A primary argument for university involvement in the management of federally supported inventions was that university agents were reporting something like 30% of their inventions under management were being placed with commercialization partners, while the federal agencies’ rate was something like 2%.
What is not so generally discussed is why the university agent rate was so much better. At the time, university agents (and with only a few exceptions, not universities directly, and as far as I know, no public universities at all) operated on a basis that was highly selective. That selectivity in turn was comprised of five components:
- Inventor or Investigator interest in the invention
- Inventor choice of agent
- Review by agent for match with expertise and operating model
- Review by agent for commercialization opportunity
- Negotiated management arrangement
In the thirty or so years since Bayh-Dole, we have seen this selectivity disappear from university practice as technology transfer has moved from an outside practice to an administrative function, and move as well from voluntary to compulsory, and from narrow to comprehensive in claims. The first to go as these moves took place were the choice of agent (the universities began choosing themselves) and the negotiated arrangement (policies moved to pre-set royalty sharing schedules, which also then pre-set the level of inventor involvement in subsequent efforts to develop the invention). The issue of inventor interest was disposed of by making ownership assignments compulsory. A few institutions preserve the idea that inventor interest matters. Stanford, despite eliminating the inventor owns “whenever possible” language, still allows inventors to dedicate inventions to the public domain rather than seek commercialization income from licensing. Wisconsin, in the case of federally funded inventions at least, lets inventors choose between assigning to WARF and dealing with the funding agency.
With the move to present assignments, we are seeing even the review for appropriateness and potential disappear. The new university approach intends to take everything, and figure out what to do with it later. This is beyond even the federal agency approach to inventions, and well beyond the corporate approach. University administrators are expanding their claims from patentable inventions to any sort of intellectual property, and are expanding the definition of intellectual property to mean any intangible asset they can imagine (and some they can’t!). Even with a corporate approach, there are bounds set on the interest the company will have in inventions–they have to pertain to the business or future business of the company. Otherwise, why invest any time or effort in them? But this limitation is not available to universities, which see the world as their field of study, and therefore everything is of potential interest for licensing opportunities.
University licensing offices have become so desperate and paranoid (I can’t tell which) that some refuse even to return assigned inventions when they clearly cannot do a thing with them. The reasoning goes like this: if the inventors want the invention, then they must know something about it that they aren’t telling, so the invention must be worth something. Either the inventors come clean so we can do the deal they must know about, or they have to pay to get the invention back.” Pretty wicked stuff. Not the stuff of trustees on behalf of the beneficiaries of publicly funded research. Instead it is institutional self-interest over all. Money is more important than opportunity. (Oddly, this is also the principle enfranchised in what goes for “state ethics laws”–which makes them, in their implementation, about as unethical as one could have, when it comes to matters of faculty research, instruction, and public service.)
Is this progress? Is it innovative in its way? Or are we simply reprising the same propensities that afflicted the federal agencies with the ramp up of federal funding for university research with the formation of the National Science Foundation? If the latter, then the problem is compounded because it appears there is no oversight of the university activity. Their self-reporting is limited and fails to provide a clear accounting of activity. Bayh-Dole itself puts the annual reports on the utilization of federally supported inventions within the scope of FOIA, and universities have no obligation to report their non-Bayh-Dole invention activity. I don’t know of a university that reports its operating expenses for technology transfer, or reports the status or outcome of each invention that it takes under management.
When the National Science Foundation was formed, it was required to report its activities to Congress. Now that the responsibility for invention management has shifted from the NSF and other funding agencies to inventors choosing agents (via Bayh-Dole), and from there to universities requiring their inventor employees to choose the university as their agent (via university policies and employment agreements), thereby frustrating the expectations of Bayh-Dole and capturing as much inventive work as possible for institutional self-interest, it is time to require whoever does end up with ownership of federally supported inventions to report to Congress their activities with each invention, every year. The report can be quite simple and need not reveal any patent prosecution or licensing financial terms: a unique title, general subject matter, and status–claimed/owned/released/assigned–year disclosed, year licensed, exclusive or non-exclusive, year of first commercial use, year of first commercial sale, recovered its patenting costs or not, year licenses generated income >$1m cumulative.
Apart from reporting, which would be a first step toward accountability, the federal government, industry, and inventors alike should take a critical look at university invention practice. It now looks remarkably like pre-Bayh-Dole federal agency practice, except it is less selective, less accountable, less concerned about exclusivity, more willing to litigate and troll, and more taken with institutional self-interest than the public good. More particularly, patent administrators have attempted to conflate institutional self-interest with the public good, as in, “it’s a public good that the university is greedy, because that greed leads to licensing income, and that licensing income… er… is used for nothing but slop money… er… but that (brightening) is slop money that otherwise would have come from taxpayers or tuition, so that’s a huge benefit for all, and besides, Emory built a whole new building with its big hit deal, so there!” Is this where we want Bayh-Dole to end up, as a federal regulation that provided a claim of institutional ownership of federally supported inventions–a claim that has now been shown to be unfounded? It would not seem to be in the list of objectives in Bayh-Dole.
The sharper problem for university licensing operations is that the failure to be selective means each office is carrying more technology by far than it can manage, let alone license, and becomes the bottleneck for all sorts of collaborations that otherwise would not carry an institutional intellectual property interest. The bottlenecks started with sponsored research with industry, and then to federal consortium with industry members, and have now spread to open software, to consulting, and now to open hardware. Whatever the “success” stories that may come from the technology transfer activity, non-selective licensing offices can only be considered fundamental obstructions of science and scholarship.
This is a real shame. The federal investment in university research was founded on faculty talent, free of institutional controls while supported by institutional resources, receiving funding by virtue of their public commitments held in place by academic conventions of public service. Inventions arising in this research form part of the library of assets that faculty create to advance research, scholarship, and training. It is from this library of assets that industry and entrepreneurs may draw to assist them in their pragmatic and creative work. The idea of “technology transfer” in its Vannever Bush form was that universities would reach out to industries that do not typically have the benefit of working with new science and technology. Biotech would be the last industry on such a list, as it already has great working access to university science. One might think, perhaps, food processing or textiles or plumbing would be good places to consider for technology transfer.
Rather than having technology transfer select a few distinctive inventions for management, we now have technology transfer diverting a substantial portion of the outputs and inventory of university research for the purpose of institutional self-interest. If one believes that innovation comes about because institutions hump up their self-interest, invoke processes, and spend millions indiscriminately filing patent applications–and calling that a virtuous service to the faculty and the public–well, then, this is nearly heaven. Otherwise, it would appear that we have reached the end of a long dead-end for Bayh-Dole. The law was a good one–exceptional in a way–but the university response has been misguided, and eventually it has to be characterized as ineffective and corrupt. Small changes in the law can mitigate the adverse effects. Require public reporting. Make the (f)(2) agreement trump any non-voluntary obligation to assign to a nonprofit employer. But these cannot get at the fundamental problem of turning technology transfer into portfolio trolling, from public spiritedness led by faculty inventors to institutional self-interest led by minorling adminstrators who have like mosquitos managed to get their noses into the bloodstream of research and are now injecting live virus that will infect the whole system.
I don’t see anything for it. Another five years of present assignments and accumulation of inventions that go unlicensed, with industry pissed off, states induced to spend millions on grand schemes to start unfundable companies realizing they’ve been had, and the public increasing disaffected with the whole overpriced, overhyped, underachieving mess. Then it will all become apparent. If folks want to change without legislative action, then universities have to return to the five principles of selectivity that justified the public’s trust as represented by Bayh-Dole. It costs next to nothing to change back. Roll back compulsory ownership claims. Make the royalty sharing schedule with inventors a default from which anyone can negotiate. Make review come before assignment. Allow choice of agent (but one can still require the use of an agent). Rejoice in the stuff that gets done outside the tech transfer office. Rejoice in the stuff that gets done within it.
Do this, and important things happen. The office portfolio is reduced to workable levels. The range of personal initiative (and personal responsibility) increases. Opportunities for supporting agents in the community grow, building additional and specialized resources. Otherwise, Bayh-Dole must be revised or repealed, or states must pass freedom to invent statutes that prohibit compulsory assignment of IP by university faculty or students based on employment, use of facilities, or participation in research. The direction things are going cannot go on much longer. Time to correct the course.