Let’s start blunt. Then extended discussion. Snark as needed.
Current university IP policies create a background rights problem that drives away collaborators, makes university-based inventions irrelevant, and makes university dealings with IP default to unreliable.
One of the dark problems with a university claiming to own all faculty, student, and staff inventions has to do with what’s called “background rights.” A background right is a claim on anything (IP or otherwise) that is needed to practice under a foreground right. Examples, a university may have two patents, one broad and one narrow operating within the claims of the broad patent–an improvement, an application, an alternative method. One might get a license to the narrow patent–but one would also need a license to the broad patent to have freedom to practice with regard to the licensor. As you can see, it makes no sense to take a license to the narrow patent and not be assured of access to the broad patent.
In a general sense, the reasonable thing to do is to make the broad patent available non-exclusively and license narrow slices of rights in foreground things as one desires (non-exclusive is also good here, but university folks cannot help themselves and default to exclusive). A less reasonable thing to do–and this is what gets done, often–is to license the broad patent and narrow patent together, exclusively, tying up all sorts of applications with a single licensee typically interested in only one or two of many possible applications. Look at a typical therapeutics patent–there can be scores of applications–disease states, methods of application, formulations, and the like. Somehow, university folks claim success if only one tiny bit of their broad patents gets used, even as all the other applications are effectively suppressed by the university’s licensing practice. No one talks about this in public. They don’t see it, or don’t want to see it, or just decline to be candid. It will never show up in a university tech transfer office annual report. Sads.
But this broad and narrow is an easy case. Consider a more typical case. The university has got a patent, but it also claims rights to any number of other research projects that may be working in the same general area as the university’s claimed invention, or working on tools that are used in working with that claimed invention, or take up that claimed invention and improve on it, or adapt it, or apply it to a particular situation (such as taking an engineering invention (image compression, say) and applying it to medical imaging for, say, tumor detection). This second invention can end up becoming a background right to the earlier patented invention–the rights to the second invention, which may not even yet be “under management” by the university, may be necessary to practice an application of the prior invention.
In this case, a typical university licensing practice is to attempt to isolate the subject matter of a license–the only thing licensed is what is expressly identified in the license–such as a technology “case” number, or a US patent or patent application, and perhaps any subsequent continuations or continuations-in-part or divisionals that all claim priorty from the licensed patent. The university’s implicit position–anyone on the other side of the table gets it immediately–is that the licensee is getting only what’s licensed even if the university might also have other IP (or claims to stuff) that would also have to be resolved before a company could practice under the licensed patent. The obvious question that gets asked of the university is “What else have you got that we will need in order to have freedom to practice?” And a typical answer is “We don’t know”–which may well be true in a way, but also smacks of a failure of diligence to know what is in the university’s “portfolio” (another awful term that gets used by university folk–a good sign they don’t understand and are poseurs). University folks pop off and file lots of patent applications, with lots of claims, and how (they ask themselves in the quiet of their minds) are they to inspect all of those claims, thousands of claims, for possible overlaps with what any one licensee might choose to do with some handful of licensed claims in some foreground patent? So rather than be diligent, they use a paper disavowal of diligence that takes the form of an express and limiting definition of the “licensed subject matter.”
Now, defining what’s licensed is a really good thing. How else to know what’s gone out the door (for the licensor) and what one is bringing home (for the licensee)? But stink about it a bit. A license is not really a permission–it is an agreement not to sue to enforce one’s exclusive right. What one hopes to bring home is freedom from being sued by the licensor for any IP or other stuff held, claimed, or owned by the university and needed to practice the foreground, expressly defined and licensed rights.
You see how this problem develops? On the company side, it is keenly obvious. On the university side, folks don’t want to talk about it and are told to exclude everything but what is expressly defined–and then hope to get away with it.
Holding multiple forms of IP (and other stuff that might involve a claim or create a risk of a lawsuit (if not for infringement, then for misappropriation, or trade secret theft, and the like) becomes a problem of royalty stacking. If the narrow patent is licensed for cash, future minimum payments, and a 1% royalty on sales, then (in certain university-side minds) a second patent should be worth at least that much if not more. The second patent’s payments get stacked on top of the first patent. From a licensee’s point of view, this kind of behavior is business-immoral. You show up willing to pay for a license, negotiate in good faith, and then find out that the university has withheld improvements, tools, applications, know-how, technical information–all because it is not within the expressly defined (and expressly limited) foreground rights in your license, which now seems stupid to have signed. You the company show up hoping to come home with freedom to practice only to discover that they university has held out stuff, and now can sue you for practicing within these unlicensed other rights. What a shut down. Worse, given that universities generally default to licensing exclusively, you either have to take an exclusive license to these other rights (as you find out about them) or watch as the university tries to license those rights exclusively to someone else–which is total horror, as then you can’t practice what you’ve licensed and you can’t hope to get that exclusive license (unless you, like, buy the licensee or beg a sublicense from that licensee, or–better, as it often is–you design around what you’ve licensed and cut the darned university free to stew in its bad faith juices).
So you’ve got either a royalty-stack problem, and you find out you are getting pushed to pay double or more than you bargained for, or you find that the university contemplates totally screwing you over (getting their patenting costs out of you plus the upfront fee and not bothering with the royalty on sales which can’t happen without a further university claim threatening to shut down sales plus all sorts of other business relationships) and going off to license background rights to some other company.
The university tends to rationalize their practice by contemplating the problems of double licensing. If they license a foreground right and include even a whiff of a first right to negotiate for even a non-exclusive license, not to mention an option to an exclusive license, to any other rights held by the university and needed to practice under that foreground right, then while that whiff is in effect, the university cannot license those background rights to anyone else. And the university folks cannot decide to license such background rights non-exclusively (even royalty-free). I mean, they cannot do it. They treat each invention on its own, as unrelated to any other invention unless it obviously is, such as a continuation, or CIP, and the like. The inventors may be different. Those inventors expect royalties from their invention and hell if they are going to be the patsies while some other team of university inventors gets the big bucks from an exclusive license that necessitates everything background to be offered for free. The worry about double licensing then is real, and even the double licensing itself is a serious problem. But it is a problem created by the university’s own IP policies–it is an administrative and legal stupidity put out there under the general heading of “technology transfer is hard.” Sort of in the spirit of the Austin Lounge Lizards, “Life is hard but it is harder if you’re dumb.”
There’s another way to look at double licensing. Let’s say that one licenses an invention rather than a patent. This is a huge distinction. An invention can be more than just the parts that are patentable. There can be parts of an invention that are not patentable–anticipated in the prior art, say, or not proper subject matter for patenting. That’s the difference between an invention (generally) and a patentable invention (specifically). To bother you, I will point out that Bayh-Dole is drafted on the foundation of inventions not patents, though it is made part of federal patent law. The operating definition in Bayh-Dole is the “subject” invention–an invention that is or may be patentable. Again, this is huge–in what way is an invention “may by patentable”? Who decides this “may”? And even if patentable, what is the extent of claims that are drafted? Is it good enough to draft just one claim and leave off all the other patentable claims? How is that different from not claiming anything at all–that is, not filing–in which case the Bayh-Dole contractor risks losing title to the *subject invention.* Luckily, Bayh-Dole is so dysfunctional that this sort of thing never comes up–the federal agency folk don’t understand it or wouldn’t care even if they did, and the university folks use it as a fear totem to enforce on inventors the stupidest and least effective of their practices even while knowing the federal agencies won’t enforce anything unless the universities go out of their way to poke federal policy eyes.
If a university makes comprehensive, even automatic claims to own everything, then that’s the surest way to create the prospect of background rights issues and everything that goes with. If a university is highly selective, and owns very little, and grants non-exclusive licenses, perhaps even at no charge–leaving money to sponsor research or join consortia or pay subscription fees for publications and update notices and whatnot, then the background rights problem and all its business-immoral complications essentially goes away. There may be other rights out there a company needs to practice–but these are not the university’s problem.
Invention, patentable invention, claimed invention. These are in general very different things. Contemplate for a moment. What is an unpatentable invention? This not idle. University policies routinely include claims to unpatentable inventions. It’s nonsense. It’s misery. But there it is. Real world situation, drafted and approved by university lawyers. Who checks the work of university lawyers? No, it’s university lawyers all the way down.
If a university licenses an invention, rather than a patent (a claimed invention), then whatever is or becomes part of that invention is implied in the license. The university agrees not to sue for claims to anything that becomes part of the invention. Even the idea of the invention then, in the license, must involve a meeting of the minds. The university cannot have a secret agenda to limit what the licensee thinks it is getting. This implicit nature of an invention license means that anything in any university lab or research project that could (reasonably, say) be included in the invention that’s licensed–to the extent that the university has a claim or right to that anything–*is* licensed. That is, the university agrees not to assert its claim or right to that anything, specified or not, already in existence or yet to be, against the licensee. From a licensee’s perspective, this is absolutely the right way to do it. Just want an agreement not to get sued, and get on with it. But universities, because of their rank stupid IP policies, cannot go this way and still do exclusive licenses. But if they license only “foreground” patents, then they have to cover up all that stuff we’ve just worked through–background rights, royalty stacking, and double licensing. And those are things that get in the way of normal, generally ethical companies, doing business with university licensing offices. Sure, if a company is so big and important to a university that a dean or vice chancellor is not about to let some tech transfer attorney sue them. But we are talking about normal, generally ethical companies. They see university IP practice is business-immoral. Why then do business with a university via their technology transfer office? Design around. Avoid. Make irrelevant.
We have then this complex. Background rights (which may come after the foreground rights!), royalty stacking (multiple IP should mean more and bigger payments than for just the foreground IP), and double licensing (if foreground and background–which we may not know about–are committed to one company, then any commitments to any other company are fraught with risk–which means, gotta set the company licensee up to take the hit by limiting the licensed subject matter. Can’t “throw in” everything else that might be needed to “commercialize” the licensed IP. Can’t license additional patents at no additional charge (so that the. licensed “invention” becomes *both* patents, and the inventors across both patents (or even more than both) all share in the licensing income from the one deal).
Once the university adopts a grasping IP policy–claiming everything and releasing only what it doesn’t want or it shouldn’t have claimed in the first place–then the background rights, royalty-stacking, and double licensing problems kick in. If a university claimed nothing, did not deal in patents, it would not have any of these problems. There would still be problems of this kind to sort through but they would be local to specific university inventors (or investigators, or specific labs, or to specific invention management agents–the smaller and more specialized, the better and easier).
Consider–if a university wants money from IP positions, it can get that money without taking either an ownership position or interfering in licensing, sale, or whatever disposition others decide upon. Just claim (it’s grasping, I know, and not a great idea for other reasons) a share of what the inventors or investigators or whomever receive. If they give away their IP, and get nothing, then the university is happy getting nothing too. If they make a pile, then the university gets a share for *doing nothing*. Think of it, there is *value* to companies and inventors for universities *not* to take ownership or get involved in licensing, not to create vast background rights issues by claiming everything that can be owned as IP and everything that cannot be owned as IP (even if some dunderheaded lawyer thinks this sounds really good because it covers all cases). If it’s money a university desires, then taking a share of what its inventors receive is better than claiming everything.
Better still than claiming a share of what inventors receive is not claiming anything at all, but rather acting as a rainmaker–creating conditions under which inventors and investigators and whomever can move their ideas, inventions, creations, collections, compositions, and whatnot to new venues where they like what happens. Historically, when university-affiliated people have made big money from deals they have done themselves, they tend to gift more to their universities than the universities with their grasping policy claims get. They might wait to make their gift, but they seem to get around to it. Just saying.
And if a university were to not worry the money-as-income-from-technology at all, I argue that that is the first step toward creating a rainmaking environment for actual business and social relevance of whatever ideas its community comes up with. A university might offer grants for inventors to take out patents, and not ask for anything back, not even reimbursement. Universities do this already, all of the time, thousands of times a year, when they file patent applications without a pig’s hope in a diner of getting anything back from licensing. It’s just that universities rationalize this practice by claiming that they own the IP and somehow must obtain patents and so exclude everyone else, including the inventors, from practicing each invention until they have negotiated an exclusive license and especially have paid. It’s not technology transfer that’s off to a good start. It’s actually quite bad. Very bad. Sads bad.
Even skip the focus on patenting. I’ve come into projects where the university is spending tens of thousands of dollars patenting and what the project needs is to finish an essential piece of code for ten thousand dollars that grant funding won’t cover. But most university licensing offices won’t pay for that work to get done, and won’t set up their relationship strategy (gifting, membership, subscription, consortium, open source, crowd source, whatever) to help generate the money for that work to get done. Indeed, university licensing offices are focused on moving all relationship income through licenses. Somehow, the number of licenses (the number of ownership claims) and the amount of money received via licenses (ignoring, say, standards licensing or open source licensing or licensing done by others not the university at all) is somehow a marker of “success” or “excellence.” But it is not.
Once I dealt briefly with the vice chancellor for research at a major research university. She agreed that the approach her university had taken to IP was unproductive. A good sign. But then she followed that with (I paraphrase) “we won’t change policy until we see something better.” Sort of like an alcoholic saying, “I get it too much drinking is bad, but I won’t quit drinking what I’m drinking now until you show me something better to drink,” My response that *doing nothing* was a better IP policy for technology transfer than what they had–but that meant not drinking at all, which as it ended up was not an acceptable “something better.”
Doing nothing–not claiming IP, not taking ownership of IP, not attempting to license IP (certainly not licensing exclusively), not taking money for licenses–these are not things that get in the way of technology transfer–these are things that enable it. And these are things, in their not-ness, that also create opportunities for folks to make money, including even money to pay for university research administrators if not also for university research if not also for–er–instruction. (When was the last time–ever?–that you heard that a university devoted its share of licensing income to reduce the cost of tuition–ha! never, right?)
(I will add here, just to address a common assertion, that Bayh-Dole does not require universities to claim ownership of anything. Indeed, if a university does not claim ownership of an invention arising in federally supported work, then that invention is not a subject invention and does not fall within the scope of Bayh-Dole. This is another instance of how university administrators create their own Pigpen duststorm of complication. See Stanford v Roche, say. And yes, a university could make its inventors agree to become parties to federal funding agreements (as if they were each small companies, sole proprietorships, as it were) and then those inventors would be contractors and when they owned their inventions those inventions would be subject inventions but then the issues in Bayh-Dole would be between them and the government and the university would have nothing to do with it. And in that inventor-government relationship, inventors get the benefit of a patent rights clause that is more favorable to them than either the small business patent rights clause or especially the nonprofit patent rights clause, which is the worst of all. And in that inventor rights clause, inventors are not forced to use the patent system. See 37 CFR 401.9 soon, before NIST edits it into oblivion.
Some complicated things are administratively created complicated things. The administrators that create such complications are not going to be the ones to ever admit to it. Some complicated things get so wound up that folks become habituated to the complexity. They find their careers depend on the complications. It’s like byzantine bureaucracy as a virtue–and then try to “benefit the public” using all this complexity. Background rights, royalty stacking, and double licensing are real issues in IP, but their are especially hard when administrators are dumb beyond recognizing they are dumb. Of course, I’m being blunt, but I’m not trashing folks–I’m just putting it out there, yes, with snark–because this sort of dumb ought to be mocked–it is not dumb of folks without the capacity but rather dumb of folks who are clever enough but instead choose to look away.
University licensing operations are grossly ineffective. Universities have got over 50,000 US patents since 1981, most unlicensed (maybe a good thing in its way, but generally it means stuff sat behind pay walls until it was useless). Most universities operate licensing offices on a de facto business model of one lucrative exclusive patent license floats the operation for twenty years–if they are lucky. In those twenty years, a big university might receive four or five thousand invention reports. They paw through these trying to find one more big hit deal for the next twenty years. The rest might have to go rot for all they care (there’s the meaning of “portfolio” in this context for you). They don’t even recognize even how they got their first big hits–generally not by claiming everything and pawing through it. They believe (in practice) that making a comprehensive claim to everything and then making a system out of licensing efforts they will create more of what happened to them before they claimed everything and had no particular system.
Doing nothing is really good institutional practice. It may not be best practice, but it is a lot closer than what we see now in most US universities. It’s said that in software code development projects the programmers know a project is dead about a year and half before their managers do. How long will it take for university administrators to realize that their approach–the approach wheeled in along with the awful thing called Bayh-Dole–is dead? Way, way longer than a year and a half, obviously.