I have spent the past few weeks working through 130 patent policies at US universities, circa 1962. I have compiled a set of notes that runs to 100 pages, and another set of notes that are notes on the notes–2nd derivative stuff, you might say. In a myth of progress, the past is the land of errors and lack of understanding and misdirections, to be replaced in the course of time by enlightenment, leadership, efficiency, and my god things that really work in modern, recognizable ways, as we have today, at the peak of all things progressive. In other myths, the myth of progress is rather silly, if not dull, because it is little more than a big self-pat on the back to whomever happens to be riding the gravy pig at the moment, thinking that whatever happens to look powerful must be the best of all things.
Just because there is a progress myth in play does not mean, of course, there is not also progress. It’s just healthy to separate myth and reality, recognizing that the representation of either is going to end up as a narrative, and have narrative artifacts that go along with it. The great mystery of the way things are is that there is no primitive narrative. There is not one true statement to which all artifacts, facts, accounts of events, and reasoning will lead, once everything is sorted out.
This idea is equivalent to what a senior federal mediator told us once, that it simply was not the case that if one side in a dispute just understood everything proper-like, then they would agree with the other side’s position. That “other side,” he pointed out, had as thorough understanding of the situation as anyone. The problem is not fault of knowledge or reasoning, but fundamental differences. The resolution is not forcing facts and reasoning into a single true narrative that turns out, glory-be, to be what one had thought all along, but rather a barter of behaviors, so that the differences do not stand in the way of transacting business. That barter may be compromise, but it also may be a willingness to let things get on despite not being satisfied on every point, even if nothing is conceded. Live to fight another day. There are multiple narratives at play, and life draws its vitality from the fact. As the Talking Heads song “Heaven” has it,
The band in heaven,
They play my favorite song.
Play it one more time,
Play it all night long.
That’s because, “heaven is a place, a place where nothing, nothing ever happens.” And that’s why, of course, “it’s hard to imagine that nothing at all could be so exciting.” Because the best of everything, when it becomes the only of everything, is nothing at all, and it really is hard to imagine that such a thing could be exciting, or fun, or true.
The history of university patent policies–and the absence of such policies–provides a set of working ideas about inventions in a research environment. Rather than being past failings, to be replaced by the strong and modern stuff of today, they are more along the line of menus of restaurants that closed, not for lack of good cooking, but for other reasons. Times change, and sometimes those reasons don’t have nearly the effect they did at the time they found themselves to be so effective.
One of the concepts that recurs throughout the patent policies circa 1962 is that of “equity” in inventions. It appears, even, that “equity” may be the driving reason to have a patent policy at all, to formalize the expectations on when and why an institution might have equity in work done by personnel under its auspices, furthering its mission, in its employ, using its facilities, and participating in contracts that it has made with others. “Equity” means something along the lines of “what is fair to have, given the circumstances.” One can claim an equity in inventions without any policy statement or written agreement whatsoever–and then it is a matter of folks’ perceptions of what is fair, and if there is no agreement, then the leverage available to bring others around to address one’s claims, legal or rhetorical or otherwise. The effect of policy is to replace this swirl of potential perceptions with a protocol, and in advanced stages of policy, to replace the protocol with something contractual, something legal, or at least legal sounding, something then to be decided not by those involved, or by a review, but by attorneys, before judges if need be.
In this way, the purpose of patent policy moves from making a position clear, to establishing a protocol for deciding what is not clear, or which should be an exception, to making an enforceable claim, to leaving the whole matter to attorneys, larding up the institutional side so that any effort to contest the institutional claims will be difficult, expensive, and doomed. It’s clear that in some versions, a patent policy may be helpful, but in mature forms, unhelpful and rather damaging.
The university patent policies identify how equity in inventions might arise (replacing folk ways with formal ways). Thus, the University of Alabama can identify “substantial contribution in time, money or facilities” while the University of Southern California policy expect that equity arises with “substantial use of University facilities or funds” but excludes “performance of normal duties.” A number of universities work to define the line between substantial use of resources and ordinary use. Southern Illinois puts it at $1500, but Alabama it is $200, and at Northeastern it is a “relatively small amount” while at Mississippi State “beyond a reasonable amount” gives rise to a claim of equity.
Michigan will have none of these refinements and asserts that an equity in any patent “issued in connection with research projects” while the South Dakota School of Mines and Technology can assert “by virtue of an enlightened patent policy” that the institution requires of each employee the assignment of “any invention made while such an employee.” The interesting issue is whether “while” means “while doing previously agreed employee things” or “during the time in which one is an employee”. In the first meaning, “while” means “within the scope of duties otherwise established” and in the second “while” means “the scope changes by policy to include whatever you happen to invent.” These are very different sorts of meaning. Fairfield University claims ownership of patentable inventions by anyone “connected with the University” and which “result from research conducted under the auspices of the University or with the use of University facilities” unless a research contract provides otherwise. Thus, here, “auspices” are distinct from the use of university facilities, and “connected with” is a rather broad gesture.
Other institutions disavow any equity in inventions. The University of Pennsylvania policy provides that “any invention or discovery which may in any manner affect the public health, such as a new drug, process or apparatus intended primarily for medical or surgical use, shall not be patented for profit, either by an individual in the employ of the University or by the University itself.” Loma Linda University, as well, recites as “the governing principle” the “medical and dental ethics which prohibit physicians and dentists from realizing any direct or indirect material return from the manufacture, sale or distribution of any product for which the patient pays, or which is used as a therapeutic device or health aid, or which in any manner affects public health.” Drexel policy finds that students pay tuition, and therefore the university has no equity in their inventions based on use of “space or facilities.” The University of Iowa finds just the opposite, claiming all student inventions if the student “receives credit toward any degree for work resulting in an invention”–and not only that, the student inventor doesn’t get any share of royalties. I guess equity goes only one way. Case Institute of Technology, along with others, makes it clear that “payment of salary for normal academic work…or provision of normal academic environment can not be claimed as grounds for equity by the Institute….”
Stanford’s policy makes it clear that it is not going claim equity in faculty work: “A faculty member is permitted to keep all rights to inventions he may make except in cases where other arrangements are required by a contract or grant for sponsored research.” Interestingly, Stanford’s policy goes on to implement exactly what will later by used by the federal government in its standard patent rights clause (f)(2) agreement:
Each faculty member executes an agreement to grant appropriate license to the sponsor, or if the sponsor requires, to grant the entire right, title and interest to the sponsor.
Princeton’s policy recites a common approach with an institutional emphasis:
The equity of any faculty member, employee or student who makes an invention and the equity of the University shall be established by the University in conference with the inventor.
Disputes over this determination are taken to a board of arbitration, which a number of policies propose as a means of settling disputes. Other policies, such as at Rhode Island, simply asserts institutional ownership:
A member of the staff of the University of Rhode Island, in consideration of the employment by the University and of the salary to be paid by said University, for the purpose of definitely eliminating any possible controversy which may arise as to the ownership of any patent which may be granted, upon and by acceptance of appointment, agrees that if any discovery or invention is conceived, devised or developed in the course of employment by or through the use of the facilities and equipment of the University, the same shall, at the option of the University, be and become the property of the University under the terms and conditions of the patent policy established by the Board of Trustees of the University.
Other than being a fine periodic sentence that captures clearly what is expected and uses a word like “controversy,” this policy is almost modern in its treatment of the matter of equity. The university has ownership by rights, no particular review of circumstances needed to decide proportionalities.
The universities’ policies collectively lay out six responses to the issue of equity, if the university is going to address equity:
1) ownership of the invention
2) reimbursement of expenses
3) participation in financial return
4) license to practice the invention
5) controls or restrictions on the inventor’s exploitation of patents
6) be happy with enhanced reputation from doing good things
There are, in the policies multiple ways to respond to an inventor’s interest in a patent, once the matter of equity is raised. Institutional ownership is only one such response, and in 1962 it was clearly not the preferred one. The response that emerged as dominant was participation in financial return, based on the decisions of an agent to which the invention was assigned, typically directly by the inventor(s). This response was subsequently replaced, after Bayh-Dole, by the claim that universities own federally supported inventions (and all the rationalizing variations of this claim, such as a right of first refusal), which led to the rapid creation of internal university technology licensing operations, which claimed ownership, reimbursement of expenses, participation in financial returns, and a license to practice the invention. That is, they took nearly every response to the question of equity, ignored examination of circumstances (other than sponsored research contracts), and repositioned inventions as institutional assets from outset, rather than as personal assets from the outset. This change has profound implications for national innovation policy. Is innovation institutional or personal? Perhaps it’s not a fair question, but consider this: institutions have no brains.
The policies appear to develop from an initial default position that universities should not own patents to ones in which payment for assigned tasks should give rise to the university also taking ownership of the work created, including inventions. In this view, the university does not pay people to provide a service to others (such as teaching, research, or consultation), but to provide that service to the university as employer. For research, the change is dramatic, moving from a public service model to a corporate model. If there are going to be patents, then a university’s “auspices” or reputation comes into the picture as a reason to take an interest in how the university’s personnel use patent rights. Medical faculties in particular went out of their way to disclaim patents. Harvard even offered legal assistance to anyone seeking to challenge medical patents.
From the auspices and reputation arguments flows also an interest in the use of external agents. A number of universities disclaim ownership of inventions, but under an equity idea require that inventions be assigned to an agent. In some cases that agent is specified in policy–often it is Research Corporation, but it also may be an affiliated research foundation, of which about 50 existed in 1962, or Battelle Development Corporation. The university typically entered into a contract with one or more of these agents, under which if they obtained ownership of an invention made with a university equity interest, they shared any proceeds from patent licensing with the university. The University of Delaware policy makes this clear. Faculty inventors are recommended to use the University of Delaware Research Foundation, Inc. But if they don’t, then:
An alternative procedure may be followed by any inventor who prefers to prosecute the patent action himself instead of utilizing the Foundation. In such case, however, the equity owning to contributions of the University and others must be considered; therefore, it is mandatory for the inventor to provide full information respecting the invention to the chairman of the Committee on Research. Representatives of the Committee on Research will then negotiate with the inventor as to the equities in the invention and arrange for proper distributions of costs and of possible income.
The contract with the research foundation takes care of the equity review. Otherwise, there must be a review for circumstances, led by a committee, followed by a negotiation with the inventors with regard to costs and income.
Throughout these policies, these three themes repeat: equity arising from a stated association with the university (and especially employment, use of facilities, or special funding), the use of agents as preferential to direct university involvement in patents, and a review for equity led by a review committee composed of members of the faculty. It is worth considering how these three concepts inter-relate.
First, it appears that a university might claim an interest in inventions made by its personnel, if for no other reason than that it dislikes the ideas of patenting relative to the purposes of scholarship and wants to prevent misuse of patents arising in scholarship, whether by faculty inventors or by roguish companies patenting the same or related stuff and disrupting academic research. But managing that interest would appear to require the university to take ownership of inventions, file patents, and get wrapped up in just what it dislikes.
The agent approach offers a range of ways of dealing with the problem. First, if the inventor uses an agent, then the burden shifts away from the direct control of the inventor and the association with the university. Think of it as a kind of escrow that creates some separation between university work and legal, financial, and commercial matters. Second, if the university has an agreement with the agent, then it can obtain some benefit for its equity without being actively involved in any of the decisions with regard to whether an invention has “potential” or “value” and if so how that potential or value is to be realized. That’s all up to the agent. By going to an agent with a contract with a university, the inventor decides to share proceeds with the university. A number of universities identified multiple agents. This permitted even greater choice for the inventor in who best to work with. Finally, the agent relationship resolved any questions regarding equity.
But what if an inventor did not use an agent, or did not choose an agent with an arrangement with the university? The primary tool used by many policies is that of review. Whatever the statements regarding use of facilities or within the scope of regular duties or with substantial financial support, the issue came down to a review of circumstances, as anything about fairness arose from those circumstances. If the university had made substantial contributions, it should share, just as co-inventors or sponsors or others providing support for an invention should have an interest. These reviews were done in a number of ways. One approach was the ad hoc committee. When an invention was made, it was reported to the University president, who would appoint a committee to review it for equity. Another was to have a standing committee that reviewed all such cases. Sometimes this committee had legal counsel attached to it to provide advice to the inventors, as a service, should the inventors choose to bring their invention to the committee. A third way, used in a few cases, was administrative review by a Director of Research or similar position.
The general shape of these reviews is that faculty, generally, decided, on behalf the university what to make of an invention made in the course of scholarship. It was faculty who decided the equity of the situation on behalf of the university when an inventor chose not to work with an agent that already had an arrangement with the university. It was, if you wish, a form of peer review, or faculty governance. The point of university policy was to identify equity as an issue, but task faculty to resolve the issue by considering the circumstances. The issue was what was fair, under the circumstances. The university’s benefit, if you will, was awarded by the faculty, based on review. As the University of Southern California Policy laments, this sort of review is difficult:
It is difficult to define the equity of the University in quantitative terms since the circumstances of each invention may differ. In most instances the University’s equity in a patented invention will be satisfied by the prestige accruing to the University as being the source of useful knowledge.
Mississippi State and the University of Mississippi (sharing a common policy) make a similar gesture. Logistically, it also becomes a challenge to review every invention for circumstances, case-by-case, as they arise. This is what made the agent approach so attractive. Instead, the committee would review for basic substance, and the invention looked decent enough, the committee would forward it to Research Corporation or another agent. If the agent did not accept the invention for management, then the university would release its interest, since whatever equity it might have would not be worth much if an agent refused to manage the invention.
The important aspect of all this effort, however, at each point was to find an impartial review of the situation–both equity and resolution of equity. A faculty committee was one way. Referral to an external agent was another. Both kept the university from a self-interested claim on inventions made in scholarship and focused on what is fair, that is, what others looking at the circumstances consider to be fair. Rather than make the case for self-interest, let others decide. Such a thing these days is entirely out of fashion, and perhaps that is an indicator of how far our society has moved in fifty years, or at least university administrators have moved.
There is a further extension of this idea, however, and that relates to how those taking licenses to university-hosted inventions pay. The standard line these days is that companies pay because they have to, because they are threatened with litigation if they do not pay, and the purpose of “increasing the value of an invention” is in part to develop the swagger that makes companies fear that litigation, or adverse licensing behaviors, and therefore will pay more, and choose to license more quickly, to ensure that they do not get punished later by the university’s patent managers. The alternative, however, is that companies pay because it is equitable to pay. That is, they choose to pay. They pay because they want to pay. I know, it sounds alien, but if a university is going to ask for an impartial determination of its equity interest in an invention, then it does not matter, really, whoever happens to control the invention. What matters is whether that whoever–inventor, agent, assignee, licensee–is willing to recognize what is fair, based on the circumstances. In a world of equity, and equitable treatment, companies choose to pay universities (and their agents) for inventions because they *want to* not because they are bullied or cornered or judicially forced to.
It really should not be alien. When we walk into a restaurant, we have already made a decision to pay. The only real question is how much we’ll order, and if we enjoy the food, the service, the ambiance, and especially our own company, then how much we will add to the tip. We do this because we have chosen to, not because we are forced to. What the current technology transfer “system” has destroyed, or has had a hand in destroying, is this idea that companies choose to pay.
I once proposed such a deal to a big company that we were doing a modest research contract with. I was at a university with a lot of tax free bonds outstanding that had been used to build research facilities and bond counsel was deathly in fear of the private use provisions of the Tax Reform Act of 1986 relating to research done in such facilities. The safe harbor provided by the IRS revenue procedure (now 2007-47 but it was the same in 97-14) was that the price of any technology developed in such a facility had to be set as if the company was not the sponsor of the work, and could only be set “at the time the license or other resulting technology was ready for use.” Utter nonsense, but in fine regulatory style. The university interpreted this clause to mean that one could not grant a royalty-free license to research upfront (in the 2007 revision, one can, but only if the university owns any patent).
So I proposed that if there was any invention, at that time the company, if it wanted a license, could determine the price for a non-exclusive license, and that would be the price, as well, that we would charge all other companies that also wanted access, with the one condition that we could charge everyone, including the sponsor, a lower price if we believed that doing so would facilitate access. Well, now, that set them back a notch for a few days. But then they came back and said they couldn’t do that. Why? I asked. Seems that they felt that if they said the price should be $0, that we would accuse them of being inequitable. We wouldn’t have–we really didn’t care–but it proved that the idea of equity was still around, just not something that gets talked about, like stewardship and delight–also off limits in technology and innovation studies and university policies. It also suggested that when a sponsor requires a royalty-free non-exclusive license to any invention up front, it might not be fully addressing the equity of the situation. But then, who does these days?
The combination of equity in inventions, determination of equity by review, and use of agents to manage patenting developed over the course of nearly 70 years until the implementation of Bayh-Dole. Bayh-Dole was one of the great national agent-destroying bits of legislation, not because it set out to do so, but once it was on place, it was used as leverage to convert university technology transfer offices, which had come to replace standing faculty committees, into patent licensing operations. The equity arrangements that universities had worked out with agents were replaced by royalty sharing schedules directly with inventors. Inventors got a boost, generally, over the sharing they got through agent contracts–which in 1962 was pretty typically 15% and now is more around 30%. On the other hand, because universities now routinely claim all inventions, not just ones made with expressly dedicated resources, equity is no longer an operative concept. What the university owes is detached by policy contrivances from the act of taking ownership and represented as a policy largess, a perk of employment, a benefit of university ownership. It is not compensation for assignment of title, because current policies claim that the consideration for that assignment is in continued employment, use of facilities (without any particular limitations, usually), and approval to participate in research at all.
The direction that universities have taken with patents in policy has been to raise equity as an issue, introduce various methods of addressing equity, finding that those methods are time-consuming and sometimes contentious, adopting contracts with agents as a proxy for review, and then taking the patent activity in-house, substituting one’s own royalty sharing schedule for the agent relationship, and making institutional ownership mandatory. They then move away from the agent model and substitute the financial portfolio model. That is, rather than representing the inventor’s interest in each invention taken under management, they instead represent the university’s financial interest, which is served by a somewhere between 1% and 6% of the inventions under management. The rest can rot.
In this way, university administrators in 100 years have come a full 180 degrees from where they started. Now it is a public good that the institution own everything. The only bit of equity left to the inventor is that Bayh-Dole requires the university contractor to share royalties with them, but only if they ever get around to royalties, not that they ever give anything away if they can help it. The universities’ own royalty sharing policies no longer reflect circumstances: the policy is “fair” because it applies to everyone regardless of circumstances. De minimis (just $1 kinds of arguments) use is treated the same as $1m in extra additional funding. Again, we are now 180 degrees out of phase with the past. It is easy to see how the current state of things is institutionally convenient, especially when considered as an alternative to conducting case-by-case reviews and trying to figure out for each just what ought to be done. However, if equity is awarded or granted, rather than claimed, then the case-by-case review also is not necessary.
At one point, the prestige of supporting inventions that made it into public use was enough for a university to be satisfied. Now we have policies like the University of Akron‘s, which stipulates that “The university will prosecute applications only for those patents that appear to be of potential economic benefit to the university.” It’s hard to chide a place for being so blunt about it, given that appears to be the standard, whether stated directly or implied, in most every university patent policy these days.
It may be that equity still operates, and is an important concept, along with stewardship, delight, and controversy, in how innovation from university research comes about, or more broadly, how university research expertise contributes to innovation. If so, then university patent policies are for the most part missing a lot of stuff that matters for innovation, while taking care to pack down well everything that matters for institutional convenience and control. One wonders which is more important to the research community, or to industry, or to the general public. After Stanford v Roche, this is a great time to revisit the 180 turn and ask if we have lost our innovation compass.