Archie Palmer’s surveys of university patent policies make clear that most universities for a long time did not have a patent policy, and when they did write a policy, often it recorded ad hoc practices–for the vast majority of universities, assignment of patents to the university was voluntary, if allowed at all. With the rise of university-affiliated “research foundations” in the 1930s through the 1950s, university patent policy preferences, where assignment of ownership of patent rights was required, was assignment to a research foundation or to a national invention management organization such as Research Corporation. Here is Palmer’s description of the situation in 1948:
Palmer repeats this paragraph, verbatim, in his 1952 and 1962 surveys of university patent policies. Even as universities developed formal patent policies, they retained the voluntary assignment conditions. A patent policy might be written to encourage university inventors to make use of university resources–for that, a royalty-sharing schedule might be published that was intended to be seen as attractive to campus inventors for a majority of inventions. For the others–it was okay to let them make their own way, at their own cost. Thus, in a thoughtful way, voluntary policies on assignment avoided the egotistic and micromanaging inventors, the pig-headed ones and the ones who thought they could do everything better than anyone. These voluntary policies also stayed clear of the inventors who really were capable and connected and wanted to manage their own work.
Even when research is involved, universities preferred voluntary assignment for faculty. Here’s Palmer from his 1962 survey:
Palmer reports there is “no uniformity” in how universities handle research–this language persists from the 1948 survey to the 1962 survey with only minor alterations. Here’s the 1948 version:
Apart from whether there is or is not a policy and whether that policy dictates research terms and conditions, Palmer’s description of the situation is instructive. Some universities require extramural research to be “definitely related to their educational programs” so there’s no alteration of their instructional programs as a result. One might think of this as instruction-based research, where students and faculty undertake the research as part of regular courses. A present-day example is the 3d printing laboratory at the University of Washington, the subject of the Open 3DP blog, where faculty and students work on projects as part of courses in mechanical engineering. A second set of universities set up “special facilities” and hire research personnel expressly for sponsored projects in these facilities–“relieving” faculty from any requirements (including assignment of patent rights provisions) in this work.
There were two areas of primary concern in such policies–(1) whether the university would claim ownership of inventions and (2) the scope of any such claims. Consider, for instance, the ownership and scope paragraph of the University of Missouri patent policy, adopted in 1955 and revised in 1956 (and many times thereafter):
Part (a) establishes a claim and scope of the claim. Part (b) confirms that everything else is not subject to a university claim. Let’s look at part (a). There is a main paragraph that states a claim and two subparagraphs that define the scope of the claim. The claim made is broad, on the face of it–“The University…shall have ownership and control of any invention developed in the course of the employee’s service to the University.” But following requirements for disclosure, the policy then defines the “course of employee’s service” as “the general scope of the employee’s duties.”
The scope, by policy definition, then consists of two possible situations. In the first, an employee must be “assigned” to “an inquiry” and the invention must be “relevant to the general field” of that inquiry, where the inquiry is “research or investigation.” That is, this first subparagraph focuses on employees who are hired or assigned to a specific research project. Such employees would generally not include regular faculty, who are not assigned to any research project by the university.
The second subparagraph concerns the use of university resources–if an invention is made “in a substantial degree” with the university’s “facilities or financing” or “on University time” or with non-public “University information,” then the invention is within “the general scope of the employee’s duties.”
These two subparagraphs are set out to define duties relative to invention obligations, not to illustrate these duties with two examples. The “shall be considered” header for the discussion of scope is not of the form of “for example” or “by way of illustration”–the wording is direction to administrators with regard to what criteria to use to establish scope of duties for the purposes of making a claim on ownership of inventions. If one is assigned to a research project and the invention is relevant to that project, then the invention is within scope of the university’s claim. If one uses substantial resources of the university–facilities, money, time, or nonpublic information–then the invention is within scope. Nothing else is.
This sort of restriction for personnel assigned to research tasks makes sense. In those cases, one might argue, the personnel are hired to invent. Assignment is a reasonable part of the bargain. A similar case, especially at state universities, might be made for significant use of university resources. If one works in a university facility set up for research, then one is either pursuing personal stuff or public stuff–and if public, then it makes sense to claim ownership of any resulting inventions (even if from a *diffusion* perspective, it makes little sense at all to make such a claim, any more than claiming ownership of copyright in a novel or screenplay because the author used a university library for research).
In any case, the Missouri patent policy restricts its claim on inventions by limiting the definition of “course of an employee’s service” to two specific situations. Even with all the changes in Missouri patent policy–turning it into a sloppy mess, consistent with the practice at most American universities–the university has retained this primary definition and set of restrictions from its original policy. Whatever an employee’s general scope of duties to the university might be for other considerations, for any university claim on invention, those duties are defined in the patent policy and restricted to two situations–relevant to an assigned inquiry or making significant use of university resources.
A number of university patent policies have similar provisions that limit the scope of claim by special definition or demarcation. If one read the Missouri patent policy’s first sentence only, it would appear that the university claimed anything any employee might do that was remotely “in the course of service” to the university. But such a reading would be misleading, and if done by an administrator, would be simple malpractice, since the burden of the policy is to make clear what, for invention claiming purposes, is meant by duties for the university.
If inventions obey a power law, then 20% of inventions ought to produce 80% of the value. Similarly, 20% of inventions will take 80% of one’s resources and time. A smart patent policy, therefore, should aim at attracting primarily the 20% of inventions that the university is prepared to handle well, and try to match those inventions with the ones that it wants to spend its time and resources on. A policy that claims everything, and then obligates the university to at least make a show of managing each and every invention, is just plain stupid. Even corporations don’t claim everything–they limit their claims to inventions of interest to the corporation’s business, they don’t file on stuff that they might own when a trade secret approach would be better, and they ignore or waive the rest. Universities would do well to learn, if not from reason and evidence, then from example.
It was not until Bayh-Dole that university patent administrators forced patent policies to change from this default. The process was not an obvious one. It was a clever scheme. The key point, however, was that some people dearly wanted the university to force assignment of patent rights to the universities, and it was these who have, for the time being, won out.
A thumbnail sketch of the change might go like this:
Most American universities involved in extramural research established formal patent policies in the 1930s through the 1960s. These policies for the most part established assignment of inventions as voluntary and set up mechanisms by which inventors had access to invention management services, often through an affiliated research foundation or a national agent such as Research Corporation (or both). Policies might allow for a university claim in inventions for employees assigned to specific research projects or working in facilities set up for contract research–that is, “scope of duties” included an expectation of invention. This approach created a working system of invention management agents, most external to the universities hosting research, to which academic inventors might turn if they decided the pursuit of patents was appropriate.
As federal government funding for university research after World War 2 expanded from a small budget routed to a handful of universities to much larger allocations and to many universities, most federal agencies used an ad hoc approach–considering each invention on its merits and relation to possible government use and regulatory interest. Some agencies were liberal–notably the Department of Defense–allowing contractors to own inventions, provided the government had a right to use the invention. After all, for most weapons systems and military applications, the government expected to procure, not sell, this stuff. Other agencies were cautious–here, notably the Public Health Service under its various names. The PHS argued that publicly supported inventions ought to be held in trust for the public’s benefit, and not become a monopoly property of anyone–not the inventor, not the university that merely hosted the work, not some company aiming to use the monopoly to force a higher price for a therapeutic compound after the key work, expense, and risk for the research had been undertaken at the public’s expense.
It was to these inventions, otherwise claimed by the government, to be licensed non-exclusively, that the university research foundations wanted access to, to license exclusively, in keeping with the preferred practice of the pharmaceutical industry. Thus, Bayh-Dole, the great work-around to PHS policy aimed at providing inventions made with its support broadly to all, over the objections of the pharmaceutical industry, which wanted exclusive positions so each company might gain “big hit” products that would earn billions for shareholders. Bayh-Dole forced government agencies to allow as the primary default of contracting for research services with nonprofits and small businesses the private ownership of inventions made with federal support.
The university-affiliated research foundations and university patent administrators then claimed that Bayh-Dole required university ownership of inventions, or mandated it, or prevented inventors from assigning other than to the university that hosted the federally supported research, or gave the university first right of refusal, or allowed the university to acquire title simply by notifying the federal government that it “elected” title. All of this was a pile of rationalizing administrative crap, and when universities got cocky enough to escalate a dispute over patent rights to the US Supreme Court, the court tossed the whole norming myth into the muck hole on the back 40. But in the thirty-odd years from the effective date of Bayh-Dole in 1981 to the Stanford v Roche decision in 2011, university patent administrators demanded that patent policies be changed to bring the university into compliance with federal law, even though no changes were required by Bayh-Dole to university patent policies.
All the standard patent rights clause in any federal funding agreement required was, in addition to reporting inventions and giving notice to the government if the university intended to retain title to any federally supported invention that had been assigned to it, that universities delegate to potential employee-inventors the responsibility to report inventions and to assist in the filing of patent applications, and the authority to convey rights in inventions to the government. Instead, however, university administrators insisted that universities must claim ownership of all inventions made using federal funds, and to ensure the proper delivery of all rights the government might claim, a university must also claim any potentially related inventions, and claim anything that might not appear patentable because it might turn out later to be patentable. Thus, universities changed their policies or changed the interpretation of their policies to expand their claims of ownership in inventions.
Rather than routing federally supported inventions suited to an external invention management agent’s expertise to that agent rather than leaving those inventions with the federal agency that provided the funding, Bayh-Dole ended up being used to claim all inventions made by university personnel. If there was a misconception in Bayh-Dole, it certainly wasn’t that universities already claimed ownership of inventions as a matter of employment contracting–anyone reading Archie Palmer’s accounts could see that this wasn’t the case. Only when specific contract provisions required university ownership (as with the PHS Institutional Patent Agreement) was there an obligation to assign inventions to the university. And Bayh-Dole did away with the IPAs and their institutional assignment requirements.
The misconception in Bayh-Dole was that it was good policy to claim everything and then sort through it for what could be profitably sold off to speculative investors. Once administrators got the smell of ownership, then they could not restrain themselves. They rationalized the importance of their task, the limits of academic freedom, the secret true purpose of federal policy, the hope for millions in licensing revenue, the desire to be seen as central to the great endeavor of research enterprise and economic vitality and public benefits. For this, taking ownership as a matter of principle seemed to be a fair thing to do.
At the University of Missouri, for instance, administrators conveniently ignored the two restrictive subparagraphs that defined “general scope of duties” for purposes of invention ownership claims and simply claimed to own any invention anyone associated with the university happened to make. They went so far as to cleverly include an assignment statement on their invention disclosure form, so that to comply with the requirement to disclose all inventions necessarily included, upfront, the assignment of the invention to the university, even if the invention had not been made in assigned research or with significant university resources. Anyone objecting would be threatened or sued, all in the name of goodness–much like the Roman empire, five years after adopting Christianity as the state religion handing out death sentences to anyone who thought otherwise. Norming myths put the moral sense to sleep, to be awakened, if ever, with a “my God, what have I done?”
Claiming everything to sort through it, just in case something of value was missed or concealed, is a rotten way to deal with innovation, even if it is one way to make a living with reasonably good pay. Earlier university approaches, involving voluntary assignment for the most part, fostered selectivity. Inventors decided how to deploy their work. If patenting was appropriate, then they might choose local resources for that purpose, or if they wanted to spend their own money, or were connected with well endowed investors or companies, then they could go that route. This selectivity was hugely important. It matched inventors with their preferred pathways. It allowed for a diversity of approaches to diffusion of invention, from venture-backed speculative monopolies to fair and reasonable non-discriminatory licenses to indifference and the public domain. It allowed invention management agents to decline the opportunity to manage. Dual selectivity–that the inventor decides to pursue patenting, and an invention management agent may decline to get involved–is far superior to dual monopoly–that universities own everything and license to create monopolies.
The university infrastructure around invention management, prior to Bayh-Dole, was effective because it was selective. In the thirty-five years after Bayh-Dole, that infrastructure has been dismantled and replaced by a model that is more restrictive than even the ones corporations generally use. As a consequence, universities claim to be the owners of most everything, and find that the pure volume of disclosures–often over 200 inventions per year–is well beyond their ability to manage. A typical load for an industry-based patent management agent might be 20 inventions. A university licensing professional may have 10x or 25x that number. I routinely had 400 to 500 cases, and my portfolio was (with effort) on the low side at the University of Washington. It’s not possible to “market” that many things, let alone stay current on the patenting issues, changes in industry, available opportunities, and new research developments.
To deal with the volume and with the plain fact that most inventions claimed by universities are never licensed, or if licensed come to nothing–or because they are licensed exclusively are assured to come to nothing–university administrators add statements to policy to make it clear that they are not agents of the inventors, have no obligation to license any given invention, have no obligation to make money on any licensing, or to maximize the money they might make, or to litigate to enforce any license they have granted, or to sue for infringement. In essence, the practice ignores the typical platitudes that university patent policies start with–about how the patent policy is there to encourage the development of beneficial products, and stimulate the economy, and recognize the interests of everyone involved. Nothing of the sort happens. Instead, one gets hundreds of inventions each year piled up on the inventions from previous years. Over twenty years, one ends up with 200 or 500 or 5000 patents, depending on how much money a university wants to burn off for pieces of paper with the word “Patent” on them.
Invention ownership volume does not increase the “odds of success”–rather, it diminishes the odds by burying the self-selecting inventors and inventions under a load of noise, of rubble, of folks that hate every little bit of you, of folks that think your only role in life is to make them rich–and the university cannot, by policy, say no to any of them. In the worst rationalizations, worthy of Catch-22, if an inventor does not want to work with a university office, that must mean that the invention is more valuable than ordinary, or that the inventor has a private conflict of interest that must be disrupted or exploited, or the inventor is deliberately concealing the true value of the invention, or the inventor must be taught a lesson about authority.
If an inventor wants back ownership of an invention that has sat for a decade in a university licensing outfit, that must mean the invention has value, or that the inventor knows where to license it but isn’t tell. So either the inventor is made to reveal the source of value or should have to pay that value upfront or should have to at least repay the university for its patenting costs (even though those patenting costs were not incurred at the request of the inventor, but as a matter of policy). And if an invention doesn’t have value, then it is not worth the time to return to the inventor. Thus, the university becomes a black hole, a kind of Hotel California, programmed to receive. But in doing so, it becomes a lousy supporter of innovation, and a lousy supporter of its own programs, and presents a lousy face to industry and to the public. Far from being the shining lights of economic prosperity through proper patent management, universities have become another source of hype, monopoly, and special promotions for speculators. Yes, there is some money in doing so, but it is lousy money, used in lousy ways, at some great, usually unreported, expense if not damage to the research community and to the broader economy of innovation.
How to change? First, recognize we are dealing with a prevailing narrative myth in which people not only take comfort because it provides them with how to think and act, but also because people have livelihoods that they believe depend on this myth staying in place, and further because there is “evidence” for the myth (because most everything else is suppressed, scorned, excluded), and finally because people have a tough time admitting they are wrong, have been duped, failed to notice, or slipped into a bad habits. Narrative myths, like failed prophecy, tend to make their adherents more determined in the face of disagreement, more evangelical.
It would appear that to make changes in university patent policies, to open things up, to get back to where things were, or to undertake some “creative destruction” that Schumpeter promised should be just around the corner any time now, we will need a revolution, or conversion, or confession, or reconciliation proceedings to release folks from their recent pasts, or a new commitment from senior administrators, or legislation in the form of a bill of rights for university inventors, such as the freedom to innovate legislation state Sen. Maralyn Chase has sought in Washington state. A university faculty could do it, if they wanted to. Otherwise, we are left with a uniformly obstructionist, bureaucratic approach to innovation based on university research–hardly something to be proud of, and something that, soon, will require federal action to replace or modify Bayh-Dole to reset the balance away from the moldering invention piles of dual monopoly toward the freedom of dual selectivity.