Working through an old misrepresentation of Bayh-Dole, 2

We have dived into an old, stinky document once on-line at the University of Rochester that misrepresents Bayh-Dole requirements. The document has since come down, but its shadow remains over Rochester invention policy and practice. Thus, it is worth reading the old and stinkies to gain a sense of perspective on the new and still stinkies.

While we are at it, Rochester misrepresented (and still does) the scope of Bayh-Dole’s claim on inventions:

creation of inventions from federally sponsored grants

Bayh-Dole makes a different claim. Bayh-Dole defines “funding agreement” to include any agreement “for the performance of . . . work funded in whole or in part by the Federal Government.” (I left out the “experimental, developmental, or research” adjectives before “work” so the grammar becomes clear–the noun is work, not “research” or some such thing. The “in part” makes it clear that the “work” involved may not be funded entirely by a federal grant. Parts of the “work” may not be supported with federal funds. Doesn’t matter. What matters is a clear understanding of the “work.” We could capitalize it–Work–it should always be a defined term. One of the greatest failings in Bayh-Dole practice is inattention to the definition of the Work. The definition includes federal objectives in providing funding, specifications of the work, statements made in proposals submitted through organizations and companies, statements made by those organizations and companies with regard to their purposes and practices. If someone wants a narrower definition of Work, they have to be clear about it, express, even to the point of stating what the Work does not include–and they must do this defining as part of their proposal, before any federal award is made.

Inventions that might come within scope of Bayh-Dole, when owned by a contractor, are not just those that were made “from” a federal grant. The inventions that come within scope are any made in the performance of the indicated Work, regardless of whether the inventive activity was paid for with federal funds or with other funds or with no funds at all. All that matters is that some part of the Work received federal funds. Some part of the Work was pitched as worthy of public support. That means, necessarily, that the public (sometimes via Bayh-Dole) should have the benefit of the entirety of the Work’s outputs. As the implementing regulations for Bayh-Dole have it (37 CFR 401.1):

Separate accounting for the two funds used to support the project in this case is not a determining factor.

If separate accounting could be used, then it would be easy to shift work that appeared to be inventive off to a non-federal account and avoid federal claims on any resulting invention. That’s what University of Southern California and its inventors did in the research that led to the Mine Safety Appliances litigation–where the court made clear that USC could not extract work from its federal grant Work, do that work on the side, and then come after the federal government to pay royalties to use that side work.

Thus, inventors have to know that the Work established in their grant proposals and in the federal request for proposals and in the university’s statements of policy all contribute to the definition of the “Work” for which the university petitions the federal government to support, at least in part.

But here’s Rochester’s current guidance in its invention disclosure form:

Was extramural funding used in the course of the work leading up to this discovery? Were extramural materials used? Please list all non UR funding sources and materials. Examples include state or federal grants, foundation grants, or corporate-sponsored research. Reference grant numbers, research agreements, and Material Transfer Agreements, if known.

Bayh-Dole does not care if federal funding was “used” in the “course of the work” “leading up to” the “discovery.” Bayh-Dole’s patent rights clause scope is based on the entire Work that the government was asked to support, at least in part. A proper disclosure request is to provide documents that specify the Work that the government would have reasonably been led to believe it was funding, in part. One would ask, then, for the federal call for proposals, the proposal’s general statement of the overall Work, and the specific statement of work for which a detailed budget was presented.

The question then is: “Is the disclosed invention within the overall scope of a project that has received at least some federal funding?” It’s one thing to attempt to find the cause of a cancer. That would be science. It is another thing to attempt to create medicines to treat cancer. That would be public health. It is yet another thing to attempt to create medicines to treat cancer than can be commercialized. That’s a matter of product development. Again, it is a choice that a university makes in what it presents to the federal government as the scope of any project for which it seeks federal dollars for some part. If a university has a standing policy to seek to commercialize its inventions, then it is reasonable to expect that the federal government relies on that policy in making its decisions on funding. For instance, Rochester has this statement in its IP Policy:

University encourages the commercial development of creative works and inventions for the public use and benefit.

Rochester does not have to “encourage” commercial development. It chooses to do so as a matter of standing policy. Further, what does it mean to “encourage” commercial development here? Rochester claims to own all inventions made in sponsored research, so who exactly does Rochester need to “encourage”? Its own administrators? What a strange policy! In any case, a federal agency might reasonably view this encouragement as broadening the scope of any federal funding agreement. “Rochester encourages the commercial development of any invention made in this project as a matter of formal policy–thus, the project necessarily includes commercial development whenever Rochester acts to encourage it.”

Bayh-Dole’s definition of funding agreement makes just such extensions of scope clear:

Such term [“funding agreement”] includes any assignment, substitution of parties, or subcontract of any type entered into for the performance of experimental, developmental, or research work under a funding agreement as herein defined.

If Rochester assigns an invention or subcontracts work or substitutes parties, the funding agreement extends to include those new participants. Anything those participants do, too, comes within the scope of the funding agreement and its Bayh-Dole patent rights clause, when their employees invent and they acquire any of those inventions. Rochester’s IP policy, again:

Whenever possible, the University endeavors to make its creative works and inventions available on a non-exclusive basis; however, in many cases an exclusive license may be necessary to make it economically feasible for an enterprise to undertake commercial development and production.

We can sort out the awful import of this statement later, but here the thing is–if Rochester grants an exclusive license to all substantial rights in an invention, that’s an assignment of the invention. Per Bayh-Dole’s definitions, the enterprise undertaking commercial development and production becomes a party to the federal funding agreement–a new contractor–and any inventions acquired by that enterprise in the course of the development and production work are also subject inventions. Rochester does not have to grant exclusive licenses. It chooses to do so. An enterprise does not have to accept an exclusive license that functions as an assignment. It chooses to do so. But in doing so, Rochester and its enterprise choose to bring the enterprise within the scope of the funding agreement.

This is all made express by 35 USC 202(c)(7)(A), which requires the nonprofit assigning any subject invention to make the assignee accept the nonprofit’s patent rights clause.

Thus, it is not up to inventors to determine whether an invention has been made “from” federal funding. Certainly, the scope of interest for federal funding agreements is not any funding “leading up to” a discovery, and inventors would have no idea what meaning to give to “used in the course of” as distinct from “made in the performance of work under” a federal funding agreement. The use of money has nothing to do with Bayh-Dole’s scope of interest in the disposition of inventions. The issue–the only issue–is whether a given invention comes within the scope of work that the federal government has been asked to fund some part of. The part the federal government funds has nothing to do with the claims that Bayh-Dole makes on any invention within the scope of the project. Fund one part, a claim to any part. The core message for Bayh-Dole management, then, is to manage clearly and firmly the scope of the project for which any part seeks federal support.

Consider, for instance, “first actual reduction to practice.” This means, essentially, testing of an invention to demonstrate that it performs as conceived in each of its elements. The case behind all of this is In re Eddy L. King. King, a civilian employee of the Air Force, invents a pallet coupler. He drafts a patent application but does not file it until after the Air Force has tested the coupler and found that it worked. King had already built prototypes (and supplied them to the Air Force). He didn’t need the Air Force testing to include data in his patent application. It did not matter, for patenting purposes, whether the Air Force liked the new coupler–it would work for anyone transporting pallets of stuff. No matter. The Commissioner of Patents ruled that because King had not filed his patent application, first actual reduction to practice took place using Air Force funding, and therefore the Air Force had a royalty-free license to use the pallet coupler.

Now look at Rochester’s guidance. An invention comes within Bayh-Dole’s patent rights clause’s scope when federal funds are used to first test the invention after it has been conceived and built, even if no federal funds were used “leading up to” the invention or in building a prototype. Bayh-Dole defines “made” to include this first actual reduction to practice testing. Instead of claiming just a license to practice any such invention, where testing but not inventing made use of federal money, Bayh-Dole expands the federal claim to that of ownership of the invention if the contractor fails to timely disclose it, or to elect to retain title, or to file a patent application, or to prosecute that application to issuance, or to maintain any issued patent, or to defend the validity of any issued patent–across all US and foreign patents. Thus, Rochester’s guidance badly misses the scope of federal interest in any invention.

It’s not a matter of following the federal money. It is not a matter of what “leads up” to inventing. It is not a matter of whether federal money has been used–the proverbial one dollar touching a research project. It is, rather, a matter of having a clear scope of an overall project, seeing that federal funding has supported at least some portion of that project, and that a given invention was made within the scope of the project, not merely in that portion supplied with federal money.

Bayh-Dole doesn’t ask for the contract, but the standard patent rights clause does:

The disclosure to the agency shall be in the form of a written report and shall identify the contract under which the invention was made and the inventor(s).

Here, “invention” means a subject invention–one “of the contractor.” Further, “made” means “conceived or first actually reduced to practice”–fully mentally realized and recognized as inventive, or tested before filing a patent application. “Under” means “in performance of experimental, developmental, or research work”–at least some portion of which is funded by the federal government. “Contract” means “funding agreement” including any extension of the funding agreement by any assignment, substitution of parties or subcontract of any type.

For the university to prepare a written report, it has to determine that a disclosed invention–which the university has acquired–has been “made” “under” “contract.” That is, the university must document the circumstances. It is not sufficient to ask inventors to make something up. It is not sufficient to decide an invention is within scope of Bayh-Dole “just in case.” Such things are non-compliant. Imagine if the federal government contested every report of a subject invention. “Prove it!” the government demands. What would a university have to show?

Specify the Work. Identify documents that support that scope.

Did the federal government support some part of the Work? Identify the grant or contract.

Was the invention conceived, or, before a patent application has been filed, first built/used and tested in the Work? Identify invention records that support the claim.

For what it’s worth, I have had inventors disclose inventions claiming federal funding before the federal funding has even been awarded–in anticipation of federal funding. Well, it’s possible that the federal funding could support “first actual” testing–but that’s a fact yet to happen, a choice yet to be acted upon. I have had inventors cite federal funding because they have had federal grants in the past, and what they learned in doing that work they have applied to a present invention–but the invention was not anticipated in that past federal Work, was not “made” in that Work.

It is a tough lesson, that the Work may be much broader than the part the federal government funds. Certainly, the Work is broader than federal funding when a university blithely adopts a formal policy of owning all inventions made with its resources or in sponsored research and seeking to commercialize all such inventions. And the Work is certainly broader than federal funding when a university licenses an invention exclusively to a company, transferring all substantial rights in the invention–to make, use, and sell–with the right to enforce any patents, even if reserving the right to use the invention for non-commercial purposes and even if reserving government rights. Fortunately, the federal government does not enforce Bayh-Dole. University administrators can make up most anything they want about what inventions must be disclosed and federal agencies don’t care.

We might say, it doesn’t matter as far as the federal government is concerned. There may as well not be a law at all. But it does matter. First, it matters to inventors–that they have their inventions taken from them by university administrators (which may violate university policy and may breach Bayh-Dole’s patent rights clause–ah, but so what there). Second, it matters because inventors then cannot practice their inventions outside the university that claims those inventions without the permission of administrators. Third, it matters because university administrators routinely dedicate inventions they acquire to exclusive licensing efforts and so refuse to give permission to inventors or anyone else to practice such inventions unless they make a commitment to commercialize them–and even then, university administrators demand that the commercialization must involve an exclusive license. If someone got a non-exclusive license first, so the reasoning goes, then the university could never offer an exclusive license. Thus, they don’t offer non-exclusive licenses except under duress–such as when a big grant requires it–and then only to companies involved in the research.

Bayh-Dole non-compliance by university administrators ends up suppressing access to inventions–even access by inventors–and suppressing competition–even in the form of competing research and uses that would compete with efforts to require exclusive rights based commercialization. In turn, suppressing access and competitive use reduces the chances of adoption, of cumulative technology, of standards, of interoperability, of the usefulness of inventions made in work receiving federal support. If such inventions are not of much use anyway, then there’s not much to worry about. But if those inventions do matter, or could, or ought to, then there’s forty years of misrepresentation and non-compliance to deal with.

This entry was posted in Bayh-Dole and tagged , , . Bookmark the permalink.