How Bayh-Dole failed to protect faculty inventors (from university administrators)

[Now with some revisions in the second paragraph that on reflection were worth making.]

There are a number of things wrong with the Bayh-Dole Act, such as the lack of accountability for the disposition of privately held patents on inventions made with federal support–agencies do not have to collect utilization reports, cannot publicly release the information in those reports via FOIA, and must navigate a mind-numbing protocol to “march-in” when nothing or worse is being done with those patent rights. But I want to look at something less obvious but just as bad in Bayh-Dole.

Bayh-Dole was motivated by university-affiliated research foundations not getting access to federally supported inventions when they thought they ought to. One big fuss was at Purdue Research Foundation. The Wisconsin Research Foundation joined in. Some universities also conducted their extramural research by means of a research foundation–SUNY’s research foundation is a notable example. Thus, as Bayh-Dole was being drafted, it could not be just about the affiliated research foundationsuniversities. The affiliated research foundations universities were included as well. In Bayh-Dole, this expansion from university to research foundation to university is found in the definition of “nonprofit organization”:

The term “nonprofit organization” means universities and other institutions of higher education or an organization of the type described in section 501(c)(3) of the Internal Revenue Code of 1986 (26 U.S.C. 501(c)) and exempt from taxation under section 501(a) of the Internal Revenue Code (26 U.S.C. 501(a)) or any nonprofit scientific or educational organization qualified under a State nonprofit organization statute.

Small businesses were added to Bayh-Dole to augment the new SBIR program that was being developed at roughly the same time. Perhaps doing so made Bayh-Dole attractive to some legislators. Bayh-Dole makes a distinction between nonprofit organizations and small business firms, which it carries throughout its text. Examples:

Each nonprofit organization or small business firm may, within a reasonable time after disclosure as required by paragraph (c)(1) of this section, elect to retain title to any subject invention: Provided, however, That a funding agreement may provide otherwise

Each funding agreement with a small business firm or nonprofit organization shall contain appropriate provisions to effectuate the following

The broadest term used by Bayh-Dole is “contractor”:

The term “contractor” means any person, small business firm, or nonprofit organization that is a party to a funding agreement.

There, one finds not only nonprofit and small business but also the addition of “person.” This usage is interesting. Typically “person” in text like this means “natural person or an entity with its own legal standing to act, such as a corporation or government.” But here, “person” appears in a list of equals–a small business and a nonprofit organization are both “persons” for legal purposes. Interpreting “person” here in its broadest meaning makes the term duplicate the following items in the list and furthermore is overly broad, as the law does not consider any form of organization–certainly not large corporations (later added by executive order–another problem we will get to in a moment). In 5 USC 551 (which includes FOIA information release requirements), “person” is defined broadly:

“person” includes an individual, partnership, corporation, association, or public or private organization other than an agency;

In Bayh-Dole, however, person appears to mean “natural person”–an individual, a man or woman, you or me, assuming you and I are not robots. This reading makes sense when one turns to 35 USC 202(d) and 37 CFR 401.9 to make sense of how inventors come to be within the scope of Bayh-Dole and its authorizes standard patent rights clauses (i.e., 37 CFR 401.14(a)(f)(2) requires contractors to make research personnel conditional contractors by delegating to them certain obligations under the funding agreement).

Bayh-Dole concerns individuals, nonprofits (universities and foundations), and small business firms.

But why the distinction between small business firm and nonprofit? There are only a couple of places where the distinction appears to matter. First, utilization reports from small businesses must be routed to the SBA. Nothing there. Second, and more interestingly, Bayh-Dole imposes a set of restrictions on nonprofit organizations that Bayh-Dole does not impose on small businesses or on individual inventors. That is, whatever you may read about how wonderful Bayh-Dole has been for university control of faculty inventions made with federal support, Bayh-Dole itself treats nonprofits (including universities) with a degree of distrust. Inventors have the most freedom under the standard patent rights clause at 37 CFR 401.9. Small businesses have the next best deal. Nonprofits bring up the rear. One reason, perhaps, was the anti-trust trouble that WARF had got itself into over its monopoly control of a process to create vitamin D.

With this background, now consider a problem in Bayh-Dole. The law is set up as a cascade of delegations. The law requires agencies to adopt standard patent rights clauses. These clauses must be created by a federal agency–now the Department of Commerce. The clauses, which may be tailored or even replaced, are inserted into federal funding agreements, where as contractors universities (and other nonprofits and small businesses) agree to the terms of these contractual requirements. In turn, these contractors are required to delegate obligations to individuals, making these “persons” conditional contractors–and actual contractors if they invent within the scope of the “planned and committed activities” of the federally funded project. An inventor so delegated becomes a contractor the moment she makes an invention–that is, fully conceives an invention, recognizes that conception as inventive, and reduces the darn thing to practice (constructively, by documenting it; or actually, by building it). A patent application does the trick, too.

For Bayh-Dole to work as intended, this 37 CFR 401.14(a)(f)(2) delegation is essential. Otherwise, when an inventor invents, it’s not a subject invention *until* the university or nonprofit or small business finds out and obtains (in some manner) title. Only then can the university or nonprofit or small business then notify the government that it wants to retain the title it has acquired. Without (f)(2), an invention made with federal funds does not become “subject” to the standard patent rights clause until an inventor assigns it to the contractor. There’s no obligation in the law for inventors to report inventions to their employers or to assign or license rights to the government. That all happens with 37 CFR 401.14(a)(f)(2)–the (f)(2) agreement, as I have been calling it.

There is another route to “make Bayh-Dole work.” This alternative route involves requiring employees to assign upfront all future inventions to their employer. That way, the moment an invention is made, it becomes the employer’s and there’s no need to delegate any reporting or rights management to potential inventors because they will never own what they invent. This alternative route is only “needed” if a contractor does not comply with the (f)(2) requirement. That is, the contractor breaches the federal funding agreement but finds a way to comply with other parts of the agreement. One can purchase an item or steal it. All the same if one can get away with it, I suppose. In the case of this alternative route, the “stealing” part is along the lines of telling employees that they must assign their inventions as a condition of federal law, when in fact what the employer is required to do by “federal law” (the federal funding agreement and 37 CFR 401.14(a)(f)(2)) is to delegate to inventors key obligations for reporting, filing patent applications, and assigning or licensing rights to the government.

Small businesses, unlike the prototypical large corporation, do not necessarily have employment agreements that demand all of an employee’s intellectual property. It may (or may not) be “best practice” to obtain all such rights, but small companies, like universities, might not have patent agreements with each employee that requires such assignment. In the case of universities, at one time, back when universities were more highly regarded by the public and by administrators, universities generally did not claim ownership of most faculty intellectual property, including inventions. The issue of university involvement generally arose when members of the faculty sought to patent their inventions. It was the act of patenting, not inventing, that received attention.

Now let’s get to the problem. If universities were operating with a diverse array of patent policies, many allowing faculty ownership, others directing faculty to work with external invention management agents, some contracting directly with one or more of those agents, and a few requiring assignment to the institution for management (and of those, only a handful trying to make money from licensing)–if all this, then the problem for Bayh-Dole was not how to make *university* patent policies uniform (and grasping), but rather how to make a uniform *federal* policy not disrupt the diversity exhibited by university patent policies.

In particular, this was the problem for the implementing regulations and the standard patent rights clauses. Bayh-Dole itself was actually not helpful. Clearly, there was no intent in the law to impose patent policies on universities. That’s not in the statement of objectives, not in the law, not in the regulations, not in the standard patent rights clauses.

There is nothing to indicate that Bayh-Dole intended to supplant university patent policies and practices with a federal requirement. There is nothing to indicate that Bayh-Dole saw the diversity of university patent policies as a problem requiring federal intervention. There is nothing to indicate that the federal government even required a university to have a patent policy.

The entirety of what a university was obligated to by receiving federal funding is wrapped into the standard patent rights clause, and even there, the actual obligations are sparse–education about timely reporting, delegation of reporting, filing, and government rights, timely reporting of inventions. That’s about it. Nothing about compliant patent policies, certification of properly executed patent agreements with potential inventors, uniform assignment of all federally supported inventions, and the like.

Furthermore, there is nothing in the law to indicate that small businesses were to adopt university procedures, or that universities were to adopt small business procedures, or that either were to adopt procedures of nonprofits, including the university-affiliated research foundations.

But what has happened? Universities have adopted large corporation methods: demand ownership of everything, let management sort out what’s of value, hold everything else just in case. They justify this based first on the overturned notion that Bayh-Dole vested ownership of inventions with the employer. When that explanation failed, they argued that to comply with Bayh-Dole, they had to implement present assignments. Or, much more conveniently, if they had changed their patent policy to require assignment (following the argument that doing so conformed to federal law), they simply pointed to the policy without indicated that the policy was the way it was based on an argument that was simply false.

The coarsest way, though, is simply to claim that university ownership of faculty work is a very good thing, and no matter the accidents by which such a policy has come about, it is a public good that administrators may attach themselves to just about anything a faculty member might invent, discover, author, create, collect, compose, or consider, so long as there appears to be either a way to own it or a way to make money from it. Anyone who resists is selfish, corrupt, pig-headed, and aiming to defraud the public for personal gain. Only universities should be allowed to do such things, and (so the argument goes), universities are much better at such things than are mere faculty members. This is the gist, by the way, of AUTM’s amicus briefs on the Stanford v Roche case.

The hole that Bayh-Dole left open in trying to navigate from a uniform federal policy on inventions to a diverse university patent practice was that universities would move to the corporate model of IP ownership rather than that small businesses and nonprofits would explore voluntary and open models pioneered by universities. The shift from uniform federal policy to uniform university policy was too easy a slight of mind. The lack of clarity in the law with regard to compliance left inventors at universities exposed to administrative power, and the outcome was that the law was used to destroy the diversity of practice, undercut the network of invention management agents, bring things in-house that had no business being in-house, and create a backlog and gridlock of fragmented intellectual property. In the current university patent licensing model, one financially lucrative license per decade is success, no matter if 2,000 inventions had to be sequestered for possible monopoly licensing to get there.

Bayh-Dole, by mixing university protocols with small business and nonprofit protocols, set up the administrative destruction of faculty research and innovation freedom. The attraction of turning all faculty discoveries into *capital* assets to be offered first (and generally only) to speculative monopolists in the hope of making huge amounts of cash proved too great. Making money from exclusive patent licenses is a strong rhetoric. Letting faculty choose how to teach, publish, disclose, patent, or share what they discover is a weak rhetoric–but it is just this weak rhetoric that supported the success that universities (via their faculty inventors and the invention management agents chosen by those faculty inventors) that provided the argument for Bayh-Dole. Not that Bayh-Dole was needed, or even that the idea that all federal agencies had to adopt a single approach to inventions made with federal support makes *any* sense–but however it was, Bayh-Dole was drafted without protection for faculty inventors. When the shadow of federal requirements passed over the university, administrators made their move to take ownership of whatever they could. In the process, they disrupted (and essentially destroyed) the innovation ecology that spawned Bayh-Dole in the first place.

What to do about it? Any university that chose to could simply roll back its patent policy and comply with the (f)(2) agreement without making other claims on faculty (or student) inventions. That would be the most creative and effective thing. But university administrators are like basset hounds that have fallen into a well. It is easy to get in, but next to impossible to get back out.

Similarly, the standard patent rights clauses could be changed to make clear that (f)(2) supersedes any university patent policy or employment obligations. By requiring a university to require each potential research inventor to make an (f)(2) agreement with the government, a university gives up any other claim the university has on the potential inventor. This, too, is the case with small businesses under the standard patent rights clause, but no one to my knowledge has considered the implications. Small businesses were invited (but not required) to participate in the same open environment for inventions that universities had created over the course of seventy years. But instead of an open environment, the forces of monopoly control moved in on the territory and imposed corporate protocols. Faculty fates were sealed when Bayh-Dole was extended by executive order to cover large corporations as well.

Clearly, by that time, large corporations were not being invited to participate in an open, diverse, largely voluntary faculty-led approach to patenting inventions made with federal support. Just the opposite happened. Faculty were told that the corporate approach was the only way to comply with Bayh-Dole. Federal law mandated they give up both academic freedom and freedom to innovate–but only for the *significant* work that they might do. The faculty could of course still hold personal rights in useless, worthless, and laughable stuff. The rest should be made into offerings for speculative capitalists–not for industries broadly, not for the public, not to assure non-discriminatory access, but rather to pitch investment funds on behalf of (largely) the already wealthy. To create a betting pool for speculators. The need for gazelle companies is not rooted in the idea that discovery is best made available through solitary fast-growing companies, but rather the need is rooted in a speculator’s hope of a substantial return on investment.

Speculation has a role to play, and certainly so does investment. But loans also play a role, without only the expectation of a repayment with interest rather than a wildly open upside. And as far as research goes, the idea of a grant–not even a loan, not even with a direct interest in repayment–is also important. At one point, grants were essentially gifts, made in recognition of the integrity and determined curiosity of investigators not working to make a buck, not trying to create a commercial product or venture–but explorers, off the beaten track, chasing down an intuition. The federal version of the gift–the funding agreement–greatly expanded after World War 2. It took twenty-five years for federal funding to become not just an accepted form of funding, but also the primary form of funding of university research. Attaching Bayh-Dole requirements to turn faculty discovery into speculative institutional capital was a genius move for capitalist-minded administrators, but a destructive move for the indigenous population of faculty researchers.

There is no going back. The lost city is lost. But there is going forward, and not in the direction suggested by present policy and practice. We can open up faculty discovery and invention. Freedom to innovate will look like administrative disorder rather than protection of inventor rights. Freedom to innovate will appear to allow mistakes and missteps–and it will. Freedom to innovate will appear to leave things “sitting”–and it will, but those things will be free things. Freedom to innovate also greatly reduces university costs, administrative overhead, institutional conflicts of interest, institutional liability, and public dissonance over the role of the university. The university need not play the huckster, nor monopolist, nor speculator, nor bully to industry, nor play favorites, nor be process-bound, nor appear the untrustworthy self-promoter, nor get caught in bed with sketchy business partners.

Universities didn’t need federal law to chart their course. But now that they have fallen down the well, they do need help to get back out. Amend Bayh-Dole to protect faculty inventors and their rights. Get university administrators out of the picture until they are invited by faculty to assist. All that’s needed is to enforce the (f)(2) agreement and make it clear–as is the case for subcontracts in Bayh-Dole–that a university cannot obtain ownership of an invention made with federal support as a condition of the award. That’s the start of a new era of open innovation supported by federal grants to faculty at universities.


This entry was posted in Bayh-Dole, History, Policy, Present Assignment, Stanford v Roche and tagged , , , , . Bookmark the permalink.