We will get into the operational details of non-exclusive, voluntary, negotiated university IP management for effective technology transfer. Short answer–everything is navigable and has already been done, even if folks have forgotten how to do it. We can look at what technology transfer is. We can fuss over the silly nonsense of that patent monopoly meme. We can wallow in Bayh-Dole compliance nonsense. We can show how things work in practice down to forms of agreement and network dynamics. But for now, here’s an overview.
Are you a vice president of research at a university? Do you want to make a lot of money from new sources of revenue? And you want good publicity? And you want to grow your research programs? And you want your faculty to stop bitching about IP and return to bitching about IT or parking, if there’s ever any need again for parking? Then adopt a patent practice directed at effective technology transfer, not at harvesting patent rights to shop exclusively to speculative investors and companies willing to take what’s on exclusive offer. You can do it now. It is easy to come in from the cold.
The appropriate purpose of university IP policy is to protect those involved in research from the claims of the institution. Think of IP policy as a bill of inventor and author and public rights, relative to claims that the institution might otherwise make. This bill of rights thing is a good model that has been used elsewhere. Follow it here, as well.
I have sketched an outline of a university patent policy for effective university technology transfer. Core elements: no exclusive licensing, no compulsory institutional ownership, no pre-set royalty schedule. These elements are directly aligned against most present university patent policies. What? No exclusive licensing? Yup, none of it. But these elements are strongly aligned with effective technology transfer. They are also aligned with the core of university research policy–especially the parts that assure faculty of freedom of research and publication. These elements then are strongly aligned with academic values and public expectations.
There is nothing especially new about these three elements. They were staple in university patent practice for decades–from, say, 1912 to the 1980s and 90s. Many great instances of university-related licensing have been non-exclusive. Insulin, the electrostatic precipitator, vitamin D inducing radiation, warfarin, gene-splicing, SPICE, X-Windows. Relatively few exclusive deals have made big bucks for a university and resulted in a commercial product–and even then, many of those products end up being sold at monopoly prices or at the loss of multiple suppliers, competition, collaboration, variations on the invention and applications in areas not developed by an exclusive licensee. Perhaps the turning point was 1977, when Research Corporation exclusively licensed Rosenberg’s (via Michigan State) cisplatin to Bristol Myers. Rosenberg later wondered why there had not been much development of his invention beyond FDA approval. Maybe, just maybe, that exclusive license for effectively the life of the patent had something to do with it.
University administrators used Bayh-Dole (1980) to argue for compulsory university ownership and exclusive licensing–neither of which is supported by the law. But they got university patent policies changed anyway “to comply with federal law.” University administrators demanded ownership of inventions made in projects with federal support. They claimed Bayh-Dole vested ownership of such inventions with the university that hosted the federally funded work, not with the inventors. That was not a true claim, as it turned out, but in making the claim for decades the administrators wreaked compulsory assignment on university policies. And to be safe–just in case there could be compliance issues, so they said, the administrators demanded ownership of everything else. In a weird twist, administrative argument was that it wasn’t properly fair if inventions in work with federal support must be managed by the institution and other inventions weren’t. All inventions should be equal that way. The administrators’ idea was that institutional control was so much better, so everyone else should be forced to the institutional ownership–but in practice things work differently. “If these inventions must get squashed by institutional control, then it’s only equitable that all these other inventions should have to also get squashed in like manner.”
You may have read that the Bayh-Dole Act has been wildly successful, and that university licensing operations have also been wildly successful. The reported metrics, however, do not support the claim. Even the financials hide the reality. A university can report a financial success for its licensing operations for two or three decades on the strength of a single patent family and one license, while everything else the university touches turns to stone, even with hundreds of issued patents, and plenty of other licenses. Lots of activity, but only a single deal. Lesson learned: more licensing income does not find you more lucrative deals. Running a technology transfer program on the hope of finding one lucrative anchor deal that will last two decades is the stuff of gambling, unsuited to university practice, and certainly nothing to do with federal invention policy.
Over the past forty years, we have seen universities force a compulsory ownership default and an exclusive patent licensing practice on technology transfer. The new approach does not work, has never worked–even when there are licensing “successes” the approach appears to have been parasitic rather than catalytic, and the “successes” booked as million-dollar deals are less impactful than they would have been had the parasitic university ownership and exclusive patent licensing approach not been in place.
What’s new, then, is that it’s time for universities to return to serving as a platform for non-exclusive access to technology, whether new or otherwise. There are good reasons to return–making money, for those who value that sort of thing–among them. The best reasons, however, have to do with the advantages for effective technology transfer.
Though these days it doesn’t count for much, I will add that these elements–non-exclusive, voluntary participation, negotiate each arrangement on its circumstances–also keep a university out of unmanageable institutional conflicts of interest and breaches of federal law, including Bayh-Dole. By contrast, a compulsory ownership policy backed up by a nudge-nudge-wink-wink default exclusive licensing practice with a set-in-stone royalty sharing schedule creates those conflicts of interest, violates Bayh-Dole, sets up potential unrelated business income tax problems (just you wait and see), and drives away opportunities, relationships, talent, funding, and political support for university research. And that’s just getting started. But since this sort of stuff doesn’t count for much, and because confirmation bias runs deep–if you hear it many times, you tend to return to it when pressured, even if you have never really believed it–we can add a kicker.
The elements for an effective technology transfer program are strongly aligned with making money–if making money is important. An effective university technology transfer program will, over time, make more money and better money and from better sources–or more accurately, raise more money, rally more money, assemble more useful resources that do useful things at lower costs and more quickly and more opportunistically and with a much greater sense of what is worth doing–than does any patent licensing program rooted in compulsory institutional ownership and a default practice of seeking to license patents exclusively.
There are other advantages to adopting effective technology transfer policy and practices. You get much better relationships with university talent. You have an easier program to operate. You have a less expensive program to operate. You have lower liability exposure and way less bitterness in negotiations. In an effective technology transfer program, people want to pay because they are getting good value for their money, because they want the technology transfer program to be effective. And (did I mention?) you make money if you want that sort of thing.
The best way to implement an effective technology transfer policy is to repeal the existing crappy policy. Nothing at all is better than a crappy policy, and most universities have crappy policies–poorly drafted, inconsistent, stupid, and directed at really awful practices for technology transfer. But repeal is slow and impractical. A vice president for research can implement a decent policy for technology transfer within existing crappy policy with only a little effort.
University patent policies routinely claim ownership of all sorts of assets, but these same policies do not require university administrators to enforce those claims. While a number of universities have built out policies on how they might “give back” patent rights once the university has taken them (and those procedures are not pretty ones), there is nothing that precludes a university from declining to take the ownership position that policy claims for it. Just exhaust the policy claim: “If you disclose an invention to the university as a requirement of university policy, the university will waive its claim. If there are other reasons why the university must own the disclosed invention, then those other reasons, not the university’s policy ownership claim, control.” Thus, if a university has agreed to own inventions made in a given research program–those involved have agreed to this requirement–then that agreement controls, not any policy statement. Otherwise, the university owns only what it choose to own, and what it chooses to own should be what is suited to be made available non-exclusively on fair, reasonable, and non-discriminatory terms. If some invention is not suited to non-exclusive access, then there’s a different administrative checklist, but that checklist does not include taking ownership and trying to grant an exclusive license.
This is a fundamental: a university should never attempt to own what it does not want to own. And a university should not want to own anything that it cannot make broadly available better than others could make that thing broadly available. That is, a university should not want to own any technology that it cannot, by owning, transfer effectively.
This approach works, as well, for the current craze of adding “present assignment” language into policy statements with argument that then immediately upon creation any given invention has been assigned to the university. By implementing a default waiver of ownership, the present assignment is as if it had never had effect, even though it is sitting there in policy. There’s a good argument such present assignment wording has no effect anyway, but the point here is that such wording is ineffective technology transfer practice, has nothing to do with Bayh-Dole compliance, and (if this is what makes you tick) gets in the way of the university making money from IP positions.
Once the policy ownership claim has been exhausted, then the university and inventors can address any other reasons why the university ought to own a given invention, and what arrangements should be in place if an invention will be assigned to the university. If an inventor insists that the university must default to exclusive licensing, then the inventor should hire someone to do that sort of thing. For instance, if an inventor wants to start a company and move patent rights to that company, it is way more efficient for the inventor to deal directly with the startup than for the inventor to assign to the university and then the university licenses or assigns to the startup.
There is an objection around that universities cannot do royalty-free, non-exclusive licensing because they have a royalty sharing schedule that creates the expectation that the university will make money through licensing. But there’s nothing in university patent policies that stipulates that universities must make money on each invention they claim. Quite the opposite. The policies state making money as an aspiration but do not commit to making money for any particular invention. The policies also disclaim that the university acts as an agent of the inventor or and when they bother with it, disclaim any fiduciary duty to inventors, but for sharing whatever royalties the university might receive. If inventors don’t have to assign by policy (because university administrators uniformly waive the policy ownership claim), then inventors have nothing to fuss over that the university is not trying to make money for their invention.
Furthermore, there’s nothing at all in the notion that exclusive licenses make more money than non-exclusive licenses–Cohen-Boyer clocked in at $260 million, all non-exclusively done. And universities routinely grant industry research sponsors a royalty-free non-exclusive license to inventions made within any given grant. The NSF all but requires (“all but,” to circumvent Bayh-Dole) royalty-free non-exclusive licenses to industry members of cooperative research centers. Universities do non-exclusive licenses all the time–but not as the default.
And if inventors were to get all worked up that a university is not making money on their inventions, then they would do it for the present situation under a compulsory ownership policy with an exclusive licensing default. Subtract out the licenses under industry sponsored research agreements and NSF CRCs. Look only at the inventions the university has acquired because policy claims them. How many of those inventions are even licensed? Have made any money beyond the costs of patenting? Maybe 10% are licensed. Of those, maybe a handful have earned more than their patenting costs plus some minor upfront fee. Everything that’s not being used because the university starts with a default expectation of trying to get an exclusive license rather than announcing immediately FRAND non-exclusive access is ineffective technology transfer.
This same line of practice extends to inventions that a university does have ownership of. No university patent policy directs a licensing office to default to exclusive licensing of patents. I’ve been through hundreds of university patent policies. There’s never a requirement that a university must start by trying to find an exclusive licensee. So, as a matter of practice, forbid the exclusive license as the default position. If you have an invention in-house, then offer it for non-exclusive license. If the invention has been patented, then check to see what patents cite your patent, and see what companies own those patents. Offer those companies a royalty-free, non-exclusive, paid-up license for the life of your patent. Waive any claims for any past infringement. If you can, offer each company access to the inventors or research team, to current research involving the invention, to demonstrations of the invention in action, and the possibility of getting updates as the research progresses. Consider an annual research review meeting. Company folks may be fine paying for any of these things–they do so all the time.
Think about it this way. If you get ten companies to sign up for a research review meeting plus a year of reports and updates from the team for $5,000, you have $50,000 of revenue for $10,000 in costs. And you have at least ten licensees. If you try to get ten licensees, each paying $5,000 with free assistance thrown in, the process goes to review at the company’s legal department. The review starts with “do we really have to take a license” and “is their patent even valid” and goes downhill from there. Not effective technology transfer. Worse if you contact ten companies and tell them that only one of them will get an exclusive license. What do the rest think? Yes–how do we avoid these university jokers and their crap patent? Again, not great technology transfer.
Nothing in a university’s patent policy, crappy as it is, is so crappy as to require a university to take ownership of everything the policy claims, and nothing in a university’s patent policy requires the university to offer as a default an exclusive license. A vice president for research can implement effective technology transfer practice within a crappy patent policy by making it the university’s default practice to waive the university’s policy ownership claim upon disclosure of any invention, and to require that for any invention the university does agree to take in, it will immediately offer the invention non-exclusively, on FRAND terms, and better yet, royalty-free, at least for early adopters, such as anyone taking a license in the first year, or for the first ten companies in. Recover patenting costs from an annual research review, or from an update service, or charge a one-time fee of $5,000 to enable the license to sell, and recover patenting costs from that fee, if more than a few companies ever find reason to sell product within the claims of the patent.
Vice presidents for research could implement these practice changes immediately. They don’t involve changing policy. These practice changes involve changing default practices under existing crappy policy–to establish practices that are directed at effective technology transfer, and (on the side) at way better ways to make money, make friends, and build a reputation for meaningful outcomes from research. Of course, one could change policy, too. But that will take five years, require trustees to be persuaded to undo what they have done, and involve listening to the screeching howls of protest from university licensing folks and the speculative investors that have been attracted to crappy university exclusive licensing practices. Instead, change three practices. Exhaust the policy ownership claim. Institute default immediate non-exclusive licensing. Since whatever IP you take in then is not a matter of the patent policy, negotiate how any licensing income gets dealt with. Now you have the foundation for effective technology transfer.