Of the seven pillars of university new innovation practice, freedom to innovate is the most important. University policies on research and invention were at one time liberal. That is, faculty and students had the freedom to publish, experiment, discuss, collaborate, and commercialize as they chose. University administrations did not demand ownership of inventions unless someone was expressly hired to invent on behalf of the institution. Faculty generally were not, and are not, hired to invent on behalf of the institution.
Equitable Sharing Rather Than Compulsory Ownership
The default premise of university invention management once was “equitable sharing.” If a university administration put in extraordinary resources to support a faculty member’s efforts to develop an invention, then the faculty (generally) were to decide whether and how much and by what means the university’s support should be recognized. That recognition might take the form of acknowledgement, repayment, a financial interest, a shop right, a license to practice, an equity interest, a co-ownership position, or assignment of ownership, depending on the arrangements that had been negotiated. In a few cases, university faculty banned patenting (mostly in medical schools), and in a few others, university administrations might hold patent rights to ensure that everyone had access to the invention–that is, not for “commercialization” or as “a source of revenue” but rather so those working with the invention could push back if anyone tried to patent improvements and seek to monopolize a line of natural development and application of the invention. In that old way of thinking, commercialization is more likely to take place because everyone has access, not because a speculative investor pays for exclusive rights and locks up the invention for a couple of decades.
It is somewhat horrifying to find that in fifty years, university administrators can go from a commitment to a public mission of broad access to making it a matter of policy to sell out university-hosted discoveries to monopolist speculators.
The Old Model Was the Basis for Changes that the New Model Exploited
You may be skeptical or you may believe that we have come a long way since those days of last century. But you need to understand that it was the freedom to innovate model that was producing the statistics that were used to push through Bayh-Dole. The model that was proving itself productive was the freedom to innovate model, not an institutional ownership and control model. The approach that Vannevar Bush used in World War 2, and advocated as the foundation of what would become the National Science Foundation, was one of basic research and lateral, unexpected application. It was not a model of ownership and development in the direction of the subject area of the research. It was a model of opportunity, adaptation, coordination, and diversity. It was an approach–or rather a range of approaches–that “worked.”
We know more now than we knew in 1980. We also know much less, given what has been forgotten, disrupted, and suppressed. We know that institutional ownership of inventions for the purpose of exclusive licensing shows no prospect of performing at anywhere near the claimed levels of productivity. The “successes” that are promoted at present are the ones that “made it through” the gauntlet of bureaucratic adversity. There is no evidence to indicate that university-hosted discoveries are advanced rather than diminished by a compulsory program of institutional ownership and control. Indeed, universities have provided virtually no relevant information regarding the discoveries, inventions, software, datasets, materials, and information that they have demanded to own and subsequently have done nothing with. When normally self-promoting administrations don’t publish the backing data, you can assume the worst.
The current university compulsory ownership approach is an administrative process imposed on creative work that prevents access and dissuades interest by technical and investment audiences that might immediately work with reported results. The delay, uncertainty, and expense of obtaining a proper license (remember, “license” to a university licensing office means legal paperwork, not a simple permission or a promise not to sue) has quietly degraded the interaction–and the trust–between faculty and their counterparts in industry and in professional practice. We also know that organizationally open approaches can be highly productive, even if there is no contract demanding a financial return to the institution as a pre-condition of use.
But for all this, we have forgotten a great deal, too. We have lost the infrastructure that existed to manage inventions and commercial activity within academia. What once was a time-to-time review by faculty of their peers’ efforts to obtain patents has become a fifty-patent-a-year habit run by administrators, often with minimal communication with anyone regarding the status of the patents that have been obtained. And no, the AUTM licensing survey does not provide the requisite information. We have also lost the diversity of opportunities, as everything is now channeled through a bureaucracy. Yes, it presents itself as a caring, hopeful, supportive bureaucracy, but it is still a bureaucracy. Finally, we have lost the self-governance by which investigators and their peers debated and managed their approaches to patenting, commercialization, and research. That self-governance has been replaced by policies now mostly imposed on the faculty by administrators. Often these policies ignore intellectual property law, federal contracting requirements, and market realities. Most of the policies are so sloppily drafted that they border on incompetent. What is typically super clear, however, is that administrators are determined to own everything they can, and believe that by owning, they will create through licensing a new and substantial source of revenue for the institution.
What To Do? 1) Get the Data
What then to do? How to break the stranglehold of university bureaucracy on research findings? First, there’s the data. Make public records requests (these will be stonewalled, delayed, appealed, and redacted, but make them anyway). Ask for information pertaining to the status of each technology that the university has claimed ownership of over the past three years. For each, list an identifying case number, the general nature of the case (invention, software, material, other), date of patent application, date of patent issue, date of license, whether the license had any exclusive component, and finally the date of first commercial sale or use. This information does not reveal anything about the technology itself, does not identify the licensee, does not even deal with money. The issue is, does the university even have this information on file? Surely it must, other than the date of first commercial sale or use–which is required information for Bayh-Dole reporting, but federal agencies have to ask for it.
With this data, a picture of how much of what is claimed is making it through “the process” to commercial use will become clear. From what I can tell, most university-owned faculty and student work goes nowhere. Most of it “sits.” Worse, because it is being held for ransom, it sits in poorer condition than if it had not been owned at all. Yes, it might get a patent attached to it, but that patent increases the barrier to access and reduces the number of possible early adopters and delays initial interactions. A lab may have only a matter of weeks to months that it will keep an experimental set up in place. After that, the lab moves one, students graduate or shift to other projects, and all one is left with are the photographs and the memories. Missing that first critical window of opportunity is all but guaranteed by the administrative “process” of claiming ownership.
What To Do? 2) Change Policy Interpretations
Second, one can push for changes in the interpretation of policy. Most university patent policies are written so badly that there is always more than one interpretation available. Exploit this defect. Adhesion contracts are generally interpreted against the drafting party. Not all university patent policies are contracts, but those that are contracts are also, generally, of the adhesion variety. No one gets to negotiate the terms of a patent policy as part of appointment or employment arrangements. Faculty then have substantial opportunity to establish the proper interpretation of the patent policy, for it is what they reasonably understand to be the case, not what administrators who wrote the policy claim.
Check your patent policy. If it permits administrators to interpret the policy, or provides an appeal process with a university administrator as final authority, then the policy has a structural weakness. If appeals are only of the form of a waiver from policy, then the policy might hold up. But if the university can waive its own obligations, or suspend the policy, or decide on interpretation, or change the policy in the future at its discretion, then the policy is of the form of an “agreement to agree”: in essence, the policy simply delegates authority to administrators to decide what to do. That’s no contract–it’s just an assertion of power. The Shaw v. Regents court made clear that a university cannot assert what it intended for policy at some later date, and also that faculty have a right to rely upon a policy as it is set forth. If a patent policy has any of the above weaknesses, it is likely not a contract, or is an adhesion contract, and in either case, faculty have plenty of room to require immediate reform in interpretation of its words.
Key terms are “official duties,” “scope of employment,” and “use of resources.” Official duties are those identified in one’s appointment letter or employment contract, or have been formally assigned. Such duties rarely if ever include making inventions for the institution, or writing software for the institution. Same for “scope of employment.” Employment is not everything a faculty member decides to do. Employment, for intellectual property purposes, requires more than simply payment for a certain set of services. For copyright, especially, employment includes concepts of agency–over what actions does the employer exercise control? For faculty, the university rarely controls anything–does not require work to be done in a university lab, does not review and approve publications, does not choose or direct the research. In short, university employment for faculty is an amazingly narrow term. All you need is for university administrators to agree to stop abusing their power by expanding the definition beyond what actual written agreements with faculty set forth. Of course, one has to put it diplomatically: administrators are not required by written policy to expand claims beyond those that have been mutually agreed upon with each faculty member in the written terms of their appointment.
In dealing with scope of employment, it is important to call out federally supported research. For such research, faculty typically have to request permission from the university because extramural work is outside official duties. A faculty member asks to be released from certain of those duties. Sometimes the faculty member has to use funds to “buy out” teaching duties. Federally funded research has its own conditions for scope when it comes to inventions–these are clarified in 37 CFR 401.1, which sets out the conditions on which a federal agency is entitled to consider a given invention “subject” to the standard patent rights clause in the funding agreement that governs federal support for the research. If the invention is not within the specified scope of work, and does not distract from the conduct of the grant work that is specified, then the invention is not “subject.” If the invention also is not specified within the official duties of the faculty members involved, then the invention should be “free”–also not subject to a claim by the university under its “official duties” or “scope of employment” clause.
A similar argument holds for use of resources. If a university has accepted the funding agreement, then the university also has committed, to the federal government, to provide the resources necessary for that research to be undertaken. The university also is compensated for doing so. The university cannot then turn around and make access to those same resources a “perk” of faculty employment that requires assignment of inventions. The university cannot make those resources part of a barter with faculty researchers over rights to inventions because the university has already agreed to provide those resources, per the funding agreement. There has to be some other arrangement regarding inventions, not tied to use of resources made available for the purpose of the federal award, and not tied to employment, since work done in a federal award is almost never, formally, directed at creating assets to be owned or even used by the host institution.
Indeed, the standard university funding agreement at 2 CFR 215.37 [now 2 CFR 200.316] makes it clear that the university is to act, generally, as a trustee for inventions even when it does acquire them in the course of research. That is, if a university were to demand ownership of inventions made in federal funding, then under 2 CFR 215.37 [2 CFR 200.316], the university also agrees to hold the inventions for the benefit of the beneficiaries of the research–and that generally does not mean licensing for the purpose of generating money for the institution.
More broadly, “use of resources” clauses are generally overdrafted. What constitutes a “resource” and how is one “used.” Are office desks “resources”? How about the electricity at the wall outlets? (One university I know went so far as to argue that it was a misuse of resources to charge one’s personal laptop at a university outlet–sigh). It is an easy matter for any university administration that chooses to do so to make a narrow interpretation of both “resources” and “use.” A narrow interpretation of resources might be any resources not otherwise generally available to faculty, and for which a specific arrangement regarding access is required. Thus, if there is a sign-up protocol to use a scanning electron microscope, and the condition for sign-up is that there is a fee for use (such as a recharge center) and a faculty member wants use without paying, then perhaps that’s a use of resources that might give rise to a claim for later reimbursement from commercial activities (or ownership, however one has to do it). Rather than eliminating all claims, just narrow the scope of “resources” and “use” to reasonable, defensible interpretations.
What To Do? 3) Change Policy
Policy changes take longer, so it is better to start with changes in interpretation. But if administrators also support a change in policy (and they should! good grief, the present approach is so expensive, fraught with institutional conflict of interest and liability, and the cause of much bitterness and distrust!), then policy change might take only a few months, especially if the policy is rolling back requirements imposed on faculty that have not had the outcomes claimed for them.
A fundamental reason to review policy is the Stanford v Roche Supreme Court decision. That decision made clear that the Bayh-Dole Act never did give universities ownership of faculty inventions made with federal support, and never did require universities to require assignment of such inventions, and never did strip faculty of their rights to such inventions. Many university patent policies were changed to compulsory approaches with a false claim regarding the Bayh-Dole Act. The Supreme Court decision should all but require a policy review to eliminate any language that would be contrary to the decision. Further, it should require a review to determine the basis on which compulsory assignment language was introduced.
University compulsory claims to own inventions (and other assets) can be simply rolled back. Must and shall constructions can be replaced by may and “are encouraged to.” Copyright policies often still retain a requirement that if a university administration is to claim ownership, there must be a written agreement on the disposition of ownership up front, before the work is started and before otherwise reserved or designated university resources are made available. A similar provision works well, too, for the patent and materials side of research results.
In place of ownership claims, redevelop faculty committees–perhaps specific to each school or college in which inventions and commercialization might figure–to review the circumstances of development and make a recommendation for equitable sharing in the case of commercialization activity. There are plenty of past patent policies with the requisite policy language. It’s just a matter of selecting, adjusting, and approving that language.
What To Do? 4) Eliminate the Patent and Copyright Policies
Of course, roll back to freedom to innovate can be made even easier by eliminating the patent and copyright policies altogether. This step is not as radical as it sounds.
A patent policy is not required by Bayh-Dole. Nor is a copyright policy required for any federal funding. Compliance is generally not an issue. For grants with IP provisions that faculty investigators want and a university agrees to accept, all that is needed is for those involved in the grant work to agree in advance to comply with the terms of the award. The policy and the award are one, for the purposes of the award.
Grants aside, most university IP policies are so utterly, deeply flawed that no policy is much better than what presently exists–contrary to law, built on misrepresentations, bungled definitions, contradictory provisions, huge gaps and loopholes, unconscionable assertions, requirements never implemented or inconsistently practiced. Let me state it again: most university IP policies are rotten, foolish, nutter documents. I have read scores and scores of them. They are wretched things. Let them go.
Get rid of patent policy rather than trying to fix it. Rely on federal defaults and primary paperwork (such as appointment letters, employment agreements, and extramural funding agreements). In most cases, everyone–faculty and institution alike–are way better off.
If one wants to write a policy to govern a technology licensing office–that’s fine. Technology licensing offices in freedom to innovate will discover they are needed, their services are appreciated, and they are for once not overworked dealing with crappy stuff hauled up by the policy’s bottom-trawling net that no one wants to deal with.
Face it, really good stuff where faculty inventors have private resources available to support their efforts do not need an institutional process. They need direct action. For these, a “review for equitable sharing” policy on a case-by-case basis works just fine. There may be only a handful of these a year at even the biggest research operations, and a faculty committee will find the work rewarding and invigorating. For iffy stuff, no process is going to help–that stuff needs to circulate, to get used as it might. Any interest or use might set up some consulting work, or a bit more research support, and that’s fine. No institutional patenting needed. The sweet spot for voluntary collaboration between inventors and an institutional licensing office is for stuff that’s pretty good, but will take time, is the work of multiple hands, and will be primarily developed within the university. There, the university can act as an escrow agent, a steward, a trustee to hold rights on behalf of these involved, for the benefit of those that the research aims to help. That’s a great spot to be in. It’s even noble. It’s the kind of program I would work for in a flash, and I bet a number of other tech transfer professionals would, too. It’s the kind of program where your skills and connections, not your bluster and policy threats, get you business.
Freedom to innovate allows this sweet spot to develop. It also creates opportunities for a diversity of other licensing agents to affiliate with the university to provide custom services. Faculty with specific needs can then be referred to this network of local, national, and international agents, with the university playing the role of watchdog, looking out for faculty who choose to take their work through such a network. A key marketing strategy is to refer problem customers to the competition–what doesn’t work for one licensing shop might work for any number of others. There is no reason to try to build out a licensing operation that is “full service” if by “full service” is meant “an area of great expertise plus a lot of lousy service for everyone else because we are afraid of letting something valuable go on its own and would rather bungle it than release it.”
The simplest way to transform a patent policy is simply to eliminate the assignment requirement. It may be only a single sentence in the policy that goes. Another way is to make the scope voluntary. Inventors may choose to come within the scope of the policy, or not. That does the same thing. If an inventor wants the benefit of policy in exchange for assignment, there it is, all fitted out in its inconsistent, malevolently drafted finery for anyone who wants it.
There is nothing wrong with a licensing office identifying how it pays out on a royalty sharing schedule, or what it expects of itself and its inventors when it accepts IP for management. Just cut the compulsory bit, and then work with the rest in whatever way works best. The biggest change for a freedom to innovation IP office is that it does not have to accept work for management. It does not have to do a commercialization review of everything that might meet some minimal definition of “may be patentable” or the despairingly bad “whether or not patentable.” Imagine that–to focus on good stuff, with inventors that have chosen to work with you. What a wonderful foundation for a collaboration to get creative work in a condition for further development and use by the public. Who is it that does not want this?
What To Do? 5) Change the Law
For public universities, there is one more option. Introduce a law that prevents the state from using its power of eminent domain and its power of employment of faculty at its public universities to demand ownership of inventions and other assets created by faculty, staff, and students in the course of their university-hosted work. Think of it as part of an innovation “bill of rights” that limits the power of the state, for the benefit of the state.
As for copyright, a state claim of ownership in faculty, student, and staff scholarship strikes against academic freedom. Here I do not mean merely the academic freedom as the AAUP has developed it–a fine body of work; rather, I mean the deeper problem of the state’s control over science and technology. This debate was active in the 1960s and 70s, and has since faded but not been resolved. When the state controls scholarship, bad things happen. When the state is conflicted between reviewing claims for new work and trying to exploit those claims for money, bad things happen. When the state tries to pick winners and losers, not just regarding technology but also speculative investors, bad things happen. In these cases, the state gives up its role as adjudicator, as the foundation for appeals for fair play and justice. It becomes the self-interested state, the state mixing it up with its favored business partners, to the exclusion of the rest. Where then is the Attorney General? Where is the State Auditor? Where is the public recourse? Where is the state’s accountability?
No, all this goes, replaced by propaganda about how wonderful the state’s programs of licensing university IP have performed. The result is the fraud at the University of Utah, and at the University of Washington–and we are talking fraud in the tens to hundreds of millions of misspent, misaccounted dollars–and at any public university that publishes technology licensing and economic development spin and does not provide also the underlying data to show not only that the claims are supported, but all other data to show the context in which the claim has been made and the form of analysis that led to the claim. While a private company might spin all it wants with selective release of information, state government is not a private company, and when it starts to behave that way, we should all be concerned.
Some states have laws pertaining to employee inventions. A simple modification of these laws will make it clear that universities cannot use employment of faculty and students to claim any implied benefit under an employee invention protection statute.
Of course, inventors are always free to contract with universities to manage their work. The purpose of the law is to place that negotiation on an equal footing.
In Washington state, where freedom to innovate legislation has been introduced but not made it out of committee, the bill took this form (my emphasis):
Public universities:
16 May provide university-administered facilities and other
17 resources to students, staff, and faculty for their use, including but
18 not limited to use in extramurally funded research, such as through
19 grants, contracts, or donations, consistent with the public mission of
20 the university, if the university does not impose a requirement to
21 assign intellectual property or other intangible assets that may
22 thereby be created, or assert a financial interest in such assets,
23 except as may be required by an extramural sponsor of research;
24 (b) May require assignment of, or have a financial interest in
25 intellectual property or other intangible assets developed by students,
26 staff, or faculty only through agreements entered into on a voluntary
27 basis;
28 (c) May not, in employment agreements with student or faculty
29 employees, require assignment of or assert a financial interest in
30 their inventions;
The simplest approach is to change interpretations and see how that works. A university could make the transition in a matter of weeks, with a notice to the academic senate, a review of the new interpretations, and a date for implementation. For such a change, all that is needed is a vice provost/chancellor/president of research to step forward and engage faculty on the matter. Whatever the unknowns that show up in freedom to innovate–and there always are new things–they are well worth the freedom that the institution has released back to its creative personnel.
Yes, there will be new importance to conflict of interest issues, but the university, by freeing itself of much of the burden of its own institutional conflicts of interest, will be better positioned to address any such conflicts. Yes, there will be concerns over commercialization efforts by faculty and students. But these can be addressed by the faculty in the units involved directly, not by a licensing office, that has other matters to focus on. Yes, there will be concerns that some inventions do not make money, and had they been patented, they might have. But again, in all probability, most of those inventions would have not made money, and would have cost money, distracted good people from their good work, and been in the end a boondoogle. Better to let some things float freely than to imprison everything out of fear of not getting paid.