After 30 years of effort to build a patent practice around university research, university technology transfer is poised to shift. The pressure presently being put on central control and rigid processes in the name of efficiency is welcome. This pressure exposes the limitations of the model and the problematic intent of those behind it. Rather than developing innovation-in-research, the conventional pathway now lies in research-in-bureaucracy. Of course, it is a sincere, kitten-eyed bureaucracy, dedicated to the public benefit (meaning licensing money for universities), and defending the public (from those awful, greedy, hopelessly inept faculty researchers who could not possibly have any way to develop their research by any other means).
The present model will break because it is
*non-responsive to Bayh-Dole in practice
*compulsory rather than collaborative
*overburdened with inventions it cannot handle
*defensive, anti-innovative, and process heavy
*destroying the distinctive role for university research
That’s okay. Call some place paradise, kiss it good bye. When the change comes about, then the US will be again able to use its university research capabilities to compete once for a place in world innovation markets.
With the changes, conventional research patent methods will still be available to faculty investigators and inventors, but they will represent a resource among others rather than a monopoly franchise increasingly controlled by central administration. As a result, the conventional methods will substantially improve their practices as well as outcomes. This will be a good thing for everyone.
Lesson learned: Bureaucratic monopoly control of inventive output does not create more efficient innovation systems and does not put US researchers at the world-leading edge of expertise and relevance.
How could anyone straight-faced argue that all invention to be any good has to pass through the hands of bureaucrats? Is it better to have the government certify you as good to drive every time you get into your car? Hey, you could have defective equipment, or it could be bad weather, or you are a little tired or ill, or you have forgotten how many feet behind another vehicle to hit the low beams. You could be a bit irritable or feeling aggressive. It would be *better* to have the government check you out before allowing you to drive. Better, of course, if there’s a perfect government, always available, with perfect assessment tools, not capable of making mistakes, not itself represented by tired, irritable, distractable folks. So, back to that argument that innovation is always better if it passes through that happy, capable, altruistic, mistake-free, always helpful office run by bureaucrats. Yeah, that kind of better. The better of an impossible world that isn’t considered very deeply but sure sounds good. A tidy, organized, efficient office armed with grail-like licensing instruments that investors and companies compulsively sign as better than anything they could ever think of.
I am *not* being critical of tech transfer folk. A lot of them are talented, thoughtful, decent folks. I *am* being critical of arguments that set them up for failure. And I am *mocking* those arguments to provide them with some clarity, which they seem to lack on their own.
The current, dominant approach to university research engagement with community is patent commercialization. It is represented as an effort to place each reported invention with (usually) one industry partner willing to take a royalty bearing license, build and sell commercial product, and share income with the licensing university. If one claims an invention in this model, it is because it has commercial potential. The argument is that this is the approach mandated by Bayh-Dole. If that mandate is not expressly in the law (it isn’t), then it was the secret motivation for the law.
We have in other places picked this argument apart. It is empty. There is no juicy center to it. There isn’t another substantive argument around that I’m avoiding. Patent commercialization is not mandated by Bayh-Dole. It is not even a primary focus of Bayh-Dole. It is barely indicated in Bayh-Dole. But it is worse than that.
First, most research inventions do not have commercial potential. That is, they are not as inventions suited to supporting the formation of a product to be sold. They may be suited to other things, such as illustrating a way of going about inquiry or advancing capabilities within the research community, creating an ad hoc standard or cross-licensing to give industry players access to a broad range of pre-competitive technology or marking out incremental advances or building a cluster of claims to exclude others.
Second, some research inventions might some day have “commercial potential” but it is not present at disclosure. The potential has to be cultivated rather than ascertained, and arises based on who holds the rights and who is presented with opportunities.
The commercial potential furthermore may not be based in the sale of product but rather in an internal use for advantage, such as reducing energy costs in a production line. Still further, the commercial potential may be in a market that is not obvious, or lucrative, or may be in a form that will not return money to the university. Finally, commercial potential may depend on factors outside the control of the university, or any broker of patent rights, based on unanticipated or uncontrollable events. The point of this brief account is to highlight how ambiguous the concept of commercial potential is.
All this to say that by “commercial potential” university technology transfer offices mean, “giving us a way to make more money licensing than we spend in patenting”. What they really mean is: “giving us a way, across all of the inventions we claim and licenses we write, to make more money in licensing than we spend on patenting.” In this latter development, winners pay for losers, and losers are 80% or more of the portfolio. That is, in the model of university technology transfer about to fail, more than 80% of the US university research inventions is held *off* the marketplace, and the effort to exclude is *paid for* by the “success” of a tiny portion of the portfolio–about 1%. There are reasons why this is so. There are reasons why, as to how an IP portfolio is managed for income, this is perfectly well formed. It’s just *not* healthy for the role university research plays in the broader community. That’s why the model will fail. The engagement of university research with community is way more important than bureaucrats making money with patent licenses. The bureaucrats have the audacity to argue otherwise. That’s what will bring their model down.
The reality is that university technology transfer offices are not able to match each invention with a commercial opportunity. It’s just not possible. Hire anyone you want, pay them what you will. They cannot do it. Anyone who says they can is spouting it. 30 years shows it. It’s not done anywhere. Not in government, not in industry. It’s a thought that on inspection is empty. It is a null thought. A dead parrot.
Rather than waiving the inventions it cannot put into play immediately, however, a university technology transfer office worries that at some future point these inventions may come to have commercial value and therefore is willing to file patent applications and hold inventions for future licensing opportunities, even if it is no longer “actively marketing” the invention.
At that point, the university is no longer meeting its Bayh-Dole mandate to use the patent system to promote the use of the inventions arising under *each* federal research award. Instead, the university lapses to a portfolio model, or rather a patent accumulation model. Rather than keeping a steady state at the carrying capacity of the office, the technology transfer office maintains a robust intake function, filing on inventions just in case, making an attempt to “market” the invention by writing a non-confidential summary, posting the summary on its web site and perhaps with a broker site, and/or sending out unsolicited announcements to companies drawn from a business database. Typically there are no takers, and the invention is pushed aside by a new set of invention reports.
The patent accumulation model as practiced is not described by any university technology transfer office. No office says: “we acquire invention rights from our employees without regard to our particular expertise, we file patent applications and hold the resulting patents for their useful life, awaiting opportunities to shake down industry opportunities that eventually may take advantage of these inventions.” That, however, is the model in practice. If it were not, then a university office would generally not hold inventions with rights unlicensed longer than 60 months from the date of first disclosure. While patent accumulators do not have to become patent trolls, doing so is the time-honored way of making money using the patent system. We emphasize that Bayh-Dole does not mandate to use the patent system to make money from licensing, but rather to promote the use of research inventions by licensing. Many of the problems presently faced by university technology transfer arise from the conflation of these two goals.
We can see that the emphasis by university administrators on commercialization is mostly misplaced. It has created patent accumulation services rather than technology transfer services. They may not *intend* that, but it *is what has happened*. At some point, one has to accept the connection between choices and outcomes, quite apart from one’s protestations of intentions. Intentions without regard for the truth are just another form of bullshit. The conflation of promoting inventions and licensing patents for money has interfered with industry supported research and collaboration, has created uncertainty and risk for industry and for other research organizations in the development of new research efforts, and has shifted valuable resources to the effort to make money from licensing and away from those activities that would promote the use of inventions—that is, their transfer.
Again, because it always seems to come up, this critique does not require the displacement of commercialization by some other “better” approach. But this critique does argue that one cannot expand commercialization to include other forms of intellectual property or make a show of marketing efforts for longer periods of time. Expanded services commercialization efforts are perhaps the worst thing that the patent accumulation model could do. This critique argues that commercialization would do better if it were *radically* selective, dealing in ten inventions a year rather than 100 or 20o or 300. And making those selections on a basis other than a monopoly position, compulsion, or an oversight committee stocked with one’s venture buddies cast as experts in commercialization.
Furthermore, this critique argues for the growth of other existing methods of transferring technology, greatly reducing the routing of research events such as invention through a single commercialization office. Routing through a single office is antithetical to public research, mixes noise with signal, wastes many valuable resources in inventor reporting, triage, filing applications, and dealing with the consequences of operating with a mostly moribund portfolio of patent rights. Patent commercialization is one option among others within a technology transfer function. Policies that make commercialization compulsory, or that limit other alternatives, must be revised as non-compliant with Bayh-Dole, against public interest and the mission of universities, and a potential threat to the role and effectiveness of public research. Luckily, things are going to change. The present system has had its day and its failure is evident everywhere. That NAS can’t see it, I can’t help that. Perhaps they didn’t want to kick out the last supports for a collapsed building. Perhaps the system is so bad they just don’t care about it. Perhaps consensus is the hallmark of the cutting edge these days, so the last thing one would want to do is to be an outlier.
Other approaches are not necessarily “better” than patent commercialization in commercializing patents. They may however be “better” than patent accumulation in finding opportunity and even in making money for the university. Other approaches are most certainly better at things other than patent commercialization. One can go so far as to argue that the presence of other approaches is essential to the improvement of patent commercialization and the disbanding of large-scale patent accumulation practices.
What are these other approaches? Here are a few.
Open innovation
Open source
Research standards
Commons
Agents and other intermediaries
Intangible intermediaries
Doing nothing, nothing at all