New Mis-Guidance on Bayh-Dole for Universities

There’s a very nice guide on the internet to managing federal grants, “A Guide to Managing Federal Grants for Colleges and Universities [now behind a paywall, apparently].” The Guide is published by Atlantic Information Services and looks like it was published this year. The primary author is a research programs officer at a major public university. The Guide does a good job describing intellectual property and condensing federal requirements into understandable segments. But when it comes to Bayh-Dole, the Guide makes a number of mistakes that contribute to the continuing mischaracterization of Bayh-Dole and its proper implementation.

The Guide focuses on the conditions for a patentable invention, pointing out the new, useful, and non-obvious components of patentable subject matter. That’s fine, but the Guide misses the conception plus reduction to practice plus recognition by the inventors that their work involves an inventive element. (An inventor doesn’t go–is it new, useful, and non-obvious? Oh, it’s an invention! One goes–look what I thought up–it works! It’s an invention! Now, is it patentable?) Conception and reduction to practice are key elements that determine when an invention has been made, and therefore also when an invention is potentially within scope of the standard patent rights clause requirements. For the management of inventions made in the context of federal funding, the critical term is “subject invention”–

The term subject invention means any invention of a contractor conceived or first actually reduced to practice in the performance of work under a funding agreement

–plus stuff about plant variety protection. An “invention” is one that “is or may be patentable.” That’s a nice restriction that most university patent policies these days ignore. The phrase “of a contractor” means “owned by a contractor.” That is, an invention is not subject until it is owned by a contractor–that is, by a party to the funding agreement.

(Unless the contractor has implemented the (f)(2) written agreement requirement in the standard patent rights clause, then no invention will be a subject invention unless the contractor finds a way to obtain assignment of each invention (except in at least Florida and Ohio, where there are state laws that decree that public universities own certain inventions). If a university has obtained a patent agreement with a present assignment, that might do the trick, too–it’s just a question of whether the (f)(2) written agreement requirement–even if not implemented by a university contractor–preempts the operation of these other strategies, or whether a failure to comply with (f)(2) renders these other strategies void–but all this is for another time).

To be “made,” an invention must be conceived (all elements of the invention held in the inventor’s mind), reduced to practice (practiced to show all elements function as expected or described so one can see that all elements will so function, or described as in a patent application so that one with ordinary skill in the art might practice the invention without undue experimentation), and recognized as an invention by the inventor. Once an invention is made, it becomes “subject” if either the conception or the practice of the invention was within a funding agreement’s “planned and committed” activities. That is, the written statement of work anticipates either conception or first actual reduction to practice. If not, then the invention is not a subject invention, even if closely related to the federal funding, or even if benefiting from ideas or resources made available by the federal funding unless the work leading to the invention “diminishes or distracts” from the federally funded project. [Update: there is more to this determination–and that has to do with looking forward, rather than back in time, and turns on the meaning of federal funding “in part” supporting work.]

Thus, it is *not* the case that one merely asks whether federal funding somehow contributed to the invention. The relationship between federal funding and invention may be easy to construct–“we were working on understanding this phenomenon and we realized that there was an application for it.” But in such a case, the federal grant was directed at understanding the phenomenon, not on its application. The realization of an application was not specified in the statement of work and cost the grant nothing. It’s not a subject invention (though there are attorneys willing to take money to argue otherwise, and given the courts, they may well win their point there).

University invention disclosure forms almost never make such a distinction. University invention forms typically invite inventors to broaden the range of inventions that might be reported as “subject.” Here’s how Illinois puts it in their form:

Illinois disclosure

The form asks for everything “contributing to the research that led to the invention.” This request is *much broader* than the scope the definition of subject invention. Further, acknowledging funding in a publication may trace back to work that set the stage for the invention, work that funded someone who also collaborated on the invention. As the form’s instruction puts it, “Be liberal in the interpretation.” This advice is entirely inconsistent with the federal definition of subject invention and the guidance regarding scope set out in the implementing regulations at 37 CFR 401.1.

One reason that university administrators choose to misconstrue the definition in order to broaden what is reported as a subject invention is that for a long time they also misconstrued the Bayh-Dole Act to grant the university ownership of anything reported as a subject invention. Many universities relied (and continue to rely) on a savings clause in their research policies that requires all university personnel to comply with the terms of any research contract and that these terms supersede any conflicting terms of patent policy.

Thus, if a university patent policy might be narrow in its university claims or ambiguous as to what is claimed or liberal in its treatment of inventors, then the research savings clause could be invoked by administrators to assert university ownership. Encouraging a liberal interpretation of what has “contributed” to an invention amounts to trapping inventors into thinking they are obligated to assign their invention to the university. If the inventors recited federal funding, then the university administrative claim was that assignment of the invention was required by federal law. Such a claim is untrue. No assignment to the university is required by federal law; no assignment is required as a result of a university “electing to retain title” to any invention–even if it is federally supported. And assignments induced based on a savings clause in research policy that the assignment is required to comply with a federal contract is also, in general, based on a false representation.

What constitutes a “subject” invention is a matter of federal patent law, not administrative choice. An inventor–knowing what was intended by the grant proposal’s statement of work–might best be able to understand an invention to be anticipated by that statement or not. But university inventors often are motivated to claim more that was accomplished during a grant than what was expressly “planned and committed.” I have had university inventors wanting to report inventions as if made with federal support when it was clear the invention pre-dated the grant, but clearly was related to the subject of the grant, and was useful to the grant. It’s just that it was not made (conceived, reduced to practice, and recognized) in the performance of work under the grant’s statement of work. Thus, there is a great deal to be done to improve university practice with regard to what constitutes a subject invention.

Conception and reduction to practice are personal matters–only an inventor can do them–to have in mind all the elements of an invention, and to either “actually” reduce them to practice to see that they work as imagined or to “constructively” reduce them to practice by describing them in sufficient detail that anyone with ordinary skill can see that they will work as described (as a properly drafted patent application does). An inventor must also see her (or his) work as inventive–a crucial condition to patentability. That’s the starting point. Now one can apply the new, useful, and non-obvious tests, and the statutory subject matter tests (some things are excluded–algorithms, surgery methods). But if one skips over conception, reduction to practice, and recognition, one skips the personal conditions of invention, and sets up to miss the purpose of the federal requirements pertaining to those very conditions.

The Guide assumes that a university will decide to own inventions. That’s a good assumption these days, of course. But it also exemplifies the still common belief that the Bayh-Dole Act somehow gives university administrators the choice to own inventions made with federal support. That is the intention behind statements such as:

Bayh-Dole grants nonprofit organizations, including colleges and universities, and small businesses the right to elect to retain title to inventions made in whole or in part with federal funds and to grant exclusive licenses to practice the inventions made. In return for the election rights, the nonprofit or small business must meet certain reporting requirements, federal licensing provisions, and other requirements.

It is true that “elect to retain title” is exactly the language of the statute. But it is not true that “electing” has anything to do with obtaining title, or that “retaining” has anything to do with taking title–requesting, requiring, assuming one already has title. Here the Guide jumps from “elect to retain title” to “election rights”–and then discussing what a contractor must do having obtained rights, as if “electing to retain title” means “taking title.”

The phrase “elect to retain title” has to do with how an assignee of a federally sponsored invention–a subject invention, an invention the assignee has acquired–can preclude a federal agency from requiring the assignment of title to the federal government. Throughout universityville, accounts of Bayh-Dole use this language ambiguously to make it appear that federal law establishes conditions that allow universities simply to take ownership of inventions from inventors who have used federal funds. And this is not true. It is a misprepresentation. It is not even an unintentional mistake, after Stanford v Roche. It may well be actionable under 18 USC 241 (criminal action) or 42 USC 1983 (civil action) as a conspiracy to deny university inventors their federal right to own inventions they make.

If two or more persons conspire to injure, oppress, threaten, or intimidate any person in any State, Territory, Commonwealth, Possession, or District in the free exercise or enjoyment of any right or privilege secured to him by the Constitution or laws of the United States, or because of his having so exercised the same… (18 USC 241)

Every person who, under color of any statute, ordinance, regulation, custom, or usage, of any State or Territory or the District of Columbia, subjects, or causes to be subjected, any citizen of the United States or other person within the jurisdiction thereof to the deprivation of any rights, privileges, or immunities secured by the Constitution and laws, shall be liable to the party injured in an action at law, suit in equity, or other proper proceeding for redress… (42 USC 1983)

The right to patent is “secured” by the Constitution as well as federal patent law. University officials approving drafting that makes it appear that federal law gives them the right to take ownership of patent rights, when that is not true, have conspired to injure inventors, and if when an inventor objects, the officials threaten the inventor, there’s even more evidence for the violation of federal law. University officials have an obligation to get the law right. It is more than just an institutional conflict of interest that university officials have no interest in doing so. And it doesn’t help when university officials write Guides that get it wrong, even if they are merely repeating an error that has swept through universityville. Being part of a bozonet is no excuse. I would think any university inventor contesting a demand by university officials that she (or he) assign inventions to the university as a requirement of Bayh-Dole has a clear-cut cause of action in court.

But there’s more wrong with the Guide’s account here: “Bayh-Dole grants” nothing to universities. Bayh-Dole does not apply to universities. It is a law directed at federal agencies, requiring them to adopt a uniform contracting practice for inventions in funding agreements with nonprofits and small businesses. The law stipulates that agencies will use standard patent rights clauses developed by the Department of Commerce, with some modifications and exceptions permitted. Bayh-Dole stipulates much–but not all–of what must be in those standard patent rights clauses. The standard patent rights clauses are then placed by agencies in their funding agreements, typically by operation of their own implementing regulations.

Bayh-Dole doesn’t “grant” anything. Bayh-Dole restricts some federal agency practices and requires a single approach, with limited flexibility, regardless of the purpose of the funding, the nature of the research work, or the context of the management of other research assets. In a nut, Bayh-Dole forces federal agencies to give up much of their ability to tailor their patent clauses to meet their own requirements. The key element in Bayh-Dole was requiring federal agencies to not assert an ownership claim on any invention made with federal support if a nonprofit or small business contractor had acquired title to that invention from the inventor. The notification of “election” confirms what the university has already done. Bayh-Dole does not somehow permit the university to acquire title, nor does it require inventors to assign title to their university. Bayh-Dole does not “grant” universities anything. It provides a mechanism by which agencies are prevented in making a claim under a funding agreement for research (or experimental work, or development) with nonprofits or small businesses.

Finally, it is also a mischaracterization to portray Bayh-Dole as a barter between the federal government and universities. Bayh-Dole is a form of “pre-approval” that prevents federal agencies from reviewing any particular invention acquired by a university for whether it is good public policy to allow the university to seek a patent, or to license that patent exclusively, or to sit on the patent until industry has adopted and then go on a suing spree, as Caltech, Harvard, and WARF have done recently.

Instead of agency review on case-by-case basis, or the use of a master agreement on patents, university-by-university, Bayh-Dole sets standard terms for such review. If a university acquires an invention made with federal support, then the limits of federal agency oversight–and the conditions the agency may impose on the university-as-contractor–are set out in the standard patent rights clause, or an allowable modification of that clause. If an agency believes that special handling of inventions is required–that ownership of inventions made under a given funding agreement should be owned by the federal government, or assigned to an industry consortium, or dedicated to the public domain, or licensed FRAND as part of a standard–then the agency has to declare such “exceptional circumstances” upfront. In all of this, there is no barter with the universities–other than when universities object to a federal request for proposals that invokes exceptional circumstances.

If Bayh-Dole were a barter, then one might think that universities obtained ownership of patentable inventions and that the condition on that ownership was that they secured patents and tried to make as much money as possible from their patent positions. This is the implied public policy, as universities depict Bayh-Dole. Another way of putting it, of course, is a bureaucrat’s thumb in every innovation pie. But rather than a barter with universities, Bayh-Dole states the conditions that any federal agency must impose on nonprofit and small business contractors doing research work with federal funding.

Three distinct groups are then identified by Bayh-Dole and the implementing–nonprofits, small businesses, and–importantly–individual inventors to be treated as if they are small businesses. The most restrictive conditions are imposed on nonprofits. That’s the work of section (k) of the standard patent rights clause at 37 CFR 401.14(a). With lesser restrictions are the conditions imposed on small businesses. Finally, the fewest restrictions are placed on inventors who do not assign their inventions to their employer-contractor. The standard patent rights clause for these inventors is at 37 CFR 401.9.

If there were to be a barter construed in Bayh-Dole, it would be conditional on the disposition of ownership–if ownership ends up with a small business, one set of conditions apply; if ownership ends up with a nonprofit, additional restrictions apply; if an agency allows ownership to remain with the inventors, then a third, liberal set of conditions apply; and if an agency requires assignment by the inventors or other contractor to the federal government, then only a few residual conditions apply (for instance, see 37 CFR 401.14(a)(e)). Showing these various possibilities then shows how a university administrator’s decision to acquire ownership of an invention represents the most restrictive environment for the management of the invention.

Bayh-Dole also imposes a set of conditions on subject inventions directly, by operation of federal patent law. Bayh-Dole creates a new class of invention–“subject invention.” Subject inventions, like other inventions, may be treated as personal property:

Subject to the provisions of this title, patents shall have the attributes of personal property. (35 USC 261)

But Bayh-Dole is inserted into federal patent law. Most of Bayh-Dole is directed at the requirements for a standard patent rights clause, and what agencies can and cannot do with such clauses, and how folks get to fuss about those clauses. But Bayh-Dole also includes a set of objectives (35 USC 200) that are directed at the use of the patent system–and thus, the use of patents on subject inventions. These objectives apply regardless of who owns the patents–nonprofits, small businesses, inventors, federal government–and these objectives set boundaries on the personal property rights that attend to subject inventions. Subject inventions are “subject to the provisions of this title [35 USC]” and unlike other patents, personal property rights subject inventions are “subject” to the objectives established by Bayh-Dole and placed in 35 USC.

For the management by universities of patents on subject inventions, this distinction is huge. Subject patents (let’s call them) do not have the same property rights as other patents. Management is not restricted only by the conditions agreed to by the standard patent rights clause that applies to the owner of the subject invention. Management also is restricted by the conditions required by the objectives established by Bayh-Dole as a matter of patent law. These conditions are not merely agreed to–they are imposed on all. Consider them a form of public covenant that runs with the property, a federal lien on the personal property right. University administrators scoff at this–there is no difference in patents on subject inventions, they argue dismissively.

The only difference university administrators appear to acknowledge is that unlike the typical patent owner (as they imagine a typical patent owner), university administrators claim to be motivated by moral purposes–public benefit, benefit for the university (money, visibility, appearance of “excellence” and leadership), and a proper regulation of the greed that university inventors would otherwise evidence if allowed to continue to own their own inventions. Or, rather, administrators make the university own and act–and so claim that it is the university that is motivated by moral purposes. Administrators can’t help it that those moral purposes result in patents going unlicensed, or flipped (exclusive license, assignment) to companies that don’t use or develop them, or flipped to companies that become trolls.

Apart from such moralizing, the primary difference in practice in university patent management is the unwillingness of university administrators to share patent rights–cross-license, establish standards, license in a non-discriminatory fashion, permit broad use for research and internal purposes. The default mode of university patent management is to manage for the most money, and use the threat of exclusion or litigation to drive up the price companies must pay to avoid getting sued.

One might argue that present university patent management is the least moral, most anti-competitive form of patent management out there–pretty much the opposite of what Bayh-Dole’s public covenent that runs with the ownership of subject patents requires. The “success” of Bayh-Dole, then, that university administrators trumpet, is the failure of anyone in government to oversee Bayh-Dole and insist that universities comply with its requirements–both as law and by means of the obligations universities accept under the patent rights clauses in their federal funding agreements.

The best advice one can get regarding compliance with invention management in federal funding is to start with the actual terms and conditions of the award, to see whether the standard patent rights clause has been modified, and then ensure that the university’s patent manage remains within the public covenant that runs with each subject patent.

Now consider how the Guide treats assignment of title to subject inventions:

37 CFR §401.414(k)(1) expressly forbids the assignment of patent titles to other entities or individuals without approval of the federal funding agency. The only exception is assignment to a firm whose specific mission is patent management, such as Research Corporation Technologies, Inc. (RCT is a company whose purpose is to exploit college and university technology through licensing and other arrangements.) A college or university administrator must be mindful of this requirement when accepting funding from other entities, including grants from nonprofit institutions, sponsored research agreements with commercial concerns, and materials transfer agreements from all sources.

I have highlighted multiple problems and can’t highlight another because it is omitted. First, the standard patent rights clause is at 37 CFR 401.14(a), not 37 CFR 401.14 [and now since NIST has changed all the numbering by folding (b) into (a) and eliminating both (a) and (b) designations, it’s back to 37 CFR 401.14–the Guide was wrong but like a clock that doesn’t run, they end up being “right” about something every so often!]. That means that the citation here to (k)(1) is 37 CFR 401.14(a)(k)(1). Second, the restriction on assignment concerns subject inventions, not “patent titles.” Here is (k)(1):

(k) Special Provisions for Contracts with Nonprofit Organizations

If the contractor is a nonprofit organization, it agrees that:

(1) Rights to a subject invention in the United States may not be assigned without the approval of the Federal agency, except where such assignment is made to an organization which has as one of its primary functions the management of inventions, provided that such assignee will be subject to the same provisions as the contractor;

Again, I’ve added some emphasis. Subject invention is a broader term than “patent titles.” It’s clear that (k)(1) is drafted not to prohibit assignment but to ensure that any organization receiving assignment has the ability to manage inventions. There is nothing in the standard patent rights clause (or Bayh-Dole) that requires a university to have as one of its primary functions the management of inventions, nor to have a patent policy, nor to have any agreements regarding patents with its personnel. That makes sense. The federal government does not require that universities own and manage inventions as a condition of receiving federal grants. That’s a reasonable public policy.

The Guide is right that (k)(1) appears to prohibit assignments to individuals unless approved by the federal agency by granting an exception only to organizations as assignees. But clearly the Guide gets it wrong about the nature of an organization receiving assignment. It’s not a “firm whose specific mission is patent management” as the Guide puts it. It is an “organization which has as one of its primary functions the management of inventions.” Again, these are important differences. The assignment exempted from a requirement for agency approval is much broader than the guidance indicates–many organizations may have a primary function to manage inventions without a “specific mission” to do so. And managing inventions is broader than managing patents.

The Guide also fails to point out that any assignment without approval of the federal funding agency must also assign the standard patent rights clause along with title to the subject invention. That’s an important practice tip for university invention administrators.

A second practice tip is that exclusive licenses may also function as assignments, if the university grants substantially all its rights in the patent under the license. Courts have determined that university-granted exclusive licenses are assignments when the license is worldwide, permits sublicensing, allows the licensee to sue for infringement, and reserves for the university only a non-commercial right to use for research and educational purposes. Thus, exclusive licenses that are assignments are subject as well to (k)(1), and the exclusive license should have to pass through the terms of the standard patent rights clause. Thus, sublicensing should be subject to the requirement that the invention be substantially manufactured in the U.S.

Of greater import, however, is the stipulation that the form of the standard patent rights clause is that which applies to the contractor–to the university. That means that section (k), which applies only to nonprofits, also applies to an exclusive licensee who is really an assignee. Notable there are the requirements to give a preference to small businesses in licensing (now sublicensing) and to use any income after the costs “incidental to the management of subject inventions” are to be used for “scientific research or education.” That takes some oompf out of a commercial profit-motive. University patent administrators and company officials alike should ensure that an exclusive license is sufficiently limited in scope that it cannot function as an assignment operating under an assumed name.

It is slips and skips such as these that create the conditions for the primary problem with the guidance on federal inventions–and that concerns how a university deals with its own inventors. Let’s work through the “guidance”:

Employee Patent Agreements. Section 401.14(f)(2) of the regulations expressly require that an organization’s employees, other than clerical or nontechnical staff, must agree in writing to promptly disclose in writing any inventions made with federal funds and to execute any documents necessary to obtain patent protection. For more information on these employee agreements, see ¶1862.

The Guide leaves out key points. First, the regulations apply to federal agencies and stipulate what federal agencies must conform to by way of standard patent rights clauses. The regulations set out these standard patent rights clauses and require agencies to use them. Agencies then tailor the clauses and insert them into their funding agreements. These are then clauses–terms and conditions of federal agreements accepted by universities–not regulations imposed on universities or inventors. It’s the difference between law and freedom to contract. That’s a big difference, really, even for federal agreements. Law applies to everyone. But contracts apply only to those who choose to participate in them. And that matters as we look at who individuals participate in federal contracts made between agencies and universities.

The (f)(2) requirement–37 CFR 401.14(a)(f)(2) comes in a section headed “Contractor Action to Protect the Government’s Interest.” Here is (f)(2):

(2) The contractor agrees to require, by written agreement, its employees, other than clerical and nontechnical employees, to disclose promptly in writing to personnel identified as responsible for the administration of patent matters and in a format suggested by the contractor each subject invention made under contract in order that the contractor can comply with the disclosure provisions of paragraph (c), above, and to execute all papers necessary to file patent applications on subject inventions and to establish the government’s rights in the subject inventions.

Perhaps it is subtle to differentiate “agree in writing” and “agrees to require, by written agreement” but these are entirely different matters. In (f)(2), the contractor is required to require a written agreement to protect the government’s interest. In the Guide, this is reduced to the idea that the contractor must enter into an agreement with its employees. (f)(2) is most definitely not concerned with “employee patent agreements.” It is concerned with the rights of the federal government. Yes, those rights might be protected by the university taking title to all subject inventions for the purpose of disclosing them to the federal government, but that’s not what (f)(2) requires.

(f)(2) requires a written agreement that flows down certain contractor obligations in the standard patent rights clause, ones that only an inventor could be able to do. The Guide gets two of these–disclose the invention, so the university can comply with its own disclosure obligation and sign documents necessary for patent applications (such as the declaration and oath by inventors that they are the inventors). But the Guide skips the third: to sign papers necessary “to establish the government’s rights in the subject inventions.” One establishes the government’s rights in an invention by licensing it or assigning it. An inventor, at the time of an invention, holds personal ownership of potential patent rights in the invention and only the inventor, then, has the standing to dispose of those rights. The (f)(2) requirement is that the university require potential inventors who become inventors to have made a written agreement to do these things.

The result of the (f)(2) written agreement is that inventors become “contractors” under the definitions of the standard patent rights clause. This step is essential in completing the connection between the federal government and the inventors, a connection with passes through the management of the university-contractor. The effect of the (f)(2) written agreement is to make inventors contractors who have a direct relationship with the government for disclosing inventions, filing patent applications, and establishing the government’s rights. It works this way:

37 CFR 401.2(a) defines a “funding agreement” to include

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.

Contractor is defined, as well:

(b) The term contractor means any person, small business firm or nonprofit organization which is a party to a funding agreement.

Note the entities that may be a contractor–person, small business firm, nonprofit organization. If the university-contractor subcontracts work, substitutes parties, or assigns work, those involved come within the definition of contractor–become contractors to the federal government.

For subcontracts, the standard patent rights clause (37 CFR 401.14(a)(g)) sets out the requirements when work under a funding agreement is delegated to another organization:

(1) The contractor will include this clause, suitably modified to identify the parties, in all subcontracts, regardless of tier, for experimental, developmental or research work to be performed by a small business firm or domestic nonprofit organization.

When the standard patent rights clause is included in subcontract, the contractor referenced by the clause is the subcontractor. (g)(1) adds stipulation:

The subcontractor will retain all rights provided for the contractor in this clause, and the contractor will not, as part of the consideration for awarding the subcontract, obtain rights in the subcontractor’s subject inventions.

I’ve emphasized the last bit. The university cannot require assignment (or licensing) of any subject inventions as a condition of the subcontract. When the standard patent rights clause flows down, the university cannot demand an interest in whatever happens. Any arrangement regarding rights in a subcontractor’s subject inventions must be outside the federal funding agreement, must be voluntary (that is, cannot be a condition of the subcontract and so must involve freedom to contract), and must therefore also require consideration independent of the federal award.

The subcontracting section continues by providing in (2) that if the subcontractor is not a nonprofit, then the standard patent rights clause that flows down is the one that would apply to the subcontractor were it directly funded by the federal government. In (3), the subcontracting section makes it clear that if the primary award to the university is a contract, not a grant (and thus made under the FARs as a procurement, not under Circular A-110/2 CFR 215 as a subvention), then the subcontract results in a contract between the subcontractor and the federal government. This is nature of a federal subcontract. In the case of a subcontract to another non-profit, the resulting relationship between the subcontractor and the federal government is similarly a subvention, formed when the contractor brings the subcontractor in to become a “party to the agreement.”

Now come back to the written agreement requirement in (f)(2) of the standard patent rights clause. It has the same structure as subcontracting, but it is a requirement, and directed at individuals–persons–not organizations. The university-contractor is required (and agrees) to require these individuals to do certain things under the funding agreement’s standard patent rights clause–to disclose subject inventions, sign patent application documents, and sign documents establishing the government’s rights.

If we follow the parallel treatment in subcontracts, we also find that the university cannot as a condition of this written agreement requirement require assignment of subject inventions to the university. Any such university-directed interest is also outside the operation of the funding agreement and this (f)(2) requirement. One might read the (f)(2) requirement–which is a federal requirement to create a federal agreement–as “regardless of any private agreement regarding patent rights between a university-contractor and its employees, the university-contractor must require those employees to make a written agreement to protect the government’s interest, and the university-contractors that when individuals–persons–do so, that agreement, as part of a federal funding agreement and expressly approved by the university-contractor takes precedence over and releases individuals making the written agreement from any conflicting requirements the university may place on these individuals and any conflicting agreements that the university may have with them.

It is not a conflict for a university and an employee to agree to assign inventions to the university. But it is a conflict for the university to demand the assignment as a condition of participating in a federally funded project, just as it is a conflict for the university to demand assignment from a subcontractor as a condition of participating in a federally funded project.

One more point confirms this understanding. Bayh-Dole and the implementing regulations both anticipate that inventors may not assign title to inventions to any organization. Patents have the attributes of personal property, and when someone invents, the right to seek a patent is owned by the inventor. 37 CFR 401.9 provides a standard patent rights clause for inventors who do not assign their invention and for whom the federal funding agency declines to require assignment to the federal government. 401.9 provides that inventors will be treated by the federal government as if they are small business firms and given terms more liberal than those provided to other small businesses–and much less restrictive than those imposed on nonprofits (the most untrustworthy and suspected of the bunch!).

When inventors-to-be make the written agreement to protect the government’s interest, they become parties to the federal funding agreement–they become contractors for their little bit of protecting the government’s interest in subject inventions–and they are to be treated as if they are small business contractors with regard to the patent rights they retain.

It is evident, then, that the (f)(2) written agreement has nothing whatsoever to do with any employee patent agreement, and indeed displaces any such agreement with regard to anything that the university’s requirement of and approval of the written agreement knocks out. The federal condition for individual participation in a federal funding agreement with a university is that the university require the individual to become a party to the agreement for certain matters only an inventor can handle–only an inventor who owns his or her rights in a subject invention prior to any other disposition of those rights.

It is therefore entirely incorrect and misleading to for the Guide to tie employee patent agreements to compliance with federal funding. Employee patent agreements have absolutely nothing to do with compliance with federal funding agreements. If university administrators decide they must have all inventions made by faculty (and other research personnel), that’s a fixation they have all on their own. It has nothing to do with the federal government, federal policy, federal regulations, federal agreements. It has everything to do with a desire to own and control and profit from the personal property of others. Call it that. Own it.

Now look at how the Guide document handles this issue:

1862 Employee Intellectual Property Assignments

As part of an institution’s intellectual property policy, employees, including students, generally are required to sign patent assignment agreements. These agreements typically require an employee to acknowledge and accept the institution’s intellectual property policy, make specific reference to federal and other requirements, and acknowledge an institution’s royalty-sharing policy. There is no set format for these assignment agreements; however, Figure 1862-1 provides one example of such an agreement.

There is no federal requirement in Bayh-Dole that a university have a patent policy or have patent assignment agreements. In fact, Bayh-Dole is set up to anticipate that a university does not have, and need not have, any such policy or agreements. Everything needed to satisfy the federal government with regard to patents is contained in the standard patent rights clause–and that ends up being that a university disclose subject inventions, identify personnel responsible for patent matters, educate research personnel on the importance of timely disclosure of inventions, and require its inventors-to-be to make a written agreement to protect the government’s interest by disclosing subject inventions, signing patent application papers, and signing papers to establish the government’s rights. That’s it.

There is absolutely nothing about assignment of inventions to the university. If the government had wanted such a requirement, the government was perfectly capable of stating it, just as it had with the old Institutional Patent Agreement, on which Bayh-Dole was in part based. It is not a mistake in Bayh-Dole and the implementing regulations and the standard patent rights clauses, as some legal commentators have argued. The absence of such a requirement to assign to the university-host is a feature. The absence of a university-assignment requirement–the refusal to continue the IPA practice of requiring such assignment–is an express statement, as it were, of federal government policy not to stipulate such a requirement as a matter of federal patent law. And by not stipulating such assignment while excluding all other conditions placed by federal agencies on subject inventions, Bayh-Dole makes it outside the authority of the Department of Commerce to draft a clause that requires such assignments by inventors to the university-contractor.

More (bad) guidance:

Because assignment agreements are necessary to comply with federal patent regulations, it is important that an institution ensure that all appropriate personnel have executed the agreement.

No, assignment agreements are not necessary to comply with federal patent regulations [except now, after the NIST revisions in May 2018, NIST does require assignments–but for inventions that the contractor already owns–fathom that. The best I can make of it is that contractors holding equitable title must require inventors to assign legal title]. Requiring a written agreement to protect the government’s interest is necessary to comply with the standard patent rights clause–a federal contracting matter, not a federal regulation matter. The agreement that all appropriate personnel must execute is one that protects the federal government’s interest directly, not one that forces all federally supported inventions into the clammy hands of bureaucrats. Or, to use Phil Knight’s term in Shoe Dog, bureau-krakkens.

Many institutions require that the assignment be made at the time of employment, perhaps at the same time a faculty or staff member signs up for benefits. Acquiring student signatures is often more difficult, and many institutions transfer the responsibility for student assignment agreements to the student’s department.

The written agreement is specific to each federal funding agreement. It is a requirement of each patent rights clause as included in each federal funding agreement. One cannot execute such an agreement “at the time of employment.” One executes the agreement after the university has accepted the award, when the funding agreement is effective, and the university has agreed to comply with its terms and conditions, including (f)(2) of the standard patent rights clause.  And in any event, the (f)(2) written agreement is not an assignment agreement and it is not directed at the university, and it is not for the university’s benefit.

The Guide then gives an example of an employee patent agreement, as if such an agreement complies with federal regulations. Read and weep:

I understand that in the course of my activities at the <institution name> I may participate in a research program sponsored through agreements with government agencies, corporations, foundations, or others outside the University. I also understand that these agreements usually require that the <institution name> protect the sponsor’s rights to intellectual property that may result from such a research program.

See how the federal requirements are conflated with those of private sponsors, which may indeed stipulate that the university obtain rights from inventors? But way, way more frequently it is the university that inserts a university-ownership clause than that a sponsor does so. A company sponsor may well be *more* ready to take assignment of inventions directly from inventors than have to deal with the bureaukrakken. University administrators insist on a university-ownership clause and then (often) rely on a policy requirement that university personnel comply with whatever is in a university contract and that university contracts supersede any conflicting patent policy provisions. Adding a present assignment to future inventions (and non-inventions) is just moats and walls and bars to make clear the shackles and chains already intended.

If I receive any support from the University or other sponsoring agency through affiliation with a sponsor program, whether as salary or sharing in equipment use, expendable materials, or other support, I agree that I will:

This is messy. It is an overly broad scope for federal requirements. The scope of government interest in inventions–what makes an invention a “subject invention” is set out in 37 CFR 401.1 and in the definition of a subject invention. A subject invention has to be made in the “performance of work” in a federally supported project, and must be within the “planned and committed activities” of that work, or must have “diminshed or distracted” from that work. Sharing equipment, for instance, does not necessarily fall within the scope of the definition of subject invention. Nor does “any support” or “other support” (such random, globular, indefinite ideas written by drafters who imagine a big world but have no clue what’s in it, but if there is anything, it’s definitely “other’)–it must be funding–and the work must be planned and committed work, not just any work, or it must keep people from their planned and committed work.

1. Disclose promptly to the <office name at institution> full information concerning inventions or discoveries I may make in the course of any such sponsored research or training program;

Notice the appearance of “training program.” University drafters cannot help themselves. The recitals of the agreement identify “research programs” not training programs. And training programs rarely have patent requirements. Bayh-Dole requirements are implemented only for funding agreements that involve “the performance of experimental, developmental, or research work.” Bayh-Dole expressly exempts training programs (35 USC 212):

No scholarship, fellowship, training grant, or other funding agreement made by a Federal agency primarily to an awardee for educational purposes will contain any provision giving the Federal agency any rights to inventions made by the awardee.

No training programs here. Further, the federal stipulation for scope is “performance” not “course.” In federal awards, “course” is constrained by the “planned and committed” activities set out in the research proposal. In any event, the federal requirement for what is to be disclosed is specific (see 37 CFR 401.14(a)(c)(1)), not nebulous “full information” and pertains to subject inventions–inventions which are or may be patentable. Here’s what is required:

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). It shall be sufficiently complete in technical detail to convey a clear understanding to the extent known at the time of the disclosure, of the nature, purpose, operation, and the physical, chemical, biological or electrical characteristics of the invention. The disclosure shall also identify any publication, on sale or public use of the invention and whether a manuscript describing the invention has been submitted for publication and, if so, whether it has been accepted for publication at the time of disclosure.

To meet the federal requirement, then, an inventor would need to know these details as a condition for making the written agreement. On the other side of things, company sponsors are often not concerned specifically with inventions. They use something more like NASA’s “New Technology” reporting requirement–report anything you do, whether new or not, whether inventive or not, whether patentable or not. Whatever claims of ownership might be made in what’s reported is an entirely separate matter. Conflating an obligation to report with an obligation to assign is one of the chronic stupidities of university invention management practice.

2. Assign to the <institution name> or its designee all of my right, title, or interest to such inventions, discoveries, patent applications, or patents; and

Suddenly the list has expanded from invention to a host of things. But the whole point of right, title, and interest to inventions is patent rights. If there are no patent rights, then there’s no ownership theory available for a bare invention–unless the university is going to assert trade secret or, in some twist, copyright in the documentation of the invention. Of course, the major point is that for federal funding, assignment to the university is not required. This is just institutional grasping based on a misrepresentation of federal requirements.

3. Upon request of the <institution name> execute any documents and do everything necessary and proper to secure the issue of letters patent, United States or foreign.

The “do everything necessary and proper” is clever. Often, thoughtful drafters include “at the <institution name>’s expense.” The federal requirement is to allow patent applications “to be filed.” Anything further is a university lust, not a federal one. For federal funding agreements, the written agreement does not stipulate who makes the request for executing papers necessary to file a patent application–securing the patent is not part of the written agreement.

Nothing here complies with the (f)(2) written agreement requirement. There’s no obligation to protect the government’s interest. No obligation to execute papers to establish the government’s rights in subject inventions.

I understand that the <institution name>’s policies may provide for sharing of any income arising from inventions, discoveries, or patents that I assign to the <institution name> under this Agreement. I also understand that such sharing and other policies and practices that are outlined in <institution name>’s research policy manual which may change from time to time, I agree to abide to by the terms in effect at the time of the disclosure of an invention.

This is meaningless stuff here. Income doesn’t arise from inventions and the like–income arises from the disposition of inventions–licensing, assignment, litigation. Notice that “patent application” has disappeared from the list. Unless the intent is to incorporate by reference the royalty sharing policy into the agreement. That’s what the University of California ended up doing and a court found that the university then could not change that policy without the concurrence of an inventor. By including “other policies” as well, this agreement–clearly intended to be contractual (though there is no consideration, so is it?)–the university here locks down all policies so included. The university may change the policy statements, of course, but cannot change them with regard to an inventor who has signed this deal. The bit that says the policy may change is just a demand for an agreement to agree–unenforceable stuff. Any change in policy-in-contract must carry its own consideration, and for all that, to stick it, the contract has to make acceptance of the change a condition of employment. “Take the new contract, with its new consideration, or take a hike.”

I understand that from time to time the <institution name>’s intellectual property policy and practices are periodically reviewed for needed changes, such as those resulting from government regulations. I agree to keep informed of any changes in the intellectual property policy through revisions of the research policy manual or other announcements.

Again, these them are weasel words. The policy that matters is the patent policy, not the “intellectual property” policy. This is an example of an “employee patent agreement.” Copyrights, trademarks, and trade secrets are different, handled differently. There are no government regulations that require changes to a university’s intellectual property policy. And being informed about changes has nothing to do with this agreement anyway. And making it the employee’s obligation to find out any changes is pure bureaukrakken–the university has the obligation to track changing requirements and inform employees of the changes. The university cannot shift the responsibility to the employee, as if it is the employee’s personal obligation to do so on behalf of the university. This is an agreement regarding an employee’s personal property–it is not an agreement within the university delegating work responsibilities. Just nuts.


And we are back to “research” participant. Nothing more about those “training” programs. Must have been thrown in there to add variety or something.

In short, there’s nothing about this “employee patent agreement” that complies with the (f)(2) written agreement requirement. Further, there is much to object to even as a university employee patent agreement. The standard of practice in universities is low, the willingness (or ability) to be clear regarding federal obligations is low, the ability to draft a document that’s consistent, tracks regulations, focuses on its purpose, and remains clear–is almost non-existent. Is it too much to ask that folks who regard themselves as “professionals” and “specialists” take the time to master the fundamentals of their work?

Here is a link to a simple, compliant version of an (f)(2) written agreement. One can add more detail regarding what’s involved in a proper disclosure, of course, but simple and clear is a good way to start.

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