[Y’all are reading this one. I’ve made a few edits for clarity. Where it matters I’ve left some strike-throughs.]
A search query here at Research Enterprise recently asked how to determine if an invention is a “government subject invention.” Here is your answer. Keep in mind that Bayh-Dole is poorly drafted, not enforced, not complied with, and there’s plenty of misinformation available that is intended to keep things that way. Universities and academics, in particular, consistently misrepresent the law in policy, in guidance documents, and in articles. But hey, people still want to know what subject invention would be if anyone bothered to follow the law. So here it is.
Here are the key points. For those who want more discussion, that follows.
1. Key points
“Subject invention” is a statutory definition in the Bayh-Dole Act (35 USC 201(e)). An invention is a subject invention when it meets every element of the definition. No one–not an inventor, not a contractor, not a federal agency–can simply declare an invention to be a subject invention. But contractors and inventors can take actions that determine whether an invention meets the definition of subject invention.
The invention must be owned by a party to the funding agreement.
Contractors may receive assignment of the invention or may make inventors parties to the funding agreement. If an invention is not owned by a party to a federal funding agreement, then it cannot be a subject invention. Nothing in Bayh-Dole’s requirements for contractors applies if an invention is not a subject invention. Bayh-Dole’s policy requirements may still apply, but that does not depend on an invention being a subject invention. See Stanford v Roche.
The invention must be patentable or possibly patentable.
If the inventor does not recognize work as inventive, then it is not patentable. If the invention is not directed at patentable subject matter, it is not patentable. If the invention is not new, useful, and non-obvious, it is not patentable. If the invention has been disclosed publicly for more than a year, it is not patentable. If the invention has not been described in sufficiently complete detail or actually reduced to practice and shown to work in each of its elements as claimed, it is not patentable. If an invention is not in a condition to be patentable, then it is not a subject invention.
The invention must have been made–conceived or testing completed–in a project that has received at least some federal funding for research, experimental, or developmental work.
Wording here is important. Define a project. If some part of that project receives federal funding, then any invention made in the project is a subject invention if it meets the other elements of the definition. Conception is more than just having an idea or insight–it involves having in mind every element of an invention. It is difficult for conception to be “funded” unless one is hired to invent or a project expressly requires invention to succeed. First actual reduction to practice means testing necessary to demonstrate that an invention satisfactorily performs each function as conceived. An invention is not “made” if it has merely been “conceived.” Thus, what someone has thought up cannot be a subject invention because it has not been reduced to practice–either by description as in a patent application or by actual reduction to practice to demonstrate that what has been thought up actually works. Since a subject invention must be “made” mere conception is not sufficient.
Any party to the funding agreement may make a subject invention, even one recruited to develop a commercial product, if the recruitment takes the form of assignment, substitution of parties, or subcontract.
The scope of the definition of subject invention depends on the scope of the funding agreement. Contractors may add parties to the funding agreement, and these parties are also contractors. Everything in the law that applies to the initial contractor also applies to each additional contractor as each becomes a party to the funding agreement.
Now for discussion. Read on, if you are curious and brave of heart.
2. Practice
Most universities ignore the definition of subject invention in the law. Many claim that any invention made with federal support is a subject invention. That’s simply not true. Bayh-Dole does not care if an invention is “made” “with” federal support. It is not whether federal money is used in messing with inventive stuff. Rather, it is whether the project in which an invention arises has received federal support. What is the project that the invention arose in? Did that project receive federal support? Is the invention possibly patentable? Does a contractor own the invention?
The university ignorance goes the other way, too. Even if federal money is used to mess around with inventive stuff (building a prototype, say), that does not necessarily matter. What matters is if the invention is within the scope of the project that receives federal funding. If the invention is not within scope, then the use of federal money doesn’t change that. A project does not bend to become anything that someone spends federal money on. It may be that the federal money has been misused, but that doesn’t not change the definition of subject invention. The important point is that if the federally funded work is not distracted or diminished by spending federal money on an invention outside the scope of that work, there’s no basis for finding that the invention is a subject invention. See 37 CFR 401.1 for the regulatory discussion. Better hurry before NIST removes it.
One more bit. It also does not matter that no federal money is spent “on” a given invention. If the invention is within the scope of the project that has received federal funding, then even if all the fussing with the invention takes place on a non-federal budget, the invention, if owned by a contractor and patentable and all, is a subject invention. Some time before Bayh-Dole, USC attempted to move an invention made in a federally funded project to a side account, patented it, licensed it exclusively to a company and the company then sued the federal government for royalties for using the invention. Well, ho ho, it turned out that the court determined that the government had a royalty-free license to use the invention because the invention was within scope of the project, even if USC tried to avoid spending federal dollars on it. See Mine Safety Appliances v United States.
Other universities assert that they call inventions subject inventions “just in case” they might be. That’s malpractice and does not conform to the law. Additionally, most universities refuse to comply with the standard patent rights clause requirement that makes inventors parties to the funding agreement (37 CFR 401.14(f)(2), 35 USC 201(b) and (c)), and so inventors are not parties and their inventions, until assigned to a contractor, are not subject inventions, even if “made” in a project receiving federal support.
Finally, universities routinely assign subject inventions to companies under the label of an exclusive patent license. An assignment of all substantial rights in an invention (make, use and sell) conveys ownership of the invention. Assigning a subject invention makes the assignee company a party to the funding agreement by operation of 35 USC 201(b)–and for nonprofits, the controlling patent rights clause covering the company assignee is the nonprofit’s patent rights clause–see 35 USC 202(c)(7)(A). Inventions made by the company within the scope of the purpose of the patent license the project receiving federal funds then also are subject inventions, when owned by the company-cum-contractor. University administrators and their attorneys will deny that they are in breach of the standard patent rights clause (and the law), but they are. As one director of technology transfer told me when I asked about a clear instance of non-compliance, “We will comply with that when a federal agency makes us comply.” He didn’t care if the federal agencies didn’t care.
All this leaves us with something really simple and non-compliant:
A subject invention is anything that anyone calls a subject invention. If no-one–inventor, contractor, federal agency–calls something a subject invention, then it is not a subject invention (even if it actually is!). If any one does call an invention a subject invention, then it is next to impossible to show that it is get those involved with compliance with Bayh-Dole to admit that it is not a subject invention. (See the problems UCLA had after asserting its nicotine patch was a subject invention.) If an inventor and contractor do not call an invention a subject invention but later a federal agency does, then there’s a convoluted procedure in Bayh-Dole for administrative actions, determinations, appeals, and the like, followed by litigation, to resolve the claim. Any hint of such fuss is sure to destroy any licensing market for the invention in dispute, because if the federal agency wins on its claim, then it has the right under NIST’s new regulation, to demand title to the invention (and any patents) as a remedy for the failure to disclose timely the invention as a subject invention.
Some universities then demand inventors report as a subject invention any invention they make in a lab that has federal funding for any purpose, regardless whether the invention has anything to do with the federal funding. Other universities instruct inventors to identify any grants associated with the invention without indicating whether those grants are past or current or anticipated. I have had inventors report inventions as made with federal funding they have not yet received because they want to show how productive they are with their federal funding. Not necessarily subject inventions. Depends on the project, the scope of federal funding, and actual reduction to practice.
Thus, university practice paradoxically both over-reports and under-reports subject inventions. They report inventions that are not subject inventions as if they are–because they believe that if an invention is a subject invention, then they have a right to demand ownership of it and are exempt from federal agency demands for ownership of it. And they fail to report inventions that are subject inventions–because they misconstrue the law’s definition to mean that the work on an invention has to receive federal funding, even though that is not the definition and not the guidance in the implementing regulations.
Thus, they create the fictions that an exclusive patent license does not assign the underlying invention (it does if the exclusive license covers all substantial rights in the invention–make, use, and sell) and that therefore any exclusive licensee/assignee is free of all obligations but for an exposure to reporting on invention utilization, U.S. manufacturing (which does not apply to assignees), and march-in for failure to achieve practical application or satisfactorily meet public health needs or regulatory requirements, or failure to comply with U.S. manufacturing provisions.
In practice, a subject invention is anything a contractor says it is, unless a federal agency claims otherwise. That’s not the law–but that’s the defacto federal policy, ignoring the law. Put it another way. Take any one invention you, a federal contractor, own. If you sued the federal government in the Court of Federal Claims for compensation for its use of the invention, would the federal government be able to establish under any federal statute and by means of any documentary record that the federal government had at least the right to a royalty-free non-exclusive license? If so, then the invention is a subject invention. Why? Bayh-Dole has to be as broad as the broadest claim made by any federal statute on any invention if Bayh-Dole is to preempt such statutes on matters pertaining to federal government rights to inventions made in any funding agreement. That’s how Bayh-Dole is drafted. That’s why the rest of this article is so long and complicated.
Put it yet another way. Federal agencies don’t give a rodent’s cheeks about what’s a subject invention so long as you don’t show them up. Federal contractors are expected to want to make inventions subject inventions 1) to avoid vestigial federal agency claims left in various statutes and executive branch patent policy; and 2) to establish a demand for ownership directed to their inventor-employees based (entirely misrepresented, but they do this) on federal law entitlements. For universities with policies assuring faculty of freedom of research and freedom to publish, plus full enjoyment of their Constitutional rights (right to patent is one), this use of Bayh-Dole is especially important because it provides a (superficial, clever but corrupt) basis to suspend such policies “to comply with federal law.”
3. The law
Now consider the law. This part ends up being fantasy relative to practice, even as practice is administrative fantasy with respect to the law. Rule of law, schmaw.
Let’s deal with the structure of Bayh-Dole. Bayh-Dole is part of federal patent law. Bayh-Dole defines a new category of invention–the subject invention–and announces conditions and requirements that are specific to subject inventions. When a patent is obtained on a subject invention, Bayh-Dole stipulates the special property attributes that attend such patents. Thus, a Bayh-Dole patent is not any ordinary patent–it does not have the attributes of any ordinary patent. Bayh-Dole then does not merely place requirements on owners of subject invention but also alters the attributes of the patents that the federal government issues on those inventions.
You can see that subject invention matters, then, in a big way, at least in theory. In practice, people ignore all these distinctions and merely assert that if something is a subject invention, then they have the right to demand ownership of it, a mandate to attempt to commercialize (or at least monetize) it under a patent monopoly (or, for the delicate–an exclusive position provided by patent), and excluding all others, provided they make a show of complying to the parts of the law on which they base their right to claim ownership while ignoring all those parts that have to do with the special conditions imposed on their exploitation of subject inventions. This may sound cynical–it’s not–it’s pragmatic. This is what happens. You may as well hear it up front if you don’t already know it from your own practice.
Bayh-Dole did not repeal past laws and executive branch patent policy. Instead, it adds another administrative layer that preempts the other layers when a contractor acquires ownership of a conforming invention.
Bayh-Dole applies when a contractor owns an invention that is or may be patentable (or is eligible for plant variety protection) and that has been made in the performance of work under a federal funding agreement. 35 USC 201(e). An invention that meets this definition is a subject invention.
Bayh-Dole does not apply if an invention (or anything else) fails to meet each element of the definition. Thus, the definition is rather crucial. If Bayh-Dole does not apply, then federal statutes pertaining to invention ownership in areas of federal government activity and contracting–ones that Bayh-Dole otherwise preempts–then apply. Those statutes, however, do not deal with “subject inventions.”
If none of those statutes apply, then executive branch patent policy–the Kennedy policy as modified by Nixon and then by Reagan–applies. However, the contracting mechanism by which that policy was codified in the Federal Procurement Regulations was dismantled and replaced by the Federal Acquisition Regulations, which implement only Bayh-Dole. Thus, if Bayh-Dole does not preempt federal statutes and executive branch patent policy, and no federal statute applies, then in effect nothing applies and the government has no basis to claim an interest in or control of invention rights in whatever has been made.
Again–If an invention does not meet the definition of subject invention, then Bayh-Dole does not apply Bayh-Dole’s provisions pertaining to contracting for research and the like do not apply or preempt federal statutes on matters of federal contracting for an interest in inventions made under federal contract. Other things might apply–including 35 USC 200, but most of the apparatus for those other things has been dismantled and abandoned.