This article starts here:
In most cases, a principal investigator will know immediately whether an invention or discovery is within scope of a well drafted research agreement–is this invention something that was proposed, as the solution to a problem, say, or that might arise as a result of an investigation? Did the research propose to build or demonstrate or test something new that might have utility for the grantor? Why would a principal investigator hold out on a sponsor of research? Would such “holding out” constitute research misconduct? If so, how could university administrators properly review the situation to ensure compliance if “holding out” on a sponsor meant that the university might be in line for profits from patent licensing via an invention management agent? The review of inventions by deans and business officers only works if the university has no interest in the outcome but for compliance.
Of course, with federal funding, this entire calculus is switched around. If the government receives inventions to ensure that the patent system is used to make inventions (and rights to inventions) broadly available to all, then “holding out” on the government amounts to finding a way to prevent the government from making inventions broadly available (without charge, without playing favorites, without upsetting the competition for additional research and development funding). An inventor might “hold out” and claim an invention, then, against the interests of a federal agency because the inventor wants to use the patent system in some other way–to prevent all use, or to make money from any use, or–perhaps–to make the invention broadly available, but using some special method that’s different than the government’s methods and so does for the government more and faster and better than what the government hopes to accomplish using its methods.
The IPA then switches the calculus of this last –perhaps– into the primary position. The government, so the IPA proposes, wants private invention management agents to step in and take assignment of inventions made with federal support from inventors and use the patent system better than might the federal government to do this –perhaps– thing, to make inventions broadly available using special private methods better than the government’s own methods. In this view, if an inventor “holds out,” the inventor is now “holding out” against the assumed better use of the patent system entrusted by the federal government by federal contract operating outside the actual funding agreement to private invention management agents.
A university administration adopting this rationale, then, argues that aggressively asserting an interest in patenting most anything is an expression of the university’s commitment to assist the federal government in making inventions broadly available. The university’s methods are better than the government’s methods (and hence the continued repeating even now of the “28,000 government patents” nonsense and the unquestioned thought that Bayh-Dole has been wildly successful–these are the moralizing bedtime stories to help them to restful sleep despite their deeds–universities now hold over 120,000 US patents, over 50,000 of which are federally supported, and the vast majority of these patents are unlicensed and if licensed the inventions are unworked, and even where an invention does result in a commercial product, often 99% of the claimed invention remains unworked and unavailable even for research outside the university).
The university’s aspiration to financial reward is thus aligned with invention ownership. The patent system is to be used to “call forth risk capital” to develop federally supported inventions faster and better than could the government dedicating inventions to the public. The public, so this idea goes, will get a shinier, better invention faster and at lower cost, if private risk capital comes forward to do the work on the public’s behalf. Once the new product has been made, then the sources of the risk capital have an opportunity to recover their expenditures in the public interest, and make a “reasonable royalty” as an “incentive” to provide the risk capital in the first place.
There are two forms, then, of risk capital that come into play. The first form is that of “commercialization”–the activities by which an invention is turned into a product fit for use, with public benefits. Much attention has been paid to the costs of such efforts–almost any university-side discussion of inventions and patents involved a gesture to how much more it cost to develop products than to do the research that made inventions.
There are, of course, huge disconnects. It does not even take research to make inventions–inventions get made all sorts of ways, not only in sponsored research settings involving proposed projects. Epiphanies, accidents, messing around, designing, following goofball predictions, creating works of art or music. And inventions that get made do not have to first become commercial products in order to be widely used with public benefits. Many methods, for instance, merely have to become known to be available for practice by others.
Commercialization, if it ever needs to happen, need not happen first but might rather come later, after use is well established and new users prefer to have much of the work pre-done for them. And even if commercialization might happen in parallel with other uses (such as research uses, or custom uses internal to capable organizations–not offering anything for sale), there’s nothing at all that requires that patents should be used to support this commercialization by blocking all other uses.
These disconnects, however, come into play. To induce private risk capital to develop inventions for public use and benefit, recover expenditures, and have a reasonable return, one might use the patent system to provide a degree of exclusivity–not against all research uses or even all DIY uses, but so that sources of risk capital can recover their investments when they do step forward when no one is willing or able to develop an invention in an open environment. Even then, their commitment is to bring something into existence for the public, recover a reasonable return, and then step away from the monopoly and allow “free competition and enterprise.” That’s the social theory, anyway. It’s rather of the form that we would now call “social ventures”–efforts to create something of public value without the requirement that the effort should also maximize profits for the owners or shareholders of the venture. Other than the problems such an idea presents for the public investment corporation, it would appear that such social ventures, even with for-profit profiles, are entirely possible and not outlandish fiction.
Indeed,the idea that university-affiliated patent agents might do a better job than the federal government in just this thing is the fundamental premise of the IPA program. It’s the reason for the public covenant that runs with patents on subject inventions–that the patent system is to be used in particular ways, and not in other ways that are otherwise legal but not appropriate to the purpose (such as suppressing all use or licensing exclusively simply to maximize profit at public expense). It’s the reason for all the apparatus for reporting on invention use, for limiting exclusivity, for march-in procedures.
The second source of private risk capital does not get much attention: the cost of reviewing invention reports and filing patent applications. This is the “risk capital” expended by invention management agents. This is also the “risk capital” that sets the institutional conflict of interest in motion. If a university (or its designated agent) decides to file a patent application, then it will expend money on that effort–these days, the cost can be upwards of $15,000 (though the work can often be done for about half that cost, if done attentively). This “risk capital” then must be “recovered” from patent licensing. The expenditure of money on patenting is the primary argument against royalty-free licensing. If university patents were licensed royalty-free, then the expenditure to obtain the patent monopoly would be “wasted.” The patent and licensing would merely be in the public interest–publishing the invention in the patent literature to promote the progress of the useful arts, and making the claimed invention available to all that would use it.
Thus, the point of spending money on patenting is to make money back on the licensing. In that effort, one can recover the patenting expenditures as a share of income from each license or one can bill the licensees for the patenting expenditures in addition to any earned royalty from the use of the licensed invention. University patent licensing practice is almost entirely built around billing for patenting costs. And in doing so, university patent brokers set up the rationale for exclusive licenses. An exclusive licensee, if put in the position as if the patent had been issued directly to the licensee, should be willing to pay as well the full cost of obtaining the patent. Such exclusive licenses–granting all substantial rights in an invention–are in deed assignments. If an exclusive license agreement provides for the “reimbursement” of all of a university’s patenting expenditures, then the exclusive license is in essence a sale of the patent–all substantial rights are granted to the licensee and the licensee pays the legal bill as if the licensee had filed the patent application itself.
In this way, university patent administrators talk themselves into the idea that their best hope for recovering their patenting costs is to get a company to pay for those costs. Back in 1990 or so, when I started in university technology transfer, the old way was slipping away. In the old way, a university sent out a description of the invention in a “non-confidential summary” before filing any patent application. If one (or, rarely more) companies wanted a patent to be filed and were willing to pay the costs, then–and only then–did the university file the patent application. That is, universities (the ones that did not have a big-hit patent license that had given them a reserve budget to spend on new patent work–that is, nearly all of them) didn’t make a decision whether to file a patent application until they had a company willing to take a license. Again, all this led toward exclusive licensing, because the university sought recovery of patenting costs up front, rather than from earned royalties, which would come later–often many years later. Even though 15 non-exclusive licenses for $1,000 each would cover patent costs, university patent administrators were (and mostly are) unwilling to file a patent application thinking that there might be 15 companies willing to acquire a simple non-exclusive license. They’d rather have one exclusive licensee.
If one adopts the idea that patent-induced “risk capital” for commercialization is essential to the public use of university research findings, then it’s easy to see how a university patent administrator might move from exclusive license for patent reimbursement to exclusive license as the best way to gain a “return” on the “investment” in obtaining the patent and by extension the “investment” in the research that led to the invention. Thus, rather than seek to keep the costs of commercialized inventions low for broad public access and benefit, university patent administrators have gone the other way and argued that the purpose of a patent monopoly is to generate maximum value in any way that’s legal (that is, in any way to which no one with power objects), and therefore the purpose of the licensing agreement is to preserve the monopoly power of a patent while requiring that the university licensor share in the “upside” of maximum pricing preserved for the licensee in the exclusive license.
Public interest, rather than being aligned with broad access to the invention (for research use, for DIY use, for competitive uses) and with low costs–costs below what the market would otherwise bear, costs below what a monopoly position might command–instead was aligned with a share of the maximum that a licensee might make. What’s good for the monopolist is good for the university, and what’s good for the university is good for the public. In this way, the moral compass of university administrators was made to point, reliably, always at the university’s own navel. The apparatus of the IPA program, and later Bayh-Dole, had the apparent role of keeping the moral compass of university patent brokers pointed toward something other than institutional self-interest. But that apparatus was designed to fail in both the IPA program and in Bayh-Dole–and to that extent, Bayh-Dole has achieved the purposes designed into it.
It is something to find these implications in practice designed into the new 1969 Wisconsin patent policy. But it’s clear that the Wisconsin policy, by combining the requirements of the IPA program with a review of all inventions by administrators lays the foundations for making it appear that to comply with federal regulations, inventors must assign their inventions to the university (or to WARF) whenever administrators decide they must–even while the IPA program (and later, Bayh-Dole) does not require administrators to take ownership of any invention made with federal support.
Once university administrators get the idea that they are making a decision about ownership in the interests of the university rather than in the interest of compliance with a bargain between inventors and research sponsors, then it’s an easy step to argue that patents should be managed for their maximum financial value, and thus licensed exclusively, and thus licensed for their monopoly value rather than for the public benefit that might arise from access to the underlying invention. “Commercialization” becomes the term used to mean “denying public access in favor of receiving payments from a company that derives value from a monopoly position.” Most university commercialization deals don’t result in commercial products. Of those that do result in commercial products, it’s an open question whether those products are made available on “reasonable” terms. But “reasonable” is a technical detail in the IPA/Bayh-Dole apparatus that few university patent administrators worry over–and no federal agencies appear ready to step in to enforce or counter by using their government license to practice and have practiced (quite apart from march-in procedures).
But there’s not a compelling argument that university administrators must review all inventions to determine whether any given invention must be owned by the university or the sponsor. That issue can be addressed in the reporting requirements between the university investigators and sponsors, and by making the investigators parties to the funding agreement so that the sponsors and the university understand that the obligation to report inventions is with the investigators. It’s just that Wisconsin did not choose to develop its policy in this direction.
In a research procurement environment, a research sponsor seeks deliverables that have utility–application to the areas of the sponsor’s interests. Any research agreement written with any competence will specify what the sponsor desires by way of reports and what the sponsor recognizes as deliverables within those reports. There’s no need for deans and business officers and university vice presidents to scrutinize invention reports–just send the reports to the sponsors and see what the sponsor says. The university’s review, in an environment without institutional conflict of interest, is to determine that its investigators are reporting fully and not holding back inventions from the sponsor and patenting them on the sly. But sly patenting is something that will come out when the patent issues. And the university’s review may enter into it if a sponsor claims an interest in inventions that the university investigators argue was not within the scope of the sponsored research agreement, not a deliverable, not bargained for. In each of these two cases, the university retains a concern for compliance with the terms of the research contract–but the outcome has to do only with institutional compliance, not with the outright ownership by the university of inventions that figure in the determination.
The IPA, then, allowed university administrators to conflate an interest in compliance with their own interest in university (or WARF) ownership of inventions that may result from that compliance. This conflation comes about because the IPA makes it appear that the university obtains the federal agency’s interest in compliance with the federal funding agreement. The university, as far as patent rights goes, appears to act “on behalf of” or “in the place of” or “as an agent of” the federal government for purposes of patent rights. The university, in this view, is assigned a portion of the federal contract, the part pertaining to inventions, and so, in some way, becomes the interested sponsor of the research in place of the federal government, as if, for inventions, the inventors worked for the university and not for the federal government, even while the federal government supplies the money for the work–both direct and indirect costs of the university gets covered and faculty, to participate in the research, are released from their official university duties (so that the university has even less claim on their work that it would have otherwise, as a matter of employment or right to direct the work or expectation of benefit from the work or contribution of university resources for which equitable ownership of inventions might be indicated–none of this).