5. Further Implications of the Faux Bayh-Dole Act
The rise of the faux Bayh-Dole Act led university administrators with low status suddenly to see a way to acquire substantial power, using a claim that federal law gave them a mandate to take ownership of faculty inventions. Bayh-Dole was thus reconstrued as a proclamation of the supremacy of the technology licensing office over individual faculty investigators. In this faux version of Bayh-Dole, administrators ignored contractual requirements for implementation, including the key to the operation of the law, the (f)(2) requirement. As a result, the most favorable pathway for invention dissemination was suppressed by university policy. In its place arose a technology licensing office empire which has made quiet havoc of inventor initiative, the public domain for research findings, and the status of faculty members within the university. If Bayh-Dole is the killer, those constructing the faux Bayh-Dole arranged the hit.
Almost immediately after the passage of the Bayh-Dole Act, the actual text of the Act was replaced by popular accounts of what the Act was supposed to do. These accounts appear to have originated with people who were the key advocates for the law. The accounts express what these advocates apparently intended the law to do, but the accounts do not report Bayh-Dole in the form Congress ultimately approved, and certainly do not report what the law does as it was implemented by standard patent rights clauses included in funding agreements. The faux Bayh-Dole Act, however, was much more attractive to university administrators, and as promoted by the newly created organization now known as the Association of University Technology Managers (AUTM), the faux Bayh-Dole law became, if not the law of the land, certainly the law of university policy.
Because the faux version of Bayh-Dole was a made-up story sold to willing senior administrators and gullible faculty, there were various versions. One faux version claimed that federal law vested ownership of federally supported inventions with the university that hosted the research leading to the invention. Another faux version claimed that a university could “elect” title to an invention and that invention would without further formality become the property of the university. Yet another faux version claimed that federal law gave universities a right of first refusal, meaning that inventors could not assign to any other entity without the university first declining assignment. And finally, one other faux version suggests that federal law requires universities to take title to federally funded inventions, if only to offer that title to the federal government.
The mere presence of multiple accounts of the law suggests there was a problem in the university interpretation of Bayh-Dole. How could the law be so complicated that people could agree on the outcome but could not with any consistency identify a clear mechanism by which the law operates?
As we will see when we turn to the 2011 Supreme Court decision in Stanford v Roche, Bayh-Dole does none of the things claimed by the faux accounts. The law applies to federal agencies, not to universities, and not to inventors. Bayh-Dole provides no mechanism by which ownership of inventions is altered from federal patent law. Inventions are owned by their inventors until assigned in writing. Bayh-Dole requires no change in university patent policy and does not require university ownership of inventions made with federal funding. You will not find these disclaimers in most university patent or research policies or in university technology licensing office “training” or “guidance” about Bayh-Dole.
Any obligations and opportunities provided to universities and their inventors comes through a federal contract provision called a standard patent rights clause included in each funding agreement supporting faculty research. The standard patent rights clauses do not show up in the Bayh-Dole Act proper. They are part of the Code of Federal Regulations (37 CFR 401.14) and are inserted into most university funding agreements through another part of the CFR (2 CFR 215.36). As a federal contract provision, a standard patent rights clause is accepted by the organization receiving the federal funds. It is up to the organization to comply with the requirements of the clause and secure from inventors any required agreements.
As it stands, there is only one requirement in the primary standard patent rights clause that universities must make of prospective inventors: they must delegate to prospective inventors (1) the responsibility to report inventions subject to “contractor” ownership claims, and require these inventors (2) to agree to complete paperwork to allow patent applications to be filed, and (3) to agree to complete paperwork “to establish the government’s rights” in inventions. This requirement is set forth at 37 CFR 401.14(a)(f)(2), a provision that does not appear in Bayh-Dole.
(Anyone focused on the Act and not on the CFR fails to see half the situation–the part that involves implementation. The law establishes contract provisions. The contract provisions may be tailored by the agencies. The contract provisions as tailored convey the obligations to the university contractors. A number of otherwise bright academic attorneys have made this mistake and published defective analyses of Bayh-Dole. A quick way to tell whether a discussion of Bayh-Dole has a chance to be correct is to scan the article for references to the CFR. If all you see is 35 USC 200-12 citations, be skeptical. Of course, mere reference to the CFR is no guarantee that the writer is clued in, but it’s at least a positive indication they understand that the operative document is the funding agreement, not the law that authorizes funding agreements. In this discussion, where references to the CFR abound, you should at least set your mind at ease that if there are mistakes, you are on the right road to see them and call them out.)
The (f)(2) provision is the key to the operation of the standard patent rights clause, and thus of the implementation of Bayh-Dole. When this provision is implemented, under the terms of the standard patent rights clause, prospective inventors become “parties” to the funding agreement. They become “contractors” along with the university and their inventions become “subject” to the standard patent rights clause. As contractors, they agree to establish the government’s rights in inventions, which could involve either assignment of title or a grant of license for government purposes. If a university did nothing other than implementing the (f)(2) provision, it would have no other obligations under the funding agreement with regard to inventions. If inventions were made, the inventors as contractors would be obligated to report the inventions to their sponsoring agencies, and sign whatever paperwork the agency requested, including assignment of ownership. A university would not have to receive a disclosure document, obtain title, report anything to the federal government, or attempt practical application or seek to “commercialize” the invention. None of this is required under the standard patent rights clause.
Indeed, if (f)(2) is implemented, a second patent rights clause comes into play. This one is at 37 CFR 401.9. This provision allows agencies not to require assignment of inventions to the federal government–in essence, the law prevents an agency from claiming it is required to take assignment of all inventions made with federal support. The 401.9 clause is the most favorable to the contractor of all the patent rights clauses. The second most favorable is that pertaining to for-profit contractors. The most restrictive clause is that applied to nonprofit organizations–at the time, primarily universities and their affiliated invention management foundations.
Think about it. The federal policy spelled out in 37 CFR 401 is that inventors are to be given the broadest freedom to pursue their inventions and universities are to be given the least. Inventors need less oversight; universities need the most. This is not the story one gets from many university administrators–in their account, Bayh-Dole exists to protect the public from rogue, lazy, gullible, cheap, and selfish university faculty while providing a bonanza of potential value to university administrators in the form of patent rights stripped–and deservedly so–from university faculty (and staff and students and volunteers and collaborators). One can see how university administrators might claim Bayh-Dole is a wonderful law rather than the killer that it has become. In their view, Bayh-Dole gives universities absolute power to confiscate private property (personal inventions) and demand a ransom for that property from industry. This, as they would have it, is the core of “technology transfer” and the genius of Bayh-Dole. As public policy, it reads “A bureaucrat’s thumb in every possible pie.” Decide for yourself if this adds up to the “golden goose” that the Economist cooed over in a 2002 editorial.
The (f)(2) requirement, when implemented, has another important effect. Because this requirement is part of a federal contract, it takes precedence over contracts enforced at the state level, such as employment agreements and university policies deployed with contractual force. Thus, whatever a university policy might require in the way of inventions, the (f)(2) requirement supersedes that policy wherever there is overlap. A university cannot waive by policy any interest in inventions made by faculty and also require faculty inventors to sign paperwork to establish the government’s rights in inventions. A university could, however, require all faculty to assign all inventions made within the scope of their employment to the university. The (f)(2) agreement would still operate, but only when a university chose not to enforce its claim or chose later to re-assign an invention to the inventors.
But here is the irony. Universities–not a one of them, as far as I know–has ever implemented the (f)(2) requirement. The whole lot of them are in breach of their federal funding agreements, and have been for the entire 35-year run of Bayh-Dole. In place of the (f)(2) agreement, universities have substituted a patchwork of policy statements, private contracts, and various sorts of threats, including litigation against their own faculty inventors. No private agreement, however, makes prospective inventors parties to the funding agreement, obligates those inventors to establish the government’s rights, or obligates the inventors to sign paperwork needed to file patent applications for anyone other than the university.
While universities might, as an employer, claim ownership of employee inventions, also as an employer universities must follow federal patent law. And federal law is clear–mere employment is not sufficient grounds for an employer to claim ownership of an employee’s invention. There must be some other agreement, a “patent agreement,” that covers inventions. Some states have enacted laws to ensure that an employer does not use its power of employment to claim more than it properly should.
When Bayh-Dole was passed, most universities had patent policies that made only narrow claims on ownership of inventions made by faculty. Often the policies focused on equity rather than ownership, and sought a share of any commercial income obtained by an inventor, based on a showing of university support beyond that normally provided to faculty to do their teaching and research. Where the concept of employment came up, it was focused on duties that were assigned by the university, or additional duties that were specially negotiated in exchange for “substantial” university support for developing an invention. University faculty were less “employed” than “appointed.” The university did not assign them their research, review their proposed publications, decide when they had done acceptable research or not. Faculty, for their research, were free agents. Faculty did not even have to use university facilities or funding to do their research. In the case of extramural research, most universities had policies that required the external sponsor to compensate the university for the full cost of releasing faculty members to conduct the research. Hence, the division of extramural research budgets into direct and indirect costs.
The university does not assign faculty to conduct extramural research. The faculty identify a source of support and request permission to be released from “official” duties (such as teaching) to commit time to the externally sponsored work. It is the faculty member who writes the proposal, who chooses the sponsor, and who chooses her collaborators. The university obtains the funds from the sponsor on behalf of the faculty investigator. It is a matter of special performance, not university assignment of employees to conduct research. If the university does not make the investigator available, the sponsor may withdraw its support. Thus, faculty receiving funds from external sponsors are not paid with university funds nor is the university, by its own affirmation, out any resources for allowing the extramural research to be conducted by its personnel. There is thus in general neither university direction nor compensation provided to faculty for their extramural research activities.
While faculty may be called “employees” in university policy statements, they are clearly not employees for matters of their research and scholarship. Faculty serve no university master who decides what will be studied or published. They are not “employed” for these activities unless they expressly agreed to be employed in a particular situation.In the same way, members of a state legislature are paid for their service, and may even be called employees of the state, but the state asserts no control over their activities as representatives of the people. University faculty are appointed to their positions, even as they are “hired” for their “official duties.” Those “official duties” do not involve extramural research or publishing scholarly articles, though extramural research and publishing may be considered for tenure or promotion.
Universities managed this distinction between faculty member and employee for purposes of extramural research by implementing policy statements that required faculty to honor the terms of research agreements under which the faculty received support. The policy read, in essence, if you voluntarily agree to conduct research using a sponsor’s money, and the university agrees to release you from other duties to do this research, and the university receives the funds and enters into the funding agreement on your behalf, then you agree to comply with the terms of the funding agreement. This arrangement made formal a voluntary agreement under which faculty agreed to special terms and conditions on their work–in this case, terms and conditions agreed to by both the university and a sponsor.
Where a sponsored required ownership of inventions made in under the funding agreement, a university might concur or might seek to negotiate a different outcome, depending on the nature of the research, prior work done at the university, and public policy–some inventions ought not be owned at all, even if externally sponsored. In any case, no research would get done if the faculty investigator rejected the negotiated terms between the university and the sponsor. As the university interest in obtaining control of faculty inventions waxed, university administrators got the clever idea of insisting on university ownership of inventions when negotiating with an external sponsor, and forcing that condition on faculty investigators rather than responding to a sponsor’s request for that ownership on behalf of a faculty investigator’s refusal to assign inventions to the sponsor. Rather than solving select negotiation problems by serving as something of an escrow agent, university administrators sought to own for the advantage ownership might bring in making money from patent positions–even with sponsors putting up the money for the research. One might see how industry and nonprofit sponsors of research might be dumbfounded by this change of position in their collaborations with university faculty.
But federal funding was different. While faculty had the freedom to create the statement of work that they desired, universities had little room to negotiate federal funding agreement terms. They could negotiate their indirect cost rate. And a few negotiated, in the case of the Public Health Service, an Institutional Patent Agreement to cover inventions. Bayh-Dole wiped out the IPA approach and replaced it with a uniform federal policy on agency interest in inventions made with federal support.
With Bayh-Dole, there was no sponsor requirement to own inventions, nor was there any university basis to claim inventions within the research agreement. It was a federal contract, set out originally in Circular A-110 and later incorporated into 2 CFR 215. The standard patent rights clauses were incorporated by reference by 2 CFR 215.36(b). Under the standard patent rights clause, a university has absolutely no special ownership interest in work done by its faculty. This is the problem that patent administrators at universities set out to solve. But rather than address the problem directly, through faculty governance, they sought a clever work-around. Bayh-Dole, they claimed, vested (or whatever faux variation one wished–they all “worked”) ownership of inventions with the university. Thus, faculty inventors had to assign their federally supported inventions to the university. All that was needed was a change in university policy–represented as required by federal law–to confirm that university inventors understood their obligation to assign.
The policy claim, of course, was totally wrong, a misrepresentation of the law, and worse coming from those who had worked as expert advisors as the legislation became law. For shame! But it was a clever ruse, and it has worked for more than three decades, slurping up faculty inventions and handing these to bureaucratic organizations, swamped with assets, finding themselves unable to secure licenses at nearly the rate claimed as Bayh-Dole was debated in Congress. One estimate (from the University of California) is that about 0.5% of inventions result in a product commercially available. At the time Bayh-Dole was passed in 1980, university licensing agents claimed a rate of about 30%, and blamed the government for a rate of about 5%. Outcomes under Bayh-Dole are now 10x *worse* than the problem Bayh-Dole claimed to be the answer for. Surely Bayh-Dole is a killer, nothing less.
To implement Bayh-Dole, university administrators refused to implement the (f)(2) agreement. Instead–and entirely skew from the purposes of (f)(2)–they claimed federal law gave them ownership of inventions. They then often took their claim further. Where the standard patent rights clause restricts its interest to an invention that “is or may be patentable” and further must be owned by the contractor, university administrators expanded policy claims to inventions whether or not patentable (giving rise to entirely strange theories of ownership), to copyrights (except for “traditional works of scholarship,” with the aim to exclude software, databases, web sites, social media, and any other form that current scholarship might take, as if scholarship were frozen in time, circa 1980), to data, and to “know-how,” to “tangible research property”–essentially anything of any significance that a faculty member might think, invent, create, produce, develop, collect, recognize, or build–so long as the work involved university facilities, funding, or the general area of the faculty member’s expertise.
One patent administrator went so far as to explain to me that the expansion from federally funded patentable inventions to other things that could be called inventions, and to non-inventions, was to be “fair.” Why should those with federal funding get all the benefits of a university licensing operation and all the others miss out? If some should be required to participate, everyone should be required to participate. It is only “fair.” You get the flavor of the quality of thought going on in the technology transfer field.
One might pause to ask whether an assignment of personal property required under false pretenses might be void. For state universities, one might further ask whether such demands for state ownership are eminent domain actions, requiring both due process and just compensation. Given that the vast majority of university inventors who have had their inventions taken by the state have received nothing in compensation, and certainly not “just” compensation reflecting the fair market value of their inventions at the time they were taken, one might think there is a wonderful class-action lawsuit opportunity on behalf of tens of thousands of inventors. And one might well be right for so thinking. A judgment against the universities might total billions of dollars.
There have been three adverse effects of the failure of university administrators to implement (f)(2).
First, university faculty have been denied standing as parties to the funding agreements under which they receive federal support. In turn, what should have been federal obligations and opportunities have been blocked by assertions of institutional requirements. As a consequence, the most liberal pathway for the deployment of inventions through the inventors themselves, enabled by 37 CFR 401.9, was shut off and not allowed to operate. Faculty investigators have been cut off from building innovation networks of company scientists, invention management agents, and sources of investment. That network has been co-opted and displaced by an institutional network that cannot operate nearly so broadly, selectively, or opportunistically.
For all the emphasis on system and process made by institutional patent officers, much of what counts for innovation breakthroughs comes about by personal connections and chance opportunities favoring the prepared mind. The failure to respect and implement the (f)(2) agreement has meant the collapse of faculty-led networks, to the point that people wonder what it is that I must be writing about. Surely not something that exists in the real world. And they are right, of course–it is not something that exists now in the world created by university patent administrators. They destroyed it, often without even noticing, or when they did, they have called called it “conflict of interest” or “criminal” or “defiance of authority” or “insubordination” or “violation of policy” or “dangerous” or “selfish, rogue, foolish.”
These terms are used today to mark the burial place of the faculty innovation initiative that led to the creation of Research Corporation, which gave the idea for affiliated research foundations, which provided the shape for the National Science Foundation, which led to the expansion of federal funding for research with an idea to innovation beyond the ken of established interests–in the military, in government, in medicine, in publishing, in the broader corporate world. That expansion created the interest in technology transfer to mark the success of the model, the passing of discovery from lab to the public, to industry, to product, to social welfare, economic vitality, progress, and innovation–introduced changes to the established order. Hey, it’s a vision, so it has to wax a bit with the rhetoric.
The licensing offices now planted at universities everywhere should be the staunch, vocal stewards and guardians of faculty initiative, for that initiative is what has given them life and significance. They owe faculty their livelihoods. And yet it is just these offices where few are willing to defend that faculty initiative. No one speaks out. They fear they will lose their jobs, their careers. It is ironic, even sad, perhaps even shameful, how deep the disrespect for faculty initiative runs in licensing offices, offices of research, professional organizations such as AUTM, and university front organizations such as COGR, AAU, and APLU.
Second, inventions that would have moved rapidly to the public domain as a result by both faculty inventors and federal agencies not to pursue patenting were caught up in the net of university administrator claims–basically, greed. Administrators used patent rights to prevent early research use and evaluation by industry, holding out for a company or venture speculator willing to pay well to maintain a monopoly position against not only the rest of an industry but also often against any other university’s research programs and publications.
This move by administrators shut off the flow of research ideas to the public domain and to a domestic commons like the ones created by federal agencies that licensed non-exclusively–a pro-competitive approach, rather than a pro-monopolist speculator approach. For the monopolist, having possible access to most significant findings at universities supported with federal funds is manna from heaven. If the monopolists worship a god, it is Faux Bayh-Dole, and that god’s leading prophets are AUTM members. The effect on public policy was to divert federal research, as a default, to the trough of monopoly speculation rather than broad, collaborative (and competitive) public access.
Third, the work-arounds used by patent administrators aimed to destroy the independence of faculty in their research. Administrators used faux Bayh-Dole to change university patent policies to require assignment of inventions, claiming federal law. Even if they were wrong, they now argue that policy is policy, regardless of the premise on which policy was adopted. In the new regime, faculty are employees for their research because policy gives ownership of inventions and other assets to the university, and if the university owns these assets, then it has a right of control over disclosure, publication, and even use.
There are then two policies competing within the university for the intellectual life of the faculty. One policy is framed by the conventions of a faculty–independent, with academic freedom, not beholden to the institution or the state with regard to the choice or conduct or teaching or publication or use of research. The other policy is framed by the faux Bayh-Dole Act, claiming ownership of all key findings produced by faculty (and others), and with that ownership making faculty beholden to the institution or the state with regard to especially the publication and use of research.
If faculty members are appointed to their positions with one understanding–a meeting of minds between a faculty member and a dean–can the university then claim that this meeting of minds, this agreement, is merely for show and the “real” agreement is one buried deep in the patent policy of the institution, and that policy is imposed as a “condition of employment” even if it was never a “condition of appointment”?
That’s the question–which version of faculty standing will control. The answer is not clear. The law of contracts appears to be with the faculty, but law costs north of $100,000 to prove on university administrators, and administrators show no signs of changing practices even when they are declared wrong by the courts. The law of power acting with impunity then appears to be with university administrators. They may lack status, but they are determined to hold to power. Oddly, it is just this combination of low status and high power that creates the worst behaviors, and this, too, is a consequence of the Bayh-Dole Act, to turn those who would provide service into overlords.
University faculty cannot legally use their inventions (and often, that term is broadened to include non-inventions–right in the policy definition of “invention”), without bureaucratic permission, outside the university, or at any other university, or even personally, without a license. University bureaucrats are not known to offer free licenses. In Marxian terms, this is alienation of labor writ large. In the view of patent administrators, faculty researchers are no longer expert craftsman, working to produce findings they both own and are responsible for, independent from institutional (and state) control as a matter of public policy. They are now just paid to invent, and their inventions are owned and controlled by their institutions. Their only incentive is that if the institution succeeds in obtaining a monopoly deal, the inventors have a share in the spoils. They have little incentive to teach others how to practice their inventions, because that practice would be infringing. They have less opportunities to collaborate with other researchers at other institutions where patents block future work and future use.
Bureaucratic control has become the high point of university innovation policy. Research exists to make money for the institution. The money that is made is invested in more research and licensing. Almost never in instruction. Never in support for research and education outside the host institution. That is, never for an existing public need.
The patent administrators who have created this system are proud of it. They have successfully turned a gesture of public optimism in university faculty as leaders in adapting curiosity to public benefit into a self-interested institutional effort to use ownership claims to obtain money from industry, and if not industry, then speculative investors, and if not speculative investors, then state and federal investment funds for economic development, in the hope that speculative investors or companies will show up at some later date, if only startup companies can be kept alive on public welfare for three, five, ten years. The administrators have created a world starkly different from the one they put forward in their annual reports, in their praise of Bayh-Dole, in their self-congratulation on the growth of university licensing programs, measured by the number of patents (stuff kept from the public domain), the number of licenses (successful extortion of industry or complicity with speculative monopolists), and the income (often mostly from one or lucrative two deals, but made to look like hundreds of deals).
What the licensing offices do not report–and have never publicly reported–is what Bayh-Dole asks for: for each invention claimed and patented, what is the status of practical application? In this regard, Bayh-Dole itself is at fault. Reports of utilization are exempted by Bayh-Dole from public disclosure law. Reports are not required unless an agency expressly asks for them. March-in requirements to remedy failure to achieve practical application were so convoluted–unlike the IPA accountability that Bayh-Dole washed away–that no one has successfully used them in the entire history of Bayh-Dole, and only a handful of companies were fool enough to try.
Under the faux Bayh-Dole, university administrators have largely destroyed a robust research public domain, delayed by years access to research findings, and suppressed faculty initiative and personal innovation networks. In doing so, they have also refused to report publicly the core metric that Bayh-Dole expects–timely practical application of each invention claimed and patented. Thus, we have a system that puts bureaucrats in control of research findings created with billions of dollars of public funding per year, with no public reporting that would indicate whether this new system is producing the results claimed for it.
It is clear how the faux Bayh-Dole Act is wrong as a matter of public policy. Its results have been destructive and its premises run against reason. But how do we know that the faux Bayh-Dole Act is also technically wrong? That’s where we get to Stanford v Roche.