In a note on the Guide to Bayh-Dole, a reader asks:
One point I have been particularly concerned with is the position taken by virtually all of the universities that if a university employee is hired as a consultant to a private company to work on research and development activities, the sponsoring entity will not own any invention made by that employee resulting from that work unless a company coworker is a joint inventor. In such situations the university will give the company a right of first negotiation for a license but refuses to set pre-determined terms of the license or to other include any new patent as a part of the original licensed patents. They routinely claim such restrictions are imposed under the Bayh Dole Act. I have never been able to find an authority for this position.
Have you found this situation to be common and is there any real basis for such a rule other than the desire of the university to control all development and make such deals difficult if not impossible to consummate?
Yes, the situation is common, and there is no basis in Bayh-Dole for the university behavior. The administrators don’t want to make deals difficult to do–they just believe in unicorns and rainbows that companies should blandly and as a matter of course accept university ownership, university reimbursement for patenting work, and university demands for a share of the company’s income (even if the company might otherwise be *losing money*–as startups often are). A company’s acceptance of university-sided, draconic, policy-required, often silly terms is called a “negotiation.” The purpose of this “negotiation” is for the university administrators to recite policy requirements and explain to each clueless and wildly self-interested company that the company exists to feed a supply of its financial life-blood to the university administration, where whatever money is received will be thrown into a slush bucket and used for random administrative conveniences–stuff so inconsequential its use is almost never publicly reported.
“Problem” companies think a negotiation involves an exchange of views and interests, a seeking for common purposes, and an arrangement in which each side obtains something of sufficient value that it warrants giving up other things of value for the benefit of the other party. Such a wildly idealized and problematic view of “negotiation” rarely exists in university usage. Thus, even the term “negotiation” has special meaning among university administrators.
For those of you who already get it, or just want to know the basics, here they are:
- There is no authority in Bayh-Dole for what the universities are claiming. It was bombast, but after Stanford v. Roche, it is more like contempt.
- The refusal to set terms comes from a conservative reading of Rev Proc 47-2007 (or a failure to read it!). But those procedures are navigable.
- There is a sordid history of deception for the sake of the cause, and the result is compulsory IP policies that are destroying university engagement with industry.
I will deal with Rev Proc 47-2007 objections in a separate article. Here, let’s focus on Bayh-Dole objections.
Bayh-Dole does not apply to universities or inventors
First things first. Bayh-Dole Act only applies to federal agencies, not universities, so when university officials claim Bayh-Dole requires them to something, it’s simply untrue.
Bayh-Dole requires federal agencies to adopt uniform policies regarding how the federal government acquires an interest in inventions made with federal support–that is, under federal grant or contract. Bayh-Dole authorizes the Department of Commerce to create standard patent rights clauses, and the enabling regulations permit agencies to tailor those clauses in various ways. Thus, any university obligation under federal funding arises directly and expressly from the particular patent rights clause in a given federal funding agreement.
There are a few details that do apply directly. The definition of “subject invention” for instance establishes as federal law a new category of invention. See 35 USC 201(e). The category designation applies if and only if an invention meets the requirements of the definition:
(e) The term “subject invention” means any invention of the contractor conceived or first actually reduced to practice in the performance of work under a funding agreement: Provided, That in the case of a variety of plant, the date of determination (as defined in section 41(d) [1] of the Plant Variety Protection Act (7 U.S.C. 2401 (d))) must also occur during the period of contract performance.
Inventors may assign to companies, even with federal funding
The Stanford v Roche Supreme Court case (2011) makes clear that individual inventors may assign their interest in inventions to private parties, even in the presence of federal funding, and not violate the standard patent rights clause. There are reasons. For an individual’s rights to assign to be restricted under federal funding, a university contractor must comply with the patent rights clause. That clause requires universities to delegate assignment responsibilities to potential inventors (37 CFR 401.14(a)(f)(1)). I have not found a single university that implements this requirement. The universities are in breach of the federal funding agreement, though I have also not heard of an effort to enforce the (f)(1) requirement by any federal agency.
If a university were to enforce the (f)(1) agreement, then that agreement would supersede any other requirement that the university might place on its personnel under any theory of employment or use of resources. Multiple reasons here. First, federal contracts trump private ones. Second, the university cannot contract with the government to treat research personnel in one way, and then attempt to treat them in a conflicting way. Third, when the university accepts the federal funding agreement (and the requirements of 2 CFR 215), it agrees to provide the resources to support the sponsored research, and it releases its personnel to work on that research, with their salaries paid by federal funds–the university is fully compensated for both resources (F&A/indirect costs) and salary (direct costs), so the claim of use of resources or employment do not hold up. Finally, university faculty are not employees with regard to their research unless they expressly agree to do research for the benefit of, and under the control of, the university.
A university does not “employ” faculty for research, except when they expressly agree
The university does not act as an employer with regard to faculty research. Faculty choose what and how and where to conduct research; faculty decide who to collaborate with and when to publish their findings. Faculty may be employees with regard to certain work obligations. Faculty are certainly treated as employees for HR purposes. But faculty are not, as a matter of practice, employees in the conduct of their research. This is most clearly evident for copyright matters, where the work made for hire agency test simply cannot be met for faculty research, even if the university’s resources are used–because the university has no power to require the faculty to use its resources. Inventions, too, are not service inventions. Faculty are not hired to invent. It is not in their appointment letters. It is not in their collective bargaining agreement. It is not in the federal funding agreement. In fact, this is one area in which Bayh-Dole overturned federal agency contracting practice, which was, in institutional patent agreements, to require universities to require assignment of inventions made with federal support. Bayh-Dole eliminated that requirement!
How university officials subverted univesity IP policies
University officials recite Bayh-Dole because they used this excuse to circumvent their own liberal policies regarding inventions. Patent licensing officers wanted all the action, and bureaucrats wanted that action by force of policy rather than by quality of service. Most universities had a liberal patent policy allowing faculty ownership, with the qualification that the terms of extramural contracts superseded policy. Thus, it was convenient to claim that federal funding agreements required university ownership. Better, the administrators decided to claim that federal *law* required university ownership. Of course, the administrators by this time were up to their neck in the lie and could not decide quite how Bayh-Dole did this. They came up with all sorts of versions, some of which outed in the Stanford v Roche case: Bayh-Dole vested ownership directly with the university contractor; Bayh-Dole granted the university a right of first refusal; Bayh-Dole prevented inventors from assigning to anyone but the university; Bayh-Dole offered invention ownership to the university, subject to the university agreeing to accept ownership–all of which are utterly untrue.
The university administrators, having established the claim that federal law required the university to own federally supported inventions, then went about revising university policy to “comply” with the law. For this, the argument went, it is unfair for the university to take federally funded inventions (and give them such wonderful support) and not take the rest, too!
The Stanford v Roche delusion persists
But there was also a darker reason (if it can get darker). The claim was, if Bayh-Dole *required* a university to own an invention (and to commercialize it–also total garbage)–then a university might violate federal law if it permitted any invention to be held privately, which might subsequently be supported in its development by federal funds. The administrators claimed that because of the uncertainty of when an invention is *actually* reduced to practice, an invention could be conceived and constructively reduced to practice (as in a patent application) and then receive federal support and be *actually* reduced to practice, thus coming within the scope of Bayh-Dole. The universities claimed (and MIT was right there with this claim) that if even $1 of federal money touched invention development activities, the invention was subject to Bayh-Dole. Utter, total nonsense. The Supreme Court expressly rejected the argument.
University officials have no authority to expand the scope of subject invention
But even 37 CFR 401.1 makes clear that the scope of federal interest can be no greater than the “planned and committed activities” of a federal award. That is, the written scope of work, typically in the proposal (grant) or in the RFP (contract). If a funding agreement anticipates a scientific study but not development of applications, then according to 37 CFR 401.1(a)(1), any development is out of scope unless that development “diminishes or distracts” from the performance of the federally funded work. That is, someone would have to show that the application work prevented the university from completing the contracted work.
I have never heard of a university administration ever checking the statement of work of a federal award before determining if an invention is within scope of the standard patent rights clause. Invention disclosure forms (and I used to collect them–I’ve seen *many*–ask only for a federal grant number, never a statement of scope. The assumption is: if an inventor puts a federal grant number on the disclosure form, then the invention is a subject invention, without any inquiry into whether the inventor understands the implications and technical requirements. In some cases, the disclosure form may include instructions that deliberately mislead inventors. For instance (my bold):
If the invention was made in connection with any sponsored research, the appropriate grant numbers MUST be listed. (Northeastern–the disclosure form has been removed from the page originally linked to; perhaps some revisions have been considered)
List all sources of funding used in the development of this invention. (Delaware)
Name of industry, university, government, or other sponsor, if any. (Florida State—the disclosure form has been removed from the page originally linked to; perhaps some revisions here, too)
The Funding tab provides for entry of any sponsorship support received for the research that resulted in the invention being disclosed. (Oregon State—funny how often this guidance stuff changes–here is the archived file)
As funding and use of proprietary resources and materials often carries patent obligations, be sure to include all outside agencies, organizations, or companies that actually provided any supply, or expense funding to any inventor for the research that led to the conception or first actual reduction to practice of the invention. (Do not list any funding source gratuitously.) (California–here is the archived file)
These descriptions may seem innocuous. They are drafted to appear that way. There is no effort to draw the inventor’s attention to the statement of work, or to explain that the invention must be within the scope of that statement for a sponsor to be listed on the form. If the university simply wanted notice of all the extramural support an inventor has recently had, the university already has that information in its grants and financial databases. The only reason to request that information on disclosure form is to make the inventor make a decision regarding what sponsors, if any, meet the requirements of the contracts under which the inventions were made. But that decision should be based on the contracts, not on simply reporting sponsors “to be safe” or as a matter of random procedure or “to give credit” to sponsors or to graciously acknowledge support that led to an invention but did not specify the work to invent or the support ended before the invention was made.
At least the University of California instructions exclude gratuitous listing of sponsors. But even here, the university uses “led to” rather than “for which the contracted statement of work reasonable includes the work done to make the invention” and further expands the scope of funding to include “supply funding.” The language in the instructions directly mirrors the Bayh-Dole definition of a subject invention:
conceived or first actually reduced to practice
But Bayh-Dole finishes with the restriction:
in the performance of work under a funding agreement
And the University of California substitutes
funding . . . that led to
These are distinctly different descriptions of scope. In the federal case, the invention arises from the performance of work the university has contracted with the government to perform. In University of California version, the invention is whatever may later result (“led to”) from the use of funding without regard to the scope of agreement either in terms of performance or time. The effort is to expand the scope of what will be reported as sponsored. Everything sponsored is more readily claimed as the university’s. The policy positions, down to the innocuous instructions, supports the effort to implement a comprehensive, compulsory IP policy. If such a policy can be achieved in practice, then it is a simple matter to “correct” the text of the written policy to conform with such practice. And across the country, that has been what has happened. If an IP office is authorized to create faulty instructions on a disclosure form, and those instructions are accepted as policy, then all other policy guidance can be revised to conform. The policy change has happened in part with the assistance of conveniently incompetent drafting of disclosure forms.
In the case of federal funding, an agency providing equipment or funding for supplies surely has little basis to claim an invention made with such resources–the funding will be to purchase equipment or supplies–that will be the statement of work. An invention made as a result of the use of such equipment or supplies cannot be within scope. But a conscientious University of California inventor would never know that.
The implementing regulations in 37 CFR 400 are written with the expectation that contractors might attempt to claim inventions as private which should be within scope–no one apparently considered that universities might do the opposite and try to claim inventions that were not within scope actually were. There is a huge asymmetry here. If an invention is within scope, it is a subject invention–that is one definition in Bayh-Dole that applies to any invention and does not flow down through the standard patent rights clause. But if an invention is not within scope, there is no power in the university or the law that permits the university to declare an invention subject to the law.
Faculty inventor choice had to be eliminated
Why then did the university administrators, if they so wanted ownership of inventions, do this? Why didn’t they just require faculty to assign all inventions to the university, and then seek to restrict the scope of Bayh-Dole to its narrowest legally mandated limits? Let agencies appeal if they wanted. If university officials had done so, they would have had even greater freedom to manage patents–assign to anyone, license exclusively without restrictions, not share income with inventors, not report invention use or enabling publications, not have to file patent applications, not have any obligation to pursue practical application, not require licensees to report use, no march-in rights. Why did officials try to expand the scope of Bayh-Dole?
Because it was so easy to dupe the faculty into believing federal law required every invention to be touched by a university bureaucrat before it could be used by the public.
Alternative reasons are less generous
One could chalk it up to a kind of fortuitous administrative incompetence: “Oh, look, we sort of bungled our way into positions of impunity with regard to these apparently valuable inventions.” If that were the case, then once the error had been pointed out, the administrators would have quickly unwound the situation. They have not. There are better explanations than fortuitous administrative incompetence, or at least explanations that do not depend entirely on fortuitous administrative incompetence.
One explanation is that the administrators exploited loopholes in university policy. A university might find such a loophole in research policy, which typically requires that research contracts take precedence over policy defaults. A university could, effectively, contract itself into policy that faculty would otherwise never agree to–all that was needed was an external sponsor who agreed to the requirements. Then the university could impose those requirements on its own faculty. Thus, a university could demand ownership of inventions as a condition of accepting research funding, and then force university personnel to comply with the contract (even though the sponsor had imposed no such condition–the university imposed it, effectively abrogating its own policy otherwise). If a university could do this in contracting, the reasoning goes, why, then the university has the power to unilaterally impose the requirement to assign in all cases. Policy changed.
A second reason was the need, now increasingly desperate, to show that the new regime of technology licensing–licenses offered directly by universities and not through external agents such as Research Corporation or Battelle–was growing and prospering. University-affiliated organizations such as WARF were in competition with Research Corporation. University of Wisconsin faculty could take their inventions where they wished. An expanded Bayh-Dole scope pushed the decision of assignment to the university (so it was claimed) and the university would assign preferably to WARF. Thus, the expanded scope of subject invention also served an anti-competitive purpose, by giving a monopoly on licensing/deployment methods to university administrators. Faculty choice had to go.
For that, the options–such as Research Corporation or any other specialist in patent management–had to be disabled or discredited; and once these options were disabled, it was a simple matter to prevent faculty choice by policy as well–what were faculty fighting for when the only “viable” patent management organization left was the university’s own organization, whether external (like WARF) or internal (like Stanford, MIT, and UC)?
University administrators furthered the deception by arguing that the university could not assign federally supported inventions to anyone other than a non-profit invention management organization. But this, too, is simply untrue. 37 CFR 401.14(a)(k)(1) allows assignment to any organization “which has as one of its primary functions the management of inventions.” Any large corporation with a patent operation would qualify; any startup spending much of its resource on patent filings would qualify. And any organization can be an assignee with agency approval.
The simple answer is, universities, having overclaimed inventions made in environments that had federal support, also overclaimed ownership of non-federal inventions using the premise that they might violate federal law if they didn’t. Repeat the refrain: utterly bogus.
Privately supported inventions also have to be taken
So now to privately supported inventions. There are two forms, in sponsored research and by personal consulting. If the scope of work for the private agreement overlaps with the scope of work for a federal agreement, then the if an invention is within that intersection, the federal agreement will control. But even then, the federal agreement’s standard patent rights clause does not prevent assignment to the private sponsor. Again, the Stanford v Roche decision makes this doubly clear for those who cannot read or never have bothered.
In the second situation, the inventive work takes place in personal consulting. The faculty inventor works off-site, and outside sponsored research duties, and in a manner that does not affect any university responsibilities (and so stays within the one day a week consulting limit, for instance, and has all approvals for such work). Here it is difficult to see how the university has any authority to reach to such inventions. A number of states have laws that aim to prevent such a taking. Universities claim that the scope of employment is as broad as the “professional duties” that a faculty member might perform, including “public service” of which consulting might be considered a part. Thus, anything one might invent in one’s area of professional expertise could be claimed by the university, regardless of the actual work performed by a faculty member at the university. The argument is: even though you may decide what to study at the university, we will not allow you to conduct any study–of research, development, or advisory service–on behalf of anyone else. We demand ownership of all IP you might produce in such arrangements. It is an ugly employment agreement, then. But no, really it is a non-compete agreement implemented as an IP policy. In California, at least, I could imagine that it would be void as contrary to state law, though it has not been tested yet by UC faculty.
Yet another variation is the “worry” that “trade secrets” of the university might be used in making inventions in consulting arrangements. Again, it’s all floppy. First, with regard to faculty knowledge, that cannot possibly be a trade secret of a university unless the faculty member cedes his or her academic freedom to the university–mostly unheard of.
Second, and here’s the rub, with regard to trade secrets received by the university from others, there is an issue. If a faculty member has access to such trade secrets, and exploits them for another company, then there readily could be a misappropriation charge. If the university practice is to approve consulting, then the university will be easy to include in any such charge. If the university practice is not to approve consulting, then it has less exposure. What exposure it does have comes from a failure of administrative practice in how trade secret information is received. The general practice is to refuse such information, especially of a technology nature, as doing so triggers export control law, and that in turn makes foreign nationals (such as students and visiting faculty) second class research citizens and creates substantial obligations for obtaining and documenting “deemed” exports–any disclosure to unauthorized foreign nationals.
If a university is going to receive such trade secrets anyway, then it must implement a program that is reasonable under the circumstances to preserve confidentiality. Or, in the alternative, disclaim such a program as a condition of acceptance of trade secret information, so that if that information is subsequently disclosed, the provider has no cause of action against the university for failing to take action to prevent disclosure. But this sort of thing rarely happens–incoming IP practice is worse than outgoing at most universities. So administrators, finding that one good defect deserves others, try to enforce a program of trade secret compliance at consulting, and that also conveniently captures any inventions a faculty member might make in consulting.