From time to time that it really does happen that someone invents, gets patents, raises investment, and builds a product that gets sold at market. We say, this event is an instance of a class, or a metonymy for the class. The aspiration is, this happens often. The aspiration is, by our efforts we make this happen, and make it happen more often than it would otherwise, and in better ways, and in good ways. The aspiration is, how could you dare poke holes in our aspirations? How will we get anything *done* if these aspirations are not held up as the exemplars for our work?
Beating up on tech transfer office aspirational statements is like threatening to drown kittens. We will have to be careful.
We can restate the basic question as: is the class robust? That is, does invention to product happen this way all the time, or once in a blue moon, or once a decade per $2 billion of research funding? Further, if we want commercial innovation to happen this way, what are the methods to use, and what of these methods involve universities and their research personnel?
Let’s say someone believes the class is robust, and that invention to product to money should be happening all the time.
Well, it isn’t. The numbers indicate that invention to product to money is somewhere between 1 in 100 and 1 in 1000 research inventions.
One might say then, it *could* happen if we just got better at it. That’s the first line of critique. People believe the class is robust, the university technology transfer activity does not reflect the potential. Get better people, change the model, change the law, rewrite policy, always more money. And on the university side, defensiveness, dismay, and a double down public spin to show how worthy it all is, to show that orderly, process diagram inspired aspirations are a good proxy for reality.
The second line of critique goes after efficiencies. This is classic but dull management. There is a glimmer in the dullness, and that is that waste isn’t generally a good thing. It’s just that dull management can’t tell waste from quality. Thus we get a debate about whether there would be more success stories in the robust class if only universities could, say, skip the assignment step and simply get title to inventions outright. Oh, yes, it must be that assignment step that is preventing the robust class of successes to fill out.
With such efficiencies, could one go from once a decade per $2 billion of research funding to say once a year per $200m of research funding? That’s an interesting question. I doubt the answer is that it is a matter of efficiencies. Again, dull, distracted management cannot tell the difference between a slumlord efficiency—do as little as possible to have a desired cash flow—and a growth efficiency—recognize and act timely on opportunities to create new things of value and benefit from doing so. Are universities slumlord-efficient or growth-efficient when it comes to research-catalyzed innovation? How about invention-catalyzed “commercialization”?
Turning a technology transfer office into a little factory, for all the trappings of efficiency and management orderliness, is at its heart a slumlord version of efficiency in dealing with innovation. It takes the license as a commodity output, takes title to inventions as its commodity input, and designs efficient cranks to turn inventions to patents, patents to licenses, and licenses to products producing money = public good. It will call the Carolina Express license a grail because it is efficient not to have to think about an exclusive relationship to do something innovative.
If one takes on the idea of making a central university IP office efficient, things get murkier. Does it make any sense? Is it simply obvious that every research invention worth anything will do better running through an administrative office run by the inventor’s employer than by the work of anyone else in the world? It would be a good debate, but it would be between a slumlord seeking rents and anyone thinking that liberty, for all its inefficiencies, is going to be better for community. At its most pathetic, the debate will feature sincere folks who truly believe that controlling all research inventions by a central university authority is good for the public and good for the university. It is not that sincerity is pathetic, but that the aim is to replace a reasoned discussion of practice with sincerity and aspirations.
It may well be that in some circumstances, a university office may be better at handling IP than anyone else around. And it even may be that making that function compulsory has good outcomes. And it even may be that having one license agreement as a template for everything is keen. In the local, I can deal with that. In the general, however, I cannot believe it. If it were true, how utterly depressing that the one thing–innovation–that would seem to be so often unexpected and challenging and happy and personal and full of betterness would turn out to be a commodity of a factory-like office?
I’m not going to take on either critique. It’s pretty clear that after 30 years, university tech transfer isn’t producing as it claims. It’s also clear that being more efficient with more and more policy does not appear to indicate better outcomes. My take is, the whole desire for commercialization by grabbing research inventions in the form of patents gets at only a narrow part of the innovation ecology that research contributes to. University tech transfer offices better their numbers by being more selective not by grabbing more things out of fear they must not be getting the juicy stuff or out of fear that they cannot tell what’s any good, or out of fear that they have to follow their own rules no matter how defective or inadequate, or out of fear that showing any advantage to any one invention would be, like, playing favorites.
Community engagement with research is way, way bigger and richer than any patent accumulation operation to license for money. Set that little service up on the side, make it voluntary, stick it outside the university, like in the old days. Don’t try to improve it or extend it. Just make it serviceable. Now explore the rest of the space and ask how to do things so that such an office *isn’t needed or even chosen* most of the time. That’s the best way to make a research patent accumulation service *as good as it gets*.
My answer to the question, is the class robust? is: no, it isn’t. But it could have more in it than universities are going to get with their present practices; and, hey, there appear to be more classes of success here than folks are spending time on, so why don’t we explore those, too?