The current approach to university IP management implemented throughout the United State involves
- a broad definition of what constitutes an “invention” or “intellectual property”
- a broad scope for who is required to comply with the policy
- a policy demand of outright ownership, often now with a present assignment
- a practice of filing patent applications without industry sponsors
- a working preference for monopoly (exclusive) licenses
- a working refusal to license non-exclusively or royalty-free
- a blanket refusal to waive claims or re-assign claimed but unworked rights
This approach involves two monopolies–one over assets created at a university, and another in the attempt to place those assets so as to generate revenue. This is among the worst of possible university approaches to research enterprise and innovation. Why? Let’s count some of the ways. I’ve put these in a convenient and helpful list of 11 items.
- suppresses innovation
- disrupts collaboration
- undermines public support
- creates institutional conflicts of interest
- violates academic freedom
- turns policy statements to mush
- casts into doubt policy obligations
- abuses authority
- steals from faculty and students
- breaches federal funding requirements
- wastes, costs, and fails
I will work through these, but the basic idea is that the approach is awful. Yet it persists everywhere and even has expanded in recent years. Attempts to introduce different approaches have been chased out. One might think universities have hit the bottom with their current approach, but one should never rule out the depths to which a bozonet might plunge.
I will tell a story to illustrate what is going on. At a big, private research university, an expert early-stage fund manager created an external fund to manage startups based on university ideas. The fund was remarkably successful over its first three years of existence, generating a tremendous return on those investments. What did the university do? It demanded that the fund be brought in-house and run by administrators (so they could get even more of the income), let go the fund’s founding officer, and then promptly ran the whole thing into the ground. The general principle is this–whenever something in the IP area appears to be working, university administrators will find a way to try to take it over or set up something in competition with it and try to kill it. This happened with the external invention management infrastructure, with most of the university affiliated research foundations, with inventor’s professional decisions regarding patenting and licensing strategies, and with inventor’s personal ownership of their inventions. The last thirty-five years of university technology transfer has been a history of bureaucratic take-over of private initiative, cast as “success” and attributed to the Bayh-Dole Act.
The Bayh-Dole Act was never so powerful, though it was horribly drafted. Repealing Bayh-Dole won’t undo the damage that has been caused in its name, however. One will have to rebuild, not as a requirement of law, but as a sustained commitment to a different–let’s not call it new, because it is really rather old–way for institutions to approach invention, innovation, and economic vitality.
A university may implement a restorative patent policy without any changes to Bayh-Dole, of course. It starts with recognizing that a competent policy binds the university, not its faculty, students, or even employees. That is, the policy states limits for university authority rather than making those limits in essence limitless. The policy delegates institutional authority for decisions regarding patents. Where the question is one of academic freedom, the delegation should be to the faculty governing body–that is, the decision should be based on the faculty’s own determination. The policy authorizes the use of university resources beyond those already generally committed, in support of inventions and patents. The policy establishes a procedure by which it might be recognized for that support when a patent leads to a successful new product or improvement. Any such recognition should come by way of recommendation from the faculty, by way of offer from the entrepreneurs and companies involved, not as a demand from the university.
The now standard university dual monopoly patent policy:
1) Suppresses innovation
Most innovation arises from a base of available information and practices. See Steven Johnson, Where Good Ideas Come From. See Vanaver Bush, Modern Arms and Free Men or Science the Endless Frontier. University dual monopoly policies great IP gridlock. It may sound reasonable to identify potentially valuable inventions, file patent applications, and seek industry partnerships, but when each university does just this same thing, with one piece of a 20 or 200 piece puzzle, then no partnerships are possible because no one company can possibly acquire a license to all 200 or even 20 or even 10 bits of university-held IP, with each university seeking an exclusive license and a fat royalty. The effect of the dual monopoly approach is to lock out industry from fast-emerging areas of technology for the life of the core patents–20 years–and to use patent troll methods to attack companies that attempt to develop product anyway.
Archie Palmer put it succinctly in two paragraphs in 1948:
Some scientific discoveries made on a university campus are of such a character that they should be made public and be available to anyone wishing to make use of them, the university merely retaining, and issuing licenses under, the patent title in order to prevent some person or organization from taking out a patent by slightly modifying the material and thus monopolizing the discovery or invention. A fertilizer or medicine that any manufacturer could make is an illustration in point.
There are cases, however, such as the carbonization of coal or the manufacture of vaccum tubes for radio transmission in which the article can best be manufactured only by one or two establishments, because of the large amount of capital necessary or because the use of the new discoveries depends upon the utilization of materials or processes patented and owned by others. In such a case, it is manifest that the public interests may be best served by giving a license to the manufacturer of the patent or discovery, even an exclusive license if necessary, either royalty-free or on a royalty or cash basis.
The dual monopoly approach ignores the refinements of these two paragraphs. There is no public interest practice at most American universities–UC Berkeley’s program in socially responsible licensing is one shining counter-example, and even then, it operates against the formal dual-monopoly written policy of the UC system and with criticism by voices at UC and seeking to do business with UC that demand not only the dual monopoly approach but also claim that socially responsible licensing cuts into possible money to be made–untrue, or at least unfounded, but apparently catches the ear of senior university administrators.
Further, there are few instances of universities offering royalty-free licenses other than to research sponsors when forced to as part of a desired research contract. There just aren’t programs of non-exclusive, royalty-free licensing–and certainly not of exclusive, royalty-free licensing. At one point a few years ago, I tried to coax the University of Washington to grant non-exclusive, royalty-free licenses to the many companies that had cited a set of unlicensed UW patents in their own patent applications. It was a who’s-who list of industry leaders–the patents were core for a class of solid models. But UW would have none of it. They would rather sit on their patents until they expired–or sue everyone they could–than to create a non-threatening, beneficial relationship with companies that knew the work of the lab involved. The dual monopoly approach is antagonistic to innovation.
2) Disrupts collaboration
A dual monopoly approach operates on the principle of “no innovation unless touched by a bureaucrat.” By tying up a broad range of research and creative assets created by faculty, students, and staff with university claims of ownership and a presumption that the university IP licensing operation will have control over those assets until it finds a monopoly partner or renders the asset worthless, the dual monopoly approach makes most forms of collaboration unattractive–informal discussion, research collaboration, exchange of materials and information, publication of technical details and data, competition for grants, for attention, even hiring of students. The approach funnels collaboration into the narrow windows of sponsored research contracts and IP licenses, and throws up barriers to instruction, public service, and personal consulting.
In one situation that I was involved with, at UC San Diego the technology transfer office threatened a researcher who had released software open source. This, they argued, was a blatant conflict of interest, because once the source code was generally available, then this researcher could advance his career by helping others to use the code, all without paying the university anything. The Free Software Foundation was on the verge of suing UCSD over it all. I talked the UCSD folks out of persisting in their threat, but on reflection, the lawsuit would have been loads of fun to watch and might have made a difference. As it is, the monopoly claim on ownership of a broad swath of assets translates into the claim that it is a conflict of interest for any individual at the university to act so as to prevent the university from making money on (or attempting to make money on, or holding some asset that it might at some future point attempt to make money on).
There is still collaboration, but it’s a fool’s activity. Companies with any smarts advise their research scientists not to read the academic literature. It is a mine field of unannounced IP claims, nasty template research contracts and licenses, and liability. Little good comes from engaging a university administration over research or IP, and but there is the prospect of plenty of expense, delay, frustration, and bitterness. If a company thinks they are dealing with a university faculty member directly, they find out that the university has a policy that claims precedence over anything the faculty member might help the company with. If a company thinks they can sponsor research and enjoy the benefits of that work, they find out that the university will not offer background rights and will not consider improvements made in any other lab, creating unworkable uncertainties. Desperate or foolish or perpetually soft-hearted companies choose to work in such an environment. Yes, some big-time companies and clever startups love the idea of a monopoly license, and it is to these that the dual monopoly approach caters.
But even some big pharma companies wish they could break the cycle of developing a drug each on their own, for the $1b or so that it might cost. As one senior pharma exec told me (I paraphrase), “We really do need to find other ways of spreading the cost and the risk of drug development.” It’s just that the spreadsheets won’t let the big companies try anything else. Nothing seems to have changed since Middlemarch. Addictions in companies can last centuries. Rather than helping industries like pharma find new models of development that speed drugs through testing to market, universities have set up the dual monopoly model to prevent those new models from forming, on the assumption that there is better money (and public benefit, presumably) in holding 99.5% of all research assets hostage in favor of one huge monopoly deal once every twenty years. Collaboration, in such a context, is merely a “backdoor” out of which may escape the one golden goose among all the ducks, swans, and cormorants.
A university adopting dual monopoly IP management has taken a stand against collaboration.
3) Undermines public support
The dual monopoly approach is set up on the (largely untrue) premise that it is the best way for the university to make money from licensing. Reports of some big windfall patent license at any university gives the impression that universities are doing really grand on their patent licensing and don’t need public support. People reasonably turn to other areas of society in which tax dollars might have a beneficial effect. Such a concern was identified long ago by Alan Gregg, writing in Science in 1933:
If, in addition, the policy of taking out patents for revenue be interpreted as a declaration of independence the public may quite cheerfully acquiesce and leave research work to earn its own way. Why should gifts intended for the general welfare play the role of capitalizing a business? And what becomes of the peculiar function of university research as contrasted with that of the shrewdly administrated business enterprise?
And university licensing offices, for the most part, issue annual reports and put up promotional materials on the web that make them appear nothing other than “shrewdly administered business enterprises.” They have tacky “branded” names, they announce processes leading to financial returns, they promote “success” stories–meaning their licensing success, not a public benefit.
The dual monopoly approach to IP declares universities do not need public support. At the University of Washington, the licensing office, in a fit of reorganization, wrote a formal business plan that announced that it would generate such revenue that it would change the university’s financial model. Amazingly, senior university administrators–including Mark Emert and Phyllis Wise, and Michael Young–apparently believed the claptrap, or at least embraced it. Of course, the rhetoric around dual monopoly programs is not designed to appear as claptrap. It is designed to look like the greatest opportunity in the world, there for the taking.
UW then went on to spend–or waste, as it were–over $100m in six years chasing the idea of money from licensing, getting next to nothing and spending down reserves, including reserves that should have supported academics, not licensing. In this frenzy, UW sued a major collaborator (General Electric) over patents, claimed a string of sham startups that it claimed would revitalize the state’s economy, and lied (or bullshitted–it was never clear that UW cared about the truth) about the success of a state-funded program to attract “entrepreneurial” faculty to start new companies, if not industries, in the state. (When the UW was called out for their misrepresentations to the legislature, a senior UW official gave the maximum campaign donation allowed to the key legislator on the oversight committee for the program and, unremarkably, the reports were never revised to reflect accurate figures.) The multi-million dollar program was later shut down without fanfare, having accomplished nothing of note.
Universities spin their tech transfer programs to the public as successful, as excellence, as fulfilling the promise of all those hundreds of millions of dollars spent on research (until they fire their directors and bitch and moan about how unproductive their tech transfer programs are). The public, if it believes the spin, also does not see the point of funding more research when universities are out playing hardball with industry, grubbing for money from licensing.
Here’s a difficult concept for folks in university licensing to grasp, but it is essential: universities were paid for their support of patentable inventions because those making the money chose to benefit the universities, not because the universities were clever-wicked-shrewed in setting up IP as threats to inventors, entrepreneurs, and companies. Faculty inventors wanted universities to benefit; companies wanted universities to benefit; entrepreneurs wanted universities to benefit. The dual monopoly model undermines this goodwill and asserts that the innovators pay because they are forced to, or face lawsuits. As any number of tech licensing folks have told me, almost flippantly, if one does not defend one’s patents, they are worthless. For “defend” read “threat” and you can see that university IP licensing on the dual monopoly model is an imperial death star pointed at every rebellious planet of innovation. The public knows which side to be on, and it’s not the university’s.
4) Creates institutional conflicts of interest
A university cannot look out after the public interest or even the best interests of its research enterprise and at the same time grub for money by threatening industry, investors, and entrepreneurs with infringement, misappropriation, conversion, and tortious interference.
A university cannot serve both the public good and mammon. A university that claims outright ownership of faculty and student work for the purpose of “commercialization”cannot then serve as a trustee for that work.
A university cannot mediate claims made about that work (because it has a vested interest in hyping the work for possible commercial value).
A university cannot guard the interests of its faculty in their work (because the university is now attempting to make the deals itself, for its own benefit).
A university cannot manage research relationships to ensure everyone is treated equitably (because the university has asserted a financial interest and control in the work produced in those same relationships).
A university cannot negotiate research agreements on behalf of its faculty and sponsors of research and at the same time be looking out for its own financial self-interest. On this front, federal funding agreements make it express that a university must act only as a trustee on behalf of the beneficiaries of any federally supported project for any intangible assets (including patents and patent applications) that the university acquires or improves in the project.
A university cannot have an invention management practice that has a primary focus on making money from licensing for the university’s own use and comply with this federal requirement.
A university, in demanding ownership of faculty-created IP, abandons its public role as steward and tries to become king. The role of university administration is support for, not domination of, faculty work. A university does not become more effective because administrators dictate the work of its faculty and students–or pretend that their intervention is necessary for the public to benefit.
The smarts of a university, by luck and design, are at its periphery. It is a catallaxy, not an economy. Each intrusion of top-down management into the catallaxy disrupts work, creates problems, which then invites still more invasive management. The illusions of efficiency, optimization, dominance of a management elite over the affairs of workers, and a master-servant relationship in the service of a corporation’s financial investments lead immediately to the idea that it is only “right” and “natural” that a university should claim anything and everything it can from the work of faculty and students, all the better to sell it off for whatever cash can be got. Any attempt to resist this urge is an attack on the public, a personal conflict of interest, and of course, unethical. There is no room for a discussion of alternatives that does not demonize the faculty or student inventor. The university designed to protect the interests of its faculty and students, and to promote discussion of alternatives, ends up destroying those interests and opportunities for discussion in favor of its own financial self-interest.
Actually, the effort is not for the university’s financial self-interest, though that reason no doubt plays a role. The more immediate self-interest is the protection of administrators from fault-finding, should the program of licensing turn out not to result in such lucrative income. The university, by expressly adopting what amounts to a set of religious claims–about patenting, innovation, and the place of university management–cannot then provide support for any exploration of alternatives. To do so would undermine the authority of the administrators that have put the dual monopoly approach in place.
It is not that a university administration has chosen a defective version of the one true approach to invention management–and it is not even the case that there is no true approach to invention management–but rather that it is not the university administration’s proper business to decide. If the university administration desires a licensing operation that favors exclusive licensing, then the administration can dedicate resources to that end. But leave it as voluntary for anyone to use that resource. A university press is set up that way. A faculty author may publish through the university’s own press, or may select another publisher. The university administration does not compel all scholarly works to be submitted first to the university press. The same practice should apply to inventions.
A university that compels inventions to come in creates unmanageable institutional conflicts of interest.
5) Violates academic freedom
Academic freedom comes in various flavors and has had its share of changes over time. But one constant has been that faculty are free to teach and study and publish as they wish, without answering to the administration, or in the case of public universities, to the state. As Paul Feyerabend put it in Against Method, the separation of science and state is more important than the separation of church and state.
Compulsory disclosure and assignment of ownership of faculty work violates academic freedom. For inventions, a patent is a publication. When the university (or in the case of public universities, the state) demands ownership of inventions for the purpose of acquiring patents, it institutes a policy of compulsory publication. A patent is a document published in the name of the inventors. They set for their invention and make claims for exclusive control over the practice of their invention. In the case of compulsory assignment to the university, however, faculty are forced to publish and are forced to make claims to exclude others, even if the inventors do not wish to exclude anyone, and even if the inventors do not wish to publish in the forum and format of the patent literature. They must, and their names must be associated with the publication.
In one case I know of, a faculty member was included on a patent application and later realized that he was not an inventor and sought to get his name removed from the patent–and the university fought him tooth and nail to prevent it. Of course, it was the university administrators’ judgment regarding what constituted invention that was in play, and the administrators used as much legal budget as was necessary to ensure that the faculty member had to remain as a named inventor and therefore as an author of a publication that he did not believe he had any right to. Forced publication is not academic freedom. That the publication follows a process in a patent office rather than a publisher’s office makes no difference. That the university hopes to make money from the patent upon issuance also makes no difference. There is nothing in the principles of academic freedom that state it can be suspended whenever the university sees a way to make money by doing so.
Once a patent has issued, no one can practice the claimed invention without the permission of the patent owner. There is no fair use. There is no research exception. There is no room for independent development. There is no freedom to create functional equivalents. One must design around, get permission, infringe, or forgo. If permission is available only to someone willing to accept an exclusive license, pay generously, and commit to making a commercial product, then even faculty and student inventors cannot practice what they have invented if they leave the university. And at the university, they cannot teach anyone how to practice their invention (now the university’s invention) without leading those so instructed to infringement. Teaching is pointless, other than to show people how to design around, or to advertise the opportunity to bid for an exclusive license to “commercialize.” Teaching without an expectation of practice runs counter to the purpose of academics. When a university claims ownership of all faculty inventions, and does not give faculty inventors control over their work, it violates their academic freedom–they cannot teach and cannot practice their invention without the permission of the university (or the state).
When a university willfully violates academic freedom in pursuit of money from patenting, it gives up a fundamental reason for existing and receiving public support.
6) Turns policy statements to mush
University patent administrators feel so exposed to competition for how to manage intellectual property at universities that they have embarked on an effort to put their management beliefs into policy. To do this, they have chosen not to argue for the imposition of a new regime in which universities move from a condition of academic freedom and multiple pathways of innovation to one of monopoly control and monopoly licensing. Instead, they have simply rewritten history to argue that universities have always had monopoly control and the policies have just been poorly written to reflect this control. What is needed, so the argument goes, are revisions and restatements to “clarify” policy, to “modernize” policy, to “bring it into conformity with peer institutions,”
not to change it.
Thus, when Stanford v Roche made it clear universities had no legal expectation to own employees’ inventions as a result of employment, university administrators, such as at the University of California and University of Washington–among many others–added present assignment language to their patent policies and other documents. They expressly argued that they were not changing policy, just bringing it into line with what they always had intended. They were, of course, changing policy (or disregarding it). At the University of California, policy provided for review after disclosure, before a determination of ownership was to be made. A present assignment transfered ownership of an invention without any review, rendering inoperable an existing portion of policy. True, the text of the University of California patent policy was unchanged, but the policy itself suffered a substantial change.
At the University of Washington it was even worse, because the patent policy is so badly drafted that it never gets around to stating exactly what inventions the university has an interest in. But it is very forceful in asserting that whatever those inventions are, it must and will own them, and everyone is required to assign upfront and outright those future inventions to the university as a condition of any consulting, even consulting for no charge if university administrators decide that one could have charged for the consulting if one had chosen to.
For the first half of the twentieth century, universities for the most part did not have a policy on patents or if they did, they did not claim ownership of patentable inventions, and in the few instances in which they did, it was for public protection, not for making money. The Research Corporation, formed in 1912, taught university administrators that there was the potential for money from inventions. The Wisconsin Alumni Research Foundation a little over a decade later taught them that they could be provincial, encouraging the inventors at their university to fund more research only at the university that hosted their work, not for the benefit of anyone else. Many university research foundations were formed in the next two decades to take on licensing work. Some existed to keep patent complexities and conflicts out of the university itself and merely passed work to Research Corporation. Others worked in competition with research corporation, running their own licensing efforts, and pitching faculty to use their services, not Research Corporation’s.
The in-house licensing programs at MIT, the University of California, and Stanford showed administrators that they did not need the apparatus of external invention management agents, and so they began to end their contracts with Research Corporation and Battelle and University Patents and Competitive Technologies and started their own licensing offices. The misrepresentation that Bayh-Dole vested inventions made with federal support directly with the universities that hosted the research played nicely into this transformation and sped the conversion from “technology transfer” offices (set up with encouragement from Research Corporation to help move inventions from a faculty member’s lab to Research Corporation for evaluation) to “technology licensing offices” (set up to take institutional ownership of inventions and seek licensees directly). The administrators dismantled the external agent programs and sought to channel all work through their in-house licensing.
It was an easy step, then, to re-interpret policy to require university ownership of all inventions, to expand the definition of invention to include “non-inventions” and to expand the scope of claim from employees carrying out assigned duties to anything that an employee or a non-employee might do that the university might claim an interest in. Essentially, patent policy became a statement of non-competition with the university’s licensing operation. University compulsory patent policies are essentially non-compete covenants.
The problem with all this revisionist history and revision of policy is that the policies retain the core of their original purposes. If read for those purposes, they make sense. To do that, however, one has to go back to the original statements, and then watch as the layers accrue. We showed how this happened at the University of Michigan in a series of articles. A similar thing happens at other universities. A narrow claim directed at patentable inventions made in assigned programs of research or under an agreement to use substantial resources of the university–perfectly reasonable because voluntary–becomes transmogrified into a broad, sweeping claim to all forms of intellectual property regardless of whether it is intellectual property at all, for all employees regardless of whether they are actually employees (policies call out visitors, collaborators, volunteers as employees “for the purposes of the policy”).
A policy becomes unreadable when it uses “invention” in an idiosyncratic way to include “non-inventions” and “employee” to include “non-employees” and intellectual property to include “non-intellectual property.” Add to that incompetent drafting–failure to define key terms, failure to use terms in their established ways, inconsistent requirements, poor grammar, dangling and ambiguous modifiers, misstatement of patent law and federal research regulations, failure to update all portions of the law to reflect changes, failure to update policy changes across document series, failure to draft policy consistent with state law or regents-level delegations, and one has a hash of policy.
The upshot is that only university administrators can decide what a policy states. Some policies even make sure this is a formal property of the policy, by creating dispute resolution procedures that end with a final decision by an administrator. This is not the procedure by which, say, a dispute over the meaning of a contract gets resolved. In contract terms, if one party asserts the right to interpret the contract over the dissent of the other party (and the unilateral right to change the contract), then it’s simply not an enforceable provision. An agreement to agree in the future is not enforceable. Most university patent policies are a garbled mess, best read upside down and backwards. Even then, at the end, it is a university administrator who decides which parts are effective and which parts are to be disregarded, and what each part must mean. The interpretation serves the whimsy of the administrator and does not arise from clear language that shows a “meeting of the minds” or even clear guidance for any reasonable reader.
Most university patent polices have been revised from clarity to mush in an effort to turn a reasonable, specific, liberal, largely voluntary process to protect the public interest into a twisted, non-specific, bureaucratic, and largely compulsory process to prevent any competition with the university’s efforts to make money from exclusive licensing.
7) Casts into doubt policy obligations
A university policy binds the university by distributing authority, establishing the limits and procedures for that authority, and authorizing university resources to be used to achieve specified goals. A university policy binds the officers of the university to act as provided by policy. A university policy thus becomes a statement of what others can expect and rely on in their dealings with the university–faculty, students, staff, vendors, visitors, sponsors, the general public. In relying on formal statements of policy, people may recognize a form of quasi-contract.
The university, by committing itself to written procedures, forms a contract with those that rely on the university’s statement of commitment. It doesn’t work the other way: the university cannot assert a contractual relationship with others simply by stating such a contract exists or by asserting a contract will exist if some specified condition is met. If I write that I will give you a dollar if you walk across my lawn, then you have a contract when you rely on that offer and walk across my lawn. If I write that you owe me a dollar if you don’t walk on my lawn, there’s no contract, as you don’t rely on my offer to ignore me.
A university, by taking certain actions, may incorporate patent policy into a contract, even if not expressly incorporated by reference. This much was thoroughly discussed by the court in Shaw v. The Regents of the University of California. The first issue, however, is determining that a valid contract exists–that both parties intended to enter into a contract, that the contract was not made under duress, that the contract was not in violation of public policy or law, that the contract was not induced by misrepresentations or fraud, that the contract is sufficiently clear that a reasonable person can grasp its undertakings from the written words of the contract. A second issue concerns the state of the contract. The Shaw court made clear that while a university may change its patent policy whenever it wants to, if it has entered into a contract with an employee that incorporates that policy, it is not at liberty to unilaterally also change that contract. The employee has a contractual interest in the form the patent policy takes at the time the contract is formed. An employee hired in 1980 with a patent policy incorporated into a contract has the benefit of that policy regardless of later changes to the policy introduced by the university.
But if a university’s patent policy is a garbled mess, if it contradicts other portions of policy, such as policy statements on academic freedom, if it persists in contempt of the US Supreme Court in claiming the Bayh-Dole Act gives universities ownership of federally supported inventions, if it claims it can be changed unilaterally, and that employees must agree to any such future changes, or asserts that employees have assigned their rights to the university upfront, when federal patent and copyright laws require a written, signed instrument, then it is not clear at all that there is an enforceable contract.
It’s not that a university cannot force its policy on faculty–it can use threats, legal manipulation, disciplinary actions, retaliation–but it does not have an enforceable contract. Again: a university can force compliance with a demand for assignment merely out of a threat to sue–a threat to force a non-compliant faculty member to spend at least $150,000 defending himself or herself from the claim. But that does not mean that the faculty member had any obligation whatsoever to assign inventions to the university, or that some purported present assignment language in a policy statement was anything other than an item in a poorly conceived administrative wish list.
A good statement in a university patent policy would be an express prohibition by the university to sue any faculty member or student in an effort to obtain any intellectual property right. Either the right is voluntarily contracted or not at all.
The result of the garbled mess of most university patent policies–arising as administrators try to establish authority for their commitment to a dual monopoly approach–is that university inventors arguably don’t have any obligations as a result of policy statements. University administrators may well have such obligations, but the obligations are not transmitted by policy to faculty, students, or even employees. The university may not even have a legal basis to retaliate against faculty and others who refuse to accept a university’s demand or a claim that a policy statement has the effect of law. Even if a university does find a way to retaliate, it does not obtain ownership of an invention in dispute–it has the ability to terminate the inventor’s employment, or to apply sanctions. The refusal to accept the imposition of an obligation is not a crime, is not a tort. The rejection of the master-servant relationship is at the heart of academic freedom. If a university writes a patent policy that demands such a relationship, then it has created an impossible contradiction in its own policies and cannot possibly create a legal obligation on faculty to assign their inventions.
8) Abuses authority
The dual monopoly approach to university IP management sets up the conditions for university administrators to abuse their power. Even though most university patent policies are a garbled mess (Stanford’s is one of the few that’s not–not that it’s nearly as smart as it once was), university administrators recite policy as the authority for action. To violate policy claims is to incur the wrath of administrators, who may allege ethical misconduct, conflict of interest, torts of conversion and misappropriation; they may seek to damage the reputation of the non-compliant faculty member; they may deny (and have denied) the non-compliant faculty member access to awarded research funds; they can threaten legal action, for which they often have an unlimited, unregulated budget. In the absence of an actual legal right to take the personal property they claim (and patents and copyrights are personal property rights under federal law), a university administrator abuses his or her authority by asserting claims that have no foundation other than administrative desire.
9) Steals from faculty and students
The dual monopoly approach steals property, opportunities, and initiative from faculty and students. A university exists to provide opportunities for scholarship–creation of material for instruction, inquiry into the nature of things, professional development, public service. There has never been a need for the university administration to control these things for a university to operate. Indeed the idea of academic freedom carries with it an expectation that the university administration will not own and attempt to control the ideas of its faculty and students–and the addition of copyright or patent to those ideas does not change the situation in the slightest. How one exploits a copyright or, especially, how one exploits a patent might attract the interest of university administration, not for the money but for the effect on the university, on university research generally, and on the public, but even that degree of interest does not require a university to own the intellectual property in question.
The reality is that university patent policies that demand compulsory assignment of all intellectual property as a condition of employment, or of the use of resources, or made under the “auspices” of the university is a form of theft. In public universities, it is a clear exercise of eminent domain. The state is taking personal property for public benefit, for which it owes just compensation. Such compensation is not “just” in the form of a royalty-sharing schedule set up in the days of voluntary assignment, and which may never come into action. If the state takes your house under eminent domain, it owes you the fair market value of your house at the time of taking. It cannot say that it will pay you one-third of the fair market value of your house, but only if it ever gets around to selling your house, and it has no obligation to do that, and if it fails to maintain your house for twenty years until it is so dilapidated that it falls apart on its own and thus isn’t worth anything, then it doesn’t owe you anything and you are lucky you aren’t charged a fee for the removal of the remnants of your house. But that’s almost exactly how compulsory university patent policies read. It’s theft.
For the public universities, every taking of ownership of an invention under a compulsory patent policy constitutes an eminent domain action. The university owes the inventor the fair market value of the property at the time of the taking. For patents, that fair market value is the net present value of the net rents receivable from the exploitation of any patents that may issue on the invention. Not one third of those rents. Not an offer to share royalties without an obligation to secure royalties. The exposure at any public university running under a compulsory assignment policy can easily be in the hundreds of millions of dollars across a portfolio of hundreds to thousands of inventions.
Yes, administrators might argue that there was a patent contract under which the inventor agreed to assign–but if it is not true that there was such a contract, because the administrators failed to obtain an actual written contract, or obtained it by duress or fraud (such as claiming federal law required the assignment–a typical move), or because the policy statements themselves create no such enforceable obligation, then the university has no patent agreement with its inventors and is merely asserting a claim backed by threat of retaliation.
Similarly, if a university does have patent agreement, but it is contractually based on the patent policy in place when the inventor was appointed to the university, then it is that policy that controls, not a current statement of policy. Many patent policies before 1981–and generally before 1990–do not require assignment of most inventions, do not define invention as most anything an administrator decides is an invention, and do not even necessarily require assignment to the university (and instead may require the inventor to make arrangements with an external invention management agent, and even then, only after the university via a committee has determined that patenting is even in the public interest. If a university persists in demanding assignment, it’s not much different than a mobster demanding your property. The mechanism of compulsion is not a prior agreement, but the threat of harm.
10) Breaches federal funding requirements
A persistent claim of revisions to university patent policies that support dual monopoly practice is that the revisions are necessary to comply with federal law, and in particular Bayh-Dole. It’s not true. Bayh-Dole does not require that a university have any patent policy whatsoever, and most certainly does not require a university to take ownership of any invention, nor to “commercialize” any invention, nor to “offer to the invention to the government” if it chooses not to patent or commercialize the invention. This is all, to use the technical term for it, bullshit. That is, the administrators who recite these claims have no regard for the truth. They say these things because it serves their interest to say them. They have heard these things at AUTM meetings, and read these things in articles written by uncritical advocates for dual monopoly licensing. They don’t bother to read the law or the implementing regulations or the federal funding agreement terms and conditions. And if they do, they are not competent readers. They don’t want to discover they are wrong. They don’t want dual monopoly licensing to go away. They believe dual monopoly licensing gives them job security, even if it is a total disaster in all other ways.
The irony is that dual monopoly licensing policies breach the very federal regulations that university administrators make such a fetish over complying with. In the standard patent rights clause authorized by Bayh-Dole, federal agencies require that universities flow down certain duties to potential inventor employees. That flow down is essential. It delegates to the potential inventors rights in their inventions, subject to their decision to assign rights in the invention. It makes the potential inventors parties to the funding agreement should they make a patentable invention within the scope of the planned and committed activities of the federally funded portion of a project. If the inventors do not assign their patent rights, agencies are authorized to treat the inventors as small business contractors, but with fewer requirements than those made of other small business contractors, and substantially fewer requirements than those made of nonprofit organizations.
But universities adopting dual monopoly IP practice uniformly refuse to implement this requirement (see 37 CFR 401.14(f)(2) and 37 CFR 401.9). Instead, they substitute the obligation to assign inventions to the university, though that does not meet the requirements of the patent rights clause–it is generally not a written agreement made by the potential inventors, it is not an agreement to protect the government’s interest, it does not confirm that the inventor has both the right and obligation to convey rights to the government if so requested, it does not limit the inventor’s reporting obligations to the university to those necessary for the university to comply with the reporting requirements to the federal government.
Furthermore, most universities receive federal funding under a standard federal funding agreement set out at 2 CFR 215. This federal funding agreement addresses copyrights, data, and invention rights at section 36. In section 37, the federal funding agreement deals with the management of intangible property acquired or developed in the federally funded project. For such property, the university is to act as a trustee. That’s an important requirement. A dual monopoly IP management policy walks into the teeth of this requirement–and ignores it. First, by making assignment of patent rights and copyrights (such as in software, digital media, and some forms of data) a condition of employment and federal funding, a university expressly makes the acquisition of these rights a use of federal funds. The federal funds pay the workers, and the university asserts that by this pay the university also acquires ownership of the intangible assets that result. The university meets the requirements of section 37 and must act as trustee of the rights acquired on behalf of the beneficiaries of the project.
A trustee cannot act for its own benefit in managing an asset for the benefit of others. Yet that is exactly how university dual monopoly policies are set up. They provide protections for the university in its management of IP and do not consider the beneficiaries of the projects that federal funding supports. I have seen scores of university invention disclosure forms. I used to collect them. Not a one every asked inventors for the beneficiaries of the projects in which inventions had been made. University administrators simply ignore section 37 compliance and substitute the idea that they must be working in the public interest by trying to make money by seeking monopoly licensing arrangements. A trustee that aims to make money from an entrusted asset for its own purposes is not considering the interests of the beneficiaries of any specific project.
If a university did not have a compulsory dual monopoly policy–most universities did not until about twenty-five years ago–then it would not come within section 37 of federal funding agreements because it would not use federal funds to acquire assets–those involved would choose to assign voluntarily, and the consideration for that assignment would not be the pay they had received (necessarily federal funds), but rather some other consideration, mutually agreed upon.
11. Wastes, costs, and fails
The sad reality of the dual monopoly licensing model is that it is wasteful, expensive, and ineffective. By claiming most everything, universities create a huge bottleneck of disclosures of most anything–random junk, stuff that isn’t patentable, stuff that is but for no purpose, and stuff that could be practiced by anyone without the need for a patent to encourage investment, and even some few things that would benefit from patenting to encourage investment that otherwise wouldn’t happen. Lots of noise, little signal. It’s a lot of work to sort through disclosures, to get them done properly, to record them in a data base, to design a data base that can handle them all (that can run to millions at the big schools–I have had to deal with two), to assign inventions to case managers, to deal with patentability assessments, deal with patent attorneys and agents, get provisionals filed. Most of this is waste activity–yet it is marketed to senior administrators and the public as the growth of an entrepreneurial university culture and an index of innovation. What total crap! The number of disclosures in a compulsory, comprehensive IP regime is an index of bureaucratic compliance is is largely a measure of the extent of the threat felt by inventors by failing to comply.
The cost to a program that aims to manage everything is huge. We are not merely talking about the cost of intake and patenting–easily millions–but also the costs that come with forcing everything through a bureaucratic bottleneck that moves, as one senior attorney I worked with used to say, “at the speed of darkness.” Those costs include opportunity costs, loss of goodwill, loss of trust with inventors and with entrepreneurs. Implicit threats to litigate if one does not license degrade relationships with research sponsors and with corporate donors and collaborators. Visiting researchers are reduced to watching months’-long fusses between university licensing offices over obligations to assign inventions to whichever university. The dual monopoly approach to IP management is the most expensive possible way to handle rights–too much stuff, not enough judgment, defocused, overloaded, always on the defensive.
Finally, these programs are ineffective. Yes, there is the occasional big hit license–once every ten (Stanford), twenty (UC), or forty or more (UW) years–but it does not come about as a consequence of the dual monopoly approach. Often that big hit license comes about despite the dual monopoly approach. Some of the most lucrative early biotech deals done by university were non-exclusive, not exclusive, and the underlying rights were assign voluntarily, not by compulsion.
For the most part, for nearly all disclosed inventions (taken broadly, per university policy definitions), these programs are death itself. Regardless whether patent applications are filed, nothing happens. The mere fact that control has been taken from the inventors and patent applications filed is enough to stifle even research use in companies–why bother with something one will not get a license to because all one wants to do is use the invention, not make a commercial product out of it. And many research-related assets are tools, not future products to be sold to the public. Even when universities license rights exclusively, most times no product results. Companies license rights to fluff themselves for investors, to reduce possible competition from others, or to get access to inventors to design other products or design workarounds. Mostly they don’t license exclusively to build a product and pay royalties. They find a way to avoid royalties along the way–design around, wait for the patent application to fail or the patent to expire, or shift work to jurisdictions outside the reach of the patent.
Meanwhile, exclusive licensing precludes standards formation, broad company use (without product phase), opportunities for multiple applications, competition and cooperation in development, development of interoperable technologies, and the easy movement of technology between organizations. Exclusive licensing, just like rounds of investment, are not an index of innovation or even the potential for innovation. They are just transactions that change the status of the participants. The public benefit, if any, is constrained to the exchange of speculative money for an IP right. Actual product, or public use, is another thing entirely.
Bayh-Dole authorizes federal agencies to ask for specific information on the status of inventions supported by federal funding and claimed by universities: the date of first commercial use or sale, and royalties received, at least. But federal agencies generally don’t regularly ask for this information, and when they do, it is protected from public disclosure by FOIA. Universities do not report this information publicly themselves. AUTM’s licensing survey conveniently avoids collecting this information. It’s odd. If the dual monopoly approach was so darned successful, you would think that universities would report the underlying data to support their claims. That failure to toot the horn isn’t itself a proof that the programs are a big fail, but given the attention IP licensing programs give to their own self-success, it’s an interesting detail of omission.
The overall assessment is that university dual monopoly IP management programs fail. They are at an administrative sweet spot for controlling most everything, suppressing competition, avoiding accountability, having perpetual need for more money and staff, and claiming credit for whatever happens to work out, but as an enhancement to technology transfer or economic development or innovation or augmenting entrepreneurial culture, a bureaucrat’s thumb in every innovation pie is not a recipe for success.
Universities must break the habit of dual monopoly IP management. The path is simple, but it takes courage.
1. Declare–or confirm–that the patent policy is binding on university officials but voluntary for faculty, students, and employees. A university IP office may manage IP that it acquires when it does so voluntarily. If the office is at a public university, set it up so that there are not problems with laws restricting how public employees “do business” with the university-as-public-agency. It has been done with university presses. It can be done with IP shops, too.
2. Allow alternative forms of dissemination. Create an office to help with open source, open hardware, open wetware, commons, and non-exclusive projects. I ran such an office for nearly a decade at the University of Washington. We were consistently 3x to 6x more effective–in revenue generation, no less–than the patent licensing shop with its fixation on monopoly patent licensing. Create arrangements with external invention management agents–find specialists in the areas of research that your faculty do and get them in contact with your inventors.
3. Change your requirements for contract research. For federal awards–perhaps 70% of your total research base–just follow the standard patent rights clause and don’t place any additional requirements on inventors. For industry agreements, shift to non-exclusive access as a default, and let the sponsor choose the fee. Now offer the research results to everyone else at the price–unless it is too high, in which case negotiate a lower fee with your original sponsor. Value is not just in seeking speculative monopoly contracts with folks too generous or dumb or desperate or careless to be real. The real value–social value as well as financial–is in creating a number of projects that each may create beneficial relationships with organizations that seek to learn from, to use, and build on work done at the university. For that, the university does not have to own anything, and what it does own, it owns because inventors (and others) have chosen it as an escrow agent to manage what they produce together. Changing contracting requirements lowers the overhead of funding research and lowers the uncertainties that come with monopoly practices. There is still a role for IP and for monopoly positions–but it is rare, chosen, and taken out of the university so the university does not have to deal with its consequences–not its litigation, not is fuss over background rights and future rights, not its worry over double licensing, not its costs of patenting, not its fuss when things go wrong and investors are losing money, not the complications of board seats, rules on offering stock, insider trading, and attempts to time the market.
If all this is too difficult, keep in mind that having absolutely no policy at all, running with the defaults in federal patent and copyright law, and using voluntary agreements as needed is way, way better than having a garbled mess of a policy that puts everyone on edge, has a host of known, adverse consequences, and shows no signs of magically living up to its claims.