on the patent property rights in these inventions a working requirement, a free competition requirement, and an American manufacture requirement. These requirements do not apply to patents on ordinary inventions, ones made outside the scope of Bayh-Dole, which is part of federal patent law.
Bayh-Dole introduces the definition of “subject invention” into federal patent law to address inventions made in federal contracts for research and development work. The four standard patent rights clauses established under Bayh-Dole’s authority all depend on this definition of subject invention. The federal government may acquire inventions (as from its employees) that are not subject inventions. Thus, the sections of Bayh-Dole pertaining to federal administration of inventions do not use “subject invention.”
In 35 USC 201, “subject invention” is defined as a patentable invention owned by a party to a funding agreement and conceived or first actually reduced to practice in the performance of work under a funding agreement, where the work is funded at least in part by the federal government. Bayh-Dole also includes any inventions acquired by the federal government. Thus, while subject inventions form an important part of Bayh-Dole, the Act’s scope is broader than that of subject inventions.
Under Bayh-Dole, an invention does not become a subject invention until it is owned by a contractor–a party to a funding agreement. For contracting purposes, this means that the federal government has no authority to require the delivery of an invention made in work supported by federal funds unless the invention comes to be owned by a contractor–a party to a funding agreement. This result comes about because Bayh-Dole asserts precedence over all other laws on the matter but for Stevenson-Wydler, which does not establish a general government right to invention deliverables made under contract. Since Bayh-Dole states the conditions under which the government may obtain title to inventions made with federal support, no other law can assert otherwise (without following the protocol required by Bayh-Dole), and federal agencies cannot insert any other clause inconsistent with Bayh-Dole (except by declaring exceptional circumstances).
Indeed, this result could be seen as a “gap” in Bayh-Dole. The federal government only has rights–even a non-exclusive license–if a contractor acquires ownership of an invention. Otherwise, nada. But it is not a gap. It is what the Supreme Court ruled that Congress intended. It cannot be a gap or fault. If anything, there is a “gap” in the understanding of university patent administrators and even some university law professors–a gap between the ears, so to speak.
But there actually isn’t such a gap, in Bayh-Dole at least. The implementing regulations supply the additional cleverness in the form of the standard patent rights clause, which requires contractors to make their qualifying inventive personnel parties to the funding agreement through the (f)(2) written agreement requirement. Thus, when these inventors invent, they own their inventions by common law, they are parties to the funding agreement by (f)(2), and therefore their inventions, when conceived or first actually reduced to practice in work under a funding agreement, are subject inventions.
A subject invention, then, if (f)(2) is complied with, may be owned by an inventor-contractor rather than a research host-contractor. This would all be very interesting, except (f)(2) is never complied with, so inventors aren’t parties to the funding agreements, and inventions become subject inventions only when the research host-contractors acquire them. Instead, to fill the perceived gap in Bayh-Dole, university administrators concoct all sorts of policy and employment contract and conflict of interest claims to all inventions made by university personnel–even as (f)(2) would limit their claims.
Subject inventions carry the public covenant stipulated by 35 USC 200 but differ in the patent rights clause particulars that are added to this public covenant. Nonprofit organizations have the most requirements–including a prohibition on assignment of subject inventions, a requirement to pass through the nonprofit’s patent rights clause to any assignee, and a limitation on the use of not only royalties but of any income relating to a subject invention. By contrast, small businesses do not have these requirements, and inventors, if they retain ownership of their inventions, have even fewer requirements than small businesses–but all of them are dealing with subject inventions.
Since the definition of subject invention is established by federal law, it is not the subject of private negotiation, nor of federal agency preferences. Universities cannot unilaterally assert that an invention is a subject invention, nor can inventors be made to agree that an invention is a subject invention. The presence of federal funding is not sufficient to establish that a given invention is a subject invention. Even if an invention has been supported with federal funds, if the invention is not owned by a contractor, it is not a subject invention. Similarly, an invention may be made in the performance of work funded in part by federal funds, so that the portion of the work in which the invention is made has not been funded directly with federal money but the invention is still a subject invention if other parts of the work have received federal money. It is not sufficient to isolate a portion of a broader work from the use of federal money. One has to separate the inventive work entirely from the broader work that receives federal funding.
This creates a problem for universities that claim in policy that inventions they own will be commercialized. Such a claim establishes by default that the work the federal government supports includes a commercialization step. Any company receiving assignment of a subject invention for the purpose of commercial development necessarily comes within the scope of Bayh-Dole, not just for the subject invention that it has obtained (typically through an exclusive patent license that operates as an assignment of invention) but for any other inventions the company makes in the development of the invention–that development being supported in part by the federal government, which has supported in part the overall work–research and development, as asserted by the university.
University administrators will object to this account. Even federal agencies may argue that they do not desire the definition of subject invention to have such reach. But the definition is not a matter for university administrators or federal agency administrators to opine upon–it’s a definition in patent law, built up carefully by a series of definitions in Bayh-Dole (part of patent law). The definition is what it is.
In practice, the scope of subject invention establishes the scope of the government’s license to practice and have practiced. The broader that scope, the broader the federal government’s rights in the inventions. Take for instance prescription drug development. A faculty investigator discovers a class of compounds with federal support and assigns the invention (the class of compounds, their application, methods of making them, and the like) to the university. A subject invention with potentially multiple patents. The university then assigns the invention to a company using an exclusive patent license as the legal instrument. The company becomes a party to the funding agreement by assignment, the nonprofit patent rights clause applies to the company, and any development of the subject invention is part of the performance of work under the funding agreement. By the definitions in Bayh-Dole, any inventions made by the company in developing the subject invention are also subject inventions.
The result: the federal government has a right to practice and have practiced not only the university’s subject invention but also any additional inventions made by the company in development of the original subject invention. Those additional inventions are made in the same work, and are supported in part by the federal funding. The government may authorize companies to make, use, and sell the prescription drug for any government purpose without any obligation to pay royalties or other compensation to the university or company. (Of course, if the company includes its own background rights in the formulation or packaging of the prescription drug, any government authorized making, using, or selling involving those background rights may be subject to a compensation claim–but that claim would clearly be much less than if all patents involved were subject to a claim.)
The horrific part of all this, for a company, however is not that the government can practice the final product–essentially commission generic drug alternatives for government purposes–but that the company has accepted the nonprofit’s patent rights clause when it takes assignment of the university’s subject invention (exclusive right to make, use, and sell = assignment of invention). That means the company is not free to use income relating to a subject invention any way it pleases–it may deduct only the expense incidental to the management of subject inventions and use the balance for scientific research or education.
If the inspector general audited university exclusive licenses that operate as assignments, and audited the use by assignee companies of income related to subject inventions, companies would no doubt rather give up their exclusive right to make and use in favor of retaining only an exclusive right to sell–and thus also give up the right to enforce the licensed patent(s)–to avoid being subject to the nonprofit patent rights clause and bring any subsequent inventions they make in development within the scope of the definition of subject invention.
Universities and companies have a choice in how they contract for the development of subject inventions. Universities do not have to assign subject inventions to companies in order to secure private risk capital to develop the inventions. They could grant an exclusive license just to sell, for instance. But when they make their choice to bring companies within the scope of the nonprofit patent rights clause, and the companies accept that choice, then they should be made to accept the consequences of their actions: the government has a free license to make, use, and sell the resulting drug, and the company involved must dedicate all income relating to the invention to scientific research or education–no profits for shareholders, no extra bonus compensation for executives, no new buildings, no extra funds for development.