On making (Seeking Money = Public Service) federal research policy

Here is a paper from 2006 (actually, it looks like a near-final draft) by Fiona Murray on the OncoMouse licensing problem. At the time, the OncoMouse was one of the big issues for university tech transfer offices.  The paper is a great read, showing how the mouse research community, called the “Mouse Club,” organized a patenting opposition to the OncoMouse licensing effort, using patents to undermine the value of the OncoMouse patent and its outrageous licensing requirements and build an alternative social order around patenting.

There have been a string of these licensing problems, now–Harvard’s OncoMouse, WARF’s handling of stem cell inventions, and the disease assay patents licensed by Utah, Minnesota, and Cornell, to name a few.  In these licensing efforts, exclusive licenses from universities have allowed, or encouraged–or even mandated through diligence clauses–that companies attack university research and medical center services in order to generate a profit for the company and the licensing university.   Research and medical center use of university-hosted research developments is seen as infringement, not success; further, the research part is apparently seen as competitive with the host university’s own desire to corner the market on future research funding.

I hesitate to try to reconstruct the reasoning why university administrators would set up such horrendous licensing agendas–but it is clear that they can largely get away with it.  The oversight is lacking when the institution itself has created and defends its own institutional conflict of interest. “Money or public service? Aw, screw it. Go for the money. No, better, make going for the money the definition of public service–by making a lot of money, there’s more money for research, and that will result (eventually) in more products for the public, and more opportunities for us to make money. Yeah, that’s it. Money-making is public service.”  Without oversight and without a moral backbone, the university does become more “like a company”–more like, say, Enron, and less like, say, Whole Foods.  

Rather, let’s rule out some lines of thinking that are defective, and add some lines of thinking that argue against this sort of behavior.   I hesitate to imagine that university administrator thinking on these sorts of issues is rational, or that there really is any thinking.  It seems more like unreflective urges, a kind of shoplifting or random act of vandalism, done because it seems possible to do, so why not?  If actual intention might lead to an outcry of bad behavior, then why intend anything?  Just aspire to “public benefit” (meaning profit), and intend nothing but making money in whatever deal presents.  Perhaps this is what Walter Valdivia means by the “highest bidder” mindset–not the highest bid–there isn’t any–, but the most opportunistic arrangement within the few deals that do show up.  It is not that the licensee is one that offers more–that does not happen–but rather that the university administration chooses to become complicit with the licensee in an attack on the research community, the practice community, and in the licensee’s competitors–even if those “competitors” might also be important research collaborators.

I recall at one major research university, the dean of engineering announcing publicly that there were five designated “strategic partners” for the engineering school.  A competitor of one of these “partners” then set up a major research lab with the information school rather than the engineering school, outraging the dean of engineering.  How dare they!  Any exclusive patent license does the same thing–it sends a message to the industry that the university administration has chosen their dance partner for the next two decades and there’s no point in hanging around to be wing-men and wallflowers.  University administrators who care mostly about a “big hit” don’t care about much of anything else–they would feed their grandmothers to the Ravenous Bugblatter Beast of Traal, like any good Vogon, for the chance at two decades of millions in royalties.  One might call such an attitude corrupt in a public service position, but if public service is defined as making-money, um, well, it seems to go with the territory.

So, let’s rule some things out.  First, Bayh-Dole. Murray’s OncoMouse paper (2006) repeats the same vesting argument that the Supreme Court rejected (2011)–but anyone with half a head could read the Bayh-Dole Act and see that the vesting claim was not supported in the law. Here is what Murray writes:

The 1980 Bayh-Dole Act standardized practice by assigning all patents generated with Federal funding to Universities who were also charged with a duty to license and generally facilitate their translation and commercialization (Mowery et al 2001). (5)

Here is what Mowery et al. actually wrote in 2001:

The Bayh–Dole Patent and Trademark Amendments Act of 1980 provided blanket permission for performers of federally funded research to file for patents on the results of such research and to grant licenses for these patents, including exclusive licenses, to other parties. (102)

These are very different statements.  Where Mowery et al. are careful to describe Bayh-Dole as providing a blanket governmental permission, Murray has “assigning all patents.”  A permission to file does not have anything to do with the right to file.  But this is typical of what has happened in the misrepresentation of Bayh-Dole by the university patent management community.  The misrepresenting mice have gotten into all our foodstock about innovation and research and left droppings everywhere, even in really solid discussions like Murray’s.

Bayh-Dole did not assign patents to universities that hosted research.  Bayh-Dole does not apply to universities directly, for that matter.  Bayh-Dole applies to federal agencies, and requires agencies to approve retention of ownership by universities when they do come to own inventions, and take appropriate minimum actions to use the patent system to promote the use of the inventions.

And here’s the second point of difference in the two statements above.  In Murray’s statement, in the part that I did not boldface, Bayh-Dole is depicted as “charging” universities with “a duty” to commercialize.  In Mowery et al. this is simply part of the permission–that universities that file for patents also may grant licenses, including exclusive licenses.  There is no duty to license or to commercialize.   There is no such duty stated in Bayh-Dole, or in the standard patent rights clause that universities accept in receiving federal funds for faculty research.  Oddly, the “duties” (if that is what we will call them) in the standard patent rights clause have to do with filing patent applications, reporting to the government, favoring small businesses in marketing and licensing, and requiring US manufacture for exclusive licenses pertaining to the US.  There is no duty, even, to license. Nor to create commercial products.  Nor  to make money.  The emphasis throughout is on “practical application” so that the benefits of each invention made with federal support are available to the public at a reasonable cost.

It is not the “inventions” that are to be “transferred” but rather the benefits!  If one transfers the inventions, that’s because doing so is a way to also transfer the benefits.  If one trades merely in patent rights on inventions, that may do less to advance making benefits available than transferring the inventions that the patents are based on.  In fact, licensing the patent rights exclusively to a company may work against practical application and public availability of benefits–the licensee may be willing to pay for patent rights so that others don’t practice or develop the invention.  This happens all the time, in my experience, especially with university startup companies–the company takes an exclusive license (almost always), raises investment funds (in some cases), and then hires technical staff who build a product that may have nothing to do with the licensed patent (about half the time). The patent license is used to raise money, but the underlying invention is not developed, and is at the same time kept from the development by others.

One of the saddest things is to listen to faculty inventors talk about how the university licensed their core work, covered by a fundamental patent, to a startup company that now, a few years later, is not working on the invention, and is not about to give up the license. The university, too, is likely into it for the equity position and would damage the future value of the stock by pulling the license.  The invention is in essence a hostage to a money-making venture involving patent rights and stock prices.  Again, this happens all the time!  It is not a hypothetical.

The primary objective of Bayh-Dole is practical application.  That objective is Congress’s purpose.  The law applies to federal agencies in the hope that its implementation in the form of standard patent rights clauses will result in such practical application, with benefits available to the public.  A university may accomplish practical application without licensing, and without making commercial product, or requiring someone to make commercial product on behalf of the university, or the inventors, or the public.  This is the case, especially, with method patents, but also with any number of improvement patents where any practitioner with reasonable skill can implement the invented work.  3d-printing is a good working example, where the patent rights in various “improvements” are readily practiced without much of any extra investment to “develop” them.  Bayh-Dole does not care how it is done, but rather that it is, and that the patent system plays some role.

There is another place in federal regulations where universities are charged with a “duty.” That is in 2 CFR 215 [2 CFR 200 as of 2014, with some changes in wording], which is the federal funding agreement for most federal grants made to universities.  There, at .37 [.316],–the section following the implementation of the standard patent rights clause (.36(b) [.315(c)])–the government requires, and universities agree, that with regard to any patents acquired or improved with federal funds that the university will:

Real property, equipment, intangible property and debt instruments that are acquired or improved with Federal funds shall be held in trust by the recipient as trustee for the beneficiaries of the project or program under which the property was acquired or improved.

If one works through the definition of intangible property, it clearly includes patents.  If one considers now standard university patent policy claims to ownership of inventions (and associated patents) as a condition of employment, then (accepting for the moment these assertions), then the money used by a university to obtain ownership of a federally funded invention is precisely the federal money allocated for the research!  That is the money that the researchers are expressly paid to do the work. Such patents are acquired with federal funds. The university accepting the federal funds “shall” serve as trustee for the beneficiaries of the project. Unless the university is the beneficiary of a project, its role is not to make money from patent licensing for itself.  It cannot be that university public service = money making for the university.  The federal duty is to be trustee, to look to the well being of the beneficiaries.

No university I know of complies with this duty.  I will go so far as to posit that if a university administrator is presented with this clause, the response will be to dismiss it, or to argue that since it has not been enforced it has no effect.  It will not be a “OMG” moment, initiating a scramble to change licensing and royalty distribution practice.  Deeper, it is nothing that university administrators hold as an internal standard–that is, this duty imposed by the standard federal funding agreement comes not as a confirmation of university practice, but as a challenge to it.  There is a commentary on the mindset of university administrators in this matter, but I will leave it to the reader to construct.

Of course, as far as the OncoMouse was concerned, Bayh-Dole’s standard patent rights clause was not in play, since the Harvard research was funded by DuPont. (Murray represents the research as being also funded by the NIH, but the NIH memorandum with DuPont appears to be based on a finding that the OncoMouse work was funded entirely by DuPont.) Murray documents the various restrictions that DuPont put on the use, development, and transfer of transgenic mice.  What Murray does not get at, however, is the contractual interest that Harvard has in DuPont creating the “greatest value” for itself from holding the patent rights licensed from Harvard.  Again, this may be Valdivia’s meaning of “highest bidder”–not what DuPont was willing to pay Harvard for the rights, but what Harvard was willing to allow DuPont to do to make as much money as possible, to heck with academic “norms” or conventional ideas of public service.

The OncoMouse effort at Harvard stands in contrast to the Cohen-Boyer patents managed by Stanford.  In the Cohen-Boyer situation (again, pre-Bayh-Dole), Stanford consulted with HEW (for the NIH) and the industry broadly to determine whether to file patent applications, and having received positive input from industry, Stanford for a time made its patent prosecution publicly available, so companies could see the claims being sought and the Patent Office’s responses.  Stanford then turned down a request from Genentech for an exclusive license, and the Cohen-Boyer (or, from UC’s perspective, Boyer-Cohen) patents were then licensed non-exclusively for company use at a “low” royalty.   Universities, as far as I know, paid nothing, were not required to license.

Compare with the OncoMouse effort, where (a) the patent emerged as a surprise, (b) Harvard granted an exclusive license to DuPont, (c) DuPont demanded licenses and payments from universities as well as companies, and (d) DuPont “reached through” to claim an interest in what researchers did with the mice they received.  It’s more than just a “fine line,” to use the phrase from Spinal Tap.  In the Stanford case, the purpose was to “tax” the industry; in the Harvard case, the purpose, apparently, was to exploit a dominant position.  This approach had even been suggested to Stanford, as well.  Here’s Niels Reimers on the matter:

By the way, in the setting of the royalty, I got one letter from an  alumnus : “You’ve got a patent; you can dominate everything here. Why are you charging such a low royalty? You know Stanford could use the money. Charge a higher royalty.”

I had to smile because he obviously didn’t understand the full situation.

And here is Niels Reimers, in a 1980 memo, arguing against “going for the money” in an exclusive deal being proposed for hybridomas:

As Stanford is perhaps the first institution to begin filing genetic engineering patent application and to license hybridomas, we … are to a degree “leading the way” for other universities and government policy makers. We need to be very careful in our public and private actions with respect to developments in these fields, including policies of cell line distribution and licensing so that we do not inadvertently, in fact, restrict scientific collaboration and deny or delay development of benefits of this research for the public. We must also guard against “overreaching” in financial and other license terms.

The language here is similar to that of the Bayh-Dole Act’s objectives, especially:

to use the patent system to promote the utilization of inventions arising from federally supported research or development


to ensure that inventions made by nonprofit organizations and small business firms are used in a manner to promote free competition and enterprise without unduly encumbering future research and discovery

Much of this sense of what a university might do, and why, was clearly lost in a matter of a few years as Harvard got into the Oncomouse licensing with DuPont.  To its credit, WARF went so far as to sue its licensee over the scope of the license to WARF’s stem cell inventions, to create room for universities not to be required to take a license in order to practice the licensed inventions.

But exclusive licenses create substantial problems for subsequent research if the patented method or material is needed for research or practice. What is the point of doing academic research on a technology area in which a company has an exclusive position on a broad, fundamental finding?  This is where the academic researchers looking for the “effects” of patents on university researchers are, forgive me, clueless.  Of course academics might still get federal grants to do research, and may not have to pay for a license to practice an invention, but the technology transfer is fundamentally blocked.  The publications are neutered.  There is no way to practice the invention claimed by the fundamental patent outside of a university without a license from the university-owner or its exclusive licensee, and what are the chances of that?

In essence, the exclusive license destroys the purpose of academic publications, and the federal government in continuing to award grants in the area of the patent and exclusive license is throwing money away.  No one can practice what gets discovered, except in the confines of doing more research, which no one ever gets the benefit of, other than academics getting paid to study something.  I suppose it can be viewed as a kind of welfare for exclusive licensees.  That’s sweet, if one is an exclusive licensee or a complicit owner-university. Consider, however, this policy recommendation:  a university should not receive further funding from the government in any research area in which the university grants an exclusive license to a subject invention–that is, an invention made with federal support. That would not require a change to Bayh-Dole.  Any federal agency could implement such a policy directly.  Just disqualify universities who grant exclusive licenses from further funding.  If the exclusive position is necessary to attract private funding for development, then private funding it shall be!

In the case of the Oncomouse, the Mouse Club created their own policy response to Harvard’s exclusive deal with DuPont by filing patents that could be used to devalue the Dupont/Harvard position:

Why then did the mouse men patent? And why did they not feel that their patenting contradicted their outrage? In short, they redefined the meaning of their own patents. Whereas DuPont had exercised its IP rights to derive economic rents from a valuable asset, the mouse men had stripped many of the direct economic implications from the idea of patents.

The response of the scientific community, then, included more patents to defeat the fundamental blocking patent being deployed for “maximum” value.   Others used the fact of a dominant patent as an excuse to refuse to share mice with others.   What they created, however, was not an environment that was conducive to entrepreneurs but rather was antagonistic to all but the largest players in the field, who could afford the litigation, or who could cross-license to gain access, or who could lean on a dean or university president to get what they wanted.

University administrators claim exclusive licenses are necessary to attract private development funds.  This argument is not true in general, though there may be local instances.  What may be more generally true is that an offer of an exclusive license attracts speculators, who have money or are willing to promise money.  If money is what you want, then putting an exclusive license on offer will attract speculators, and speculators in turn will be interested in using the patent system to increase the financial value of their scheme.

It would not be far off to say that federal policy under the Bayh-Dole Act has been for university administrators to take control of major discoveries from faculty and student inventors, and use patent positions with federal agency approval to attract speculative investors and thereby disrupt the activities of researchers throughout the country, who are then prevented from transferring any technology they might create within the same area as the major discovery.  So much for “collaboration” or an “ecosystem” of innovation.  Of course, the Bayh-Dole Act does not set out such a federal policy, but that is the one that is implicitly claimed by university administrators as well as their legal advisors and regional economic development advocates–at least those that insist that Bayh-Dole gives them the right to take faculty inventions and a mandate to “commercialize” these inventions.  Given that the licensing exclusively to existing companies has not worked out all that well, universities now turn to their own startups.  There, the exclusive license to a fundamental discovery serves not only as a block to independent academic development and dissemination, but it also inhibits the formation of local economic clusters around a fundamental discovery.  What is the point of collaborating with a lab that has lost control of its IP?  What is the point of two labs, both committed to incompatible exclusive licenses by their respective universities, collaborating?  It is a double-licensing accident waiting to happen.

Exclusive licenses granted by universities are a problem.  There are times to do it–I’ve written my share of exclusive licenses.  But there are many times not to do it, or to reserve broad rights, such as by limiting the exclusivity to the “sell” and “import” rights, and handling the “make” and “use” rights by means of a non-exclusive position.  Yes, letting many people practice what has been invented can lead to competition–but for a university, this competition is healthy–it is, I would argue, essential to the research enterprise.  If folks are worried about stagnation, then of course they can complain about the possible lack of imagination among university faculty, or the problems of federal funding dictating what should be studied, or explore “new funding methods” like “crowd sourcing” and “grand challenges.”  But one might also look at university-controlled exclusive licensing–and even the effort to secure exclusive licenses (even when that routinely fails)–for what it has done to undermine academic publishing, collaboration, technology transfer, competition, and local economic cluster formation.

Bayh-Dole does not require exclusive licensing of research tools or fundamental discoveries, though it permits it.  Nothing at a university–not even the love of money, requires exclusive licensing.  Nothing in economic development practice indicates that exclusive licensing creates vibrant economies.  Certainly any exclusive licensing at the University of Utah does nothing for Idaho’s economy, unless the exclusive license is with, say, Micron.  Turning from established companies to local startups makes the economic impact of university research less, not greater.  Universities, having run the road of using patents to take financial positions at the expense of their academic norms and public purpose, by redefining public benefit as licensing income, and having now expanded their claims on faculty and student work to demand automatic ownership of anything that might have value within this system of IP right exploitation, have reached a dead end.  It’s a box canyon that is dead in terms of financials, dead in terms of academic practice, dead in terms of federal policy, and morally dead.  It is a zombie apocalypse that does not need to happen.  It is only mildly amusing to imagine university administrators stumbling around moaning “braaaains, braaaains” when “inner life” is also something they need.

The primary purpose of the present “trend” in university startups “as an industry” (or as an administrative process) is to snooker governments and institutional leaders into allocating millions for to university research on a premise of economic salvation–governments, on the claim that startups are such a success that government money will add to the success; institutions, on the premise that by putting in still more money, there will be such a “return on investment” that the university can shift its financial model from soaking students to hyping investors.  Given this state of conflicted institutional positions, damaging practices, and unsupportable rhetoric around innovation, patents, and economic development, there is no hope for Bayh-Dole in its present form.  It has run its course, and produced its harvest.  It is clear now what it has produced.  Instead of universities as trustees and arbiters, they are the self-centered speculators playing any IP game possible for money.  The current one is to start companies and flip them.   Not all, but enough that there’s plenty more than a critical mass.The Brookings Institute argues that the federal government and state governments should help them do this.

There is no going back, really.  But a good start at going forward would be to get universities out of the patent licensing business, except for non-exclusive licensing.  Bayh-Dole could be modified to do this.  A state legislature could do this for its public universities.  Even simple enforcement of 2 CFR 215.37 would be a good start–universities claiming ownership of federally supported inventions as a condition of employment and use of resources (where the employment is entirely federal money, and the use of resources the university has already agreed via the federal contract to provide) must show how they have acted as stewards for the beneficiaries of the funded projects. I would love to see the congressional hearings, calling vice provosts of research to account for their stewardship, when all they can point to, under oath, is that they tried to make better money by holding out for exclusive licenses and a willingness to be complicit with speculators to increase the “value” of their patent rights.


This entry was posted in Agreements, Bayh-Dole, Commons, Technology Transfer. Bookmark the permalink.