University Patent Policy for Effective Technology Transfer, 4: The Eat and Fart Model

No one in their right mind reads a book primarily because it has a copyright. “Gosh, all these books in the public domain–I need to find one with a solid, enforced copyright!” Similarly, technologists with stable brains do not seek out new inventions and discoveries because those have been patented. Other people are attracted to the patents, but not your basic or above average technologist. We don’t even buy products because they have been patented. “Oh, look at this one–it has 50 patents–it must be the one I want–this other one has only 48 patents. Not so good. And eeew–that one is infringing!” We just don’t care. No one with working brains leads technology transfer efforts with a patent or copyright. But that’s just what university licensing offices do as standard practice! Conclusions to follow.

If you consider a university research lab, there are plenty of “assets” floating around, beyond potentially patentable inventions. There are talented people, some with significant reputations and social networks. There are research sponsors and collaborators. There is technical data, and experimental setups, and loads of instrumentation made to do usual and unusual things. There’s software, procedure manuals, articles and manuscripts, demos, instructional materials, observations on progress or lack thereof, ideas about what to do next or to do better or not to do at all. Most of this is not “Intellectual Property” but rather “intangible assets”–non-IP intangible assets, or NIPIA.

We might ask the general question, then: even if a lab has made an invention, what assets should one lead with to effectively transfer technology? Put another way: what assets might the most efficiently form relationships through which technology in the lab might move into use by others? What assets–IP, NIPIA–might one choose from to form those relationships? What does one lead with? And what technology gets transferred?

Here we hit something counterintuitive to most university patent administrators. The technology that ought to transfer is not necessarily what is claimed by a patent. In fact, a patent might be useful in catalyzing the transfer of other technology assets (let’s call them) that are not covered by the patent. This observation runs entirely against now-dominant university technology transfer thinking–that all the other assets in a lab are to be put into service trying to get someone to take a license to the patent, and that brings in the money that then gets shared with the inventors and the department and might in some way indirectly compensate folks for all the other assets–including NIPIA–that were exploited to get that patent license. But it often–nearly always–is the reverse case: people who have the other assets are more likely to also develop an interest in the patented bits, provided the holder of the patent isn’t being an a-hole about it all.

From the patent licensing side of things, however, the answer is nearly unanimous–start with the patent right. Every university depiction of the “technology transfer process” starts with disclosure of a new invention, moves through commercial assessment (by university bureaucrats, not by people actually in commerce), patenting, “marketing,” and then–miracle of diagrammatic rhetoric–a license.

Here’s the University of New Orleans:

TransferProcess.gif (385×298)

Here’s University of Toledo:

Tech Transfer1.jpg (700×473)


Not to be outdone in fantasy graphics, here’s MIT:

inf_lg_base.png (736×736)

Same deal–start with a disclosure of an invention, patent stuff, market to “find or form a licensee” (note the use of the singular) and then licensing leads to “commercialization” (not, say, public benefit–just all the stuff people do who try to make money doing it), and–voila–money (here, called “revenue”). The rhetoric of these depictions of technology transfer insists that the process starts with disclosure of an invention and patenting–one leads with the patent right, one aims to make money, and the money funds research to produce more inventions to be patented. Reduce to basics–research to invention to patent to exclusive license to money, back to research. This is a vision for a self-perpetuating patent bureaucracy. Missing: public benefit as a primary outcome, industry-wide adoption (including standards and interoperability), free competition and enterprise (one of those express Bayh-Dole objectives universities conveniently ignore altogether).

Consistently, it is just one license per invention. The object of the “technology transfer process” as represented by university licensing offices across the country is the exclusive license. The idea, then, is that one leads with the patent–the opportunity to control a patent is the “incentive” offered for anyone to take the license on offer.

These fantasy depictions also misrepresent the situation even for exclusive licensing. The diagrams make it appear that this cycle happens for every invention. Every invention gets its patent. Every patent gets its license. Every license makes money. Money drives research. But it’s not that way at all. Maybe half of disclosed inventions get a patent. What happens to the rest? Of those that get patented, maybe 20% are licensed. What happens to the others? Of those that are licensed, maybe half of those ever result in money greater than the cost to get the deals (patenting costs, licensing office costs). What happens to those others, trapped in exclusive licenses that don’t result even in *money* let alone *public benefit*?  In a realistic depiction of this process of technology transfer, we would see an engine that eats research inventions, farts out most of them, and produces just enough money from a very few of them that it can continue to survive. 95% of inventions are farted out into non-managed states–of every 100 inventions, 50 don’t get patents, 40 get patents but don’t get licensed (and so have been actively suppressed), and 5 of the 10 that are licensed don’t result in products (and so in effect are also actively suppressed, but in a different way–by a licensee rather than by the university directly). This process transfers hardly any of the inventions that it demands to own. It eats and mostly farts. Again–what happens to all those inventions. They are not the ones that remain at the point that revenue funds further research and education.

If anyone wanted to see the metrics for this eat-and-fart exclusive licensing technology transfer process, they should ask for just these metrics–every year, the % of inventions for which US patents have been issued (that’s a four or five year thing from the date of disclosure–and patenting can go on much longer, what with continuations and continuations-in-part); of the inventions under management, the % that have been licensed and remain licensed (don’t count options to license or review-only licenses or licenses that have been terminated). Of those inventions that are licensed, the % that have resulted in commercial products. And for each invention, report the amount spent (spread the cost of licensing office operations across the entire portfolio of inventions–there are ways to do this so overhead for acquiring inventions is different from the overhead of patenting is different from the overhead of licensing. An invention that is acquired, patented, and licensed in generally costs a licensing office much more than an invention that is merely acquired) and the income received, with the net after payments to inventors and to the licensing office–what actually goes to “research and education.” That’s the metrics you can use to assess this eating and farting exclusive patent licensing operation. It’s not a good metrics of technology transfer–it is a metrics of eating and farting. Technology transfer–for all but the few inventions that make it through the intestinal system of this beast–is covert, informal, and even illegal.

And really, there is no actual offer when universities default to exclusive licensing. It is not like a university licensing office sets out the financial terms and development targets up front for each invention in advance and waits for takers. Universities won’t even comply with GASB Statement 51–“Accounting and Financial Reporting for Intangible Assets”–and put a value on each invention that they acquire. Rather, to back its offer of an exclusive license, a university provides–maybe–a generic template license agreement with lots of blanks and waits for someone to ask for more information. This step is called “negotiation” and it is of a sort–where company officials ask for changes in “boilerplate” of the template and university officials explain that they can’t change anything except what’s left open in the blanks, like royalty rate, upfront fees, and the names and addresses of the contacts at the company to notify if the university wants to audit the company’s books. What terms actually are agreed to–if ever–end up being treated as confidential. The public doesn’t get to know what those exclusive deals actually entail.

The standard line, then, is that a university should lead with its patent right if it wants to transfer technology. This is rarely the right choice. Leading with a patent right attracts patent speculators rather than people interested in new technology for being new. Patent speculators–whether investors or entrepreneurs or companies or patent trolls–are attracted to the patents rather than to the technology that the patents cover (and exclude others from making, using, or selling, but for a license). Thus, leading with patent rights tends to draw speculators into the mix rather than technologists–the university, by adopting the eat and fart model, invites patent speculators to compete for access with company technologists and nonprofit technologists who may want access to see what the invention does, or to test it out for their own research and development, or use it to do things that look worth doing. But technologists may not have any big need to mass produce the invention or build commercial products. Technologists are not attracted by patent rights. They have brains.

The technology transfer “process” introduced by the eat and fart model is one that uses patent rights to move technology (at least in the theory) from lab to licensing office and rarely to patent speculators and even more rarely to another set of contract technologists who then are tasked with producing new product to make money for the speculators, who kindly share a bit of their upside with the licensing offices, who use that money to show everyone how important they are or could be if they are given enough money to patent even more stuff and expanded policy to claim eat more stuff that can be farted, mostly.

The great question not addressed by most university patent policies is what happens to all the stuff–the vast majority of research outputs of the university–that don’t make it through the beast to commercial product available to the public on reasonable terms. Inventions not patented ought to be released. But are they? No, not readily, and not without nasty conditions. Inventions patented but not licensed in some given time frame–like three years from date of patent issue, ought to be released. But are they? No, never. A university, once an invention is patented (university spends $10K to $15K–in some cases much, much more for foreign patents), a university doesn’t let go, won’t license non-exclusive, royalty-free. What about patents licensed exclusively but no one develops as a product? Those licenses should be terminated or downgraded to non-exclusive immediately upon seeing that nothing is going to happen. But are they? No, rarely. Even then, how often is it that an invention licensed exclusively to anyone who fails to work the invention going to be taken up by others, years later, when it becomes available. No, doesn’t happen. If the exclusive licensee suspects people are anxious to use the invention, then the exclusive licensee or the university or both turn troll and aim to sue everyone for infringement, if for no other reason than to recover their costs of eating and farting so much.

We can then ask how it is that anyone expects such a process–technologists to licensing office to mostly patent speculators to their contracted technologists–is going to be more effective at technology transfer than, say, technologists to technologists? Even when the inventors want to start a company, the eat and fart model makes the inventors assign to the university, the university gets patents (whether the inventors want patents or not), and the university makes the inventors find someone to run the company and negotiate with the university (because the inventors aren’t allowed to negotiate with the university) to obtain a license to the invention (now saddled with all the university’s requirements that run with any such license). The inventors can’t just assign their invention to their company and get on with it–which would appear to be the much more effective procedure in transferring the technology, yes?

If a university patent policy aimed for effectiveness, then when inventors aimed to start a company, the university would waive its ownership claim, deal with the inventors’ split time with the company, and if the university needed money (why? it’s not paying for the patenting or for claiming ownership or for “marketing” the invention), it could demand a share of what the inventors ever make from their company. But why even that claim? More generally, if the inventors have an idea about how to transfer technology, then what’s the point of the university getting involved at all, unless the inventors ask for assistance with their technology transfer idea? We should not rule out that doing nothing at all with an invention may be a really great idea for effective technology transfer.

Why would any university patent policy intended to encourage effective technology transfer ever allow the default practice to be eat and fart? One would think–thinking here, I know, so outlandish–that among the approaches that a university patent policy would forbid would be eat and fart–along with selling stuff off to patent trolls or the university playing the patent troll itself. But there’s not a single university patent policy that forbids eat and fart. The only one that comes close is University of Waterloo, which does so only because in general it doesn’t claim any patent interest.




This entry was posted in Metrics, Technology Transfer and tagged , , , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.