Fixing the Flaw in Bayh-Dole with Freedom to Innovate Legislation

The essence of the Bayh-Dole Act is that government, though it supports faculty-led university research, should defer to investigators and inventors who wish to develop the inventions they make. Bayh-Dole does this by pre-approving a broad set of arrangements that inventors might make to promote the use of their inventions, subject to the condition that [if they assign their inventions or file patent applications, that] they use the patent system and [the law makes it easy if they choose to that they] work with an organization that has as a primary function the management of inventions, or with the university that hosted their work, even if the university doesn’t have anyone who manages inventions.

In this way, the Bayh-Dole Act is an insightful resolution to the debate over whether the federal government in supporting faculty research is contracting to own important research results, such as inventions, or is simply making donations and doesn’t care at all what happens with results. What is especially important is that Bayh-Dole frees up funding agencies so that they do not have to claim everything and review on a case-by-case basis every single proposal by inventors and their invention management associates to develop their inventions. As long as the invention management agent is pre-approved, the government does not have to demand a show of capability, a detailed plan of action, or a set of performance metrics by which progress can be judged. It is a decidedly non-bean-counting approach. Or, put another way, it is an approach that respects the judgment and motivations of university investigators.

Unfortunately, there is a huge flaw in Bayh-Dole, though it could not have been obvious at the time the law was drafted and passed.  That flaw is that the federal government deferring, with respect, to university faculty investigators and inventors left those very individuals exposed to predatory practices by university administrators who began to see in their futures money bags and empire-building from licensing patents.

One of the central arguments used to justify Bayh-Dole was the apparent ineffectiveness of government agencies to deploy inventions once they claimed ownership of them. The argument went something like this–[in 1976 Howard Forman, a federal patent attorney who had worked on the Kennedy and Nixon patent policies, asserted in Congressional testimony] that the government had amassed some 26,000 patents and was licensing only 4 or 5% of them for commercial development. By contrast, university officials testified that their university-affiliated invention management programs were licensing more like 30% of the inventions they chose to manage. Now, I’m not at all comfortable with this argument for a number of reasons, such as that many of the government-held patents were for weapons systems and it makes good sense for the government to hold these patents and not to encourage a private market to develop in their manufacture and trade. It also is worth observing that in the general patent population, only 4 or 5% of issued patents see the commercial light of day in the form of products.  The government was doing nominally what any patent holder could expect to do.

Despite the rhetoric, the issue wasn’t the poor patent licensing performance of the government–its performance was actually pretty good and the role of non-exclusive access to government-held patents promoted competition for bidding for federal work and helped to establish industry-wide standards, laying the foundation for new markets. The last thing one might want to see in such a situation is a demand that each new invention be licensed off exclusively to one company or another, fragmenting interests. No, worse is that a lot of the companies are startups that fail and their assets get acquired by speculators, bullies (call them orcs), and trolls (“SOT”s). A failed fragmented market for inventions becomes an anarchy that only SOTs could love. That is, a failed market for inventions can become a speculative, derivative market for trade in patent rights–the patent becomes an asset for money-making, independent of the state of development of the invention on which it is based. If you intend to make your money trading in the value of patents, then SOTs may well become your best friends, as it is such a volatile market environment that they seek.

What, then, led to a third of university-affiliated inventions getting licensed for development in the 1960s and 70s? This question is at the heart of the debate over what Bayh-Dole does or should have done and is also central to the foundations of what has become “university technology transfer.”  One answer that has been proposed is that universities were just plain better than the federal government because they could license exclusively and the government preferred non-exclusive. Another is that university-affiliated agents had a better sense of the commercial world and investors. A third, perhaps, was that university-affiliated agents had a profit motivation that the federal agencies lacked. Perhaps there is something to these answers, though I find them unpersuasive, and certainly I haven’t found anything approaching documentation that proves them out.

For each of these answers, as well, there are qualifying observations. Exclusive licenses, for instance, add contracting apparatus, put all one’s collaborations into one pot, and tend to restrict the licensee’s ability to deploy an invention as it wishes. With an exclusive license, invariably, the licensee has to create a product (not, say, a standard) and has to sell that product and pay a royalty. Diligence in this context means, “sell a product and pay up.”  There is no room for cross-licensing, standards formation, or offering a free public license. In drafting an exclusive license, one aims to put in provisions to prevent such behaviors–or to require payment for them, which amounts to the same thing. Exclusive licenses, then, add contracting overhead, restrict access, make diligence demands, and prevent a wide range of possible activities in preference for payment. As the sale of new products falls away and payment is the only concern, development of invention gives way to speculation on patent rights.

As for awareness of the commercial world and investors, there is nothing magic about universities and their agents. One might even wonder at the audacity of the claim. But in any event, any federal agency could hire or contract for similar expertise. No university monopoly there, and certainly not in the 60s and 70s. It may be, as well, that there was a profit motive, but for the most part this was not at the universities. Research Corporation was the major player in the area, with contracts for invention management with many universities. The Research Corporation model–now over 100 years old–was that faculty inventors would assign some or all of their invention rights to Research Corporation, and as Research Corporation licensed their inventions to industry, after costs and providing a share to the inventors, the remainder would go into a general fund to support research throughout the country. Originally, the Smithsonian was to be the primary manager of Research Corporation net patent royalties. Later Research Corporation handed out grants itself. My father, as a graduate student at Washington State University, was the beneficiary of one of Research Corporation’s grants, and while that grant was not a lot of money my father, now in his 80s, still speaks of how important it was to have such funding available.

At the University of California, which set up its own office in the 1960s, the issue was not profits, but an effort encouraged by the faculty, to get more revenue for research. The angle was, use the Research Corporation model, but skip the outside apparatus and focus all the funds after costs and royalty payments to research at the University, along the lines of the Wisconsin Alumni Research Foundation. The profit motive in these programs was a research motive–and it made sense where funding was scarce and royalty income from even a handful of inventions could represent a substantial increase in funds available to faculty and students.  At the University of Washington, which did not set up an internal technology transfer office until after Bayh-Dole, and did not convert it into an active licensing organization until the 1990s, royalty income received from Research Corporation or other agents (UW at one time listed four in its policy documents) went to a fund in the Graduate School, to support faculty and graduate student research. As a graduate student, I benefited from that fund, working on computer representations of medieval texts.

If there was a profit motive to license federally supported inventions, it was an odd sort of profit motive, altruistic in its way, so that income from one’s invention, as it was used by industry, went to support others doing research. Oddly, the federal government got its income from taxes and the like, and did not need to try to make money in order to have money to support faculty research. Indeed when the federal government decided to broaden its support for university research with the formation of the National Science Foundation, its support rapidly swamped out existing foundation and industry support and came to dominate university research activity, so much so that President Eisenhower in his farewell address could warn that government funding was having an adverse effect on independent research and invention:

Today, the solitary inventor, tinkering in his shop, has been overshadowed by task forces of scientists in laboratories and testing fields. In the same fashion, the free university, historically the fountainhead of free ideas and scientific discovery, has experienced a revolution in the conduct of research. Partly because of the huge costs involved, a government contract becomes virtually a substitute for intellectual curiosity. For every old blackboard there are now hundreds of new electronic computers.

The prospect of domination of the nation’s scholars by Federal employment, project allocations, and the power of money is ever present – and is gravely to be regarded.

Thus, I’m not taken with these possible answers for why the university-affiliated invention management programs were doing so well. I would like to propose another reason: that these university programs that were doing so well were self-selected by faculty inventors, worked responsive to the inventors’ interests and requirements, and took on only what they felt they could reasonably place with industry. In short, these programs were successful because, for the most part, they were voluntary, matched to inventor interests, and highly selective. I will leave it as an exercise for the skeptical reader to argue that this could not be the explanation.  It is, however, pretty clear that many programs did have these properties, and government programs did not, and university programs today do not.

All this leads to the flaw in Bayh-Dole. While the Act allowed investigators and inventors to decide if they wanted to develop a particular invention, it also did not protect faculty from their own institutions, which have become increasingly, shall I say, predatory on faculty inventions. The rise of university technology transfer activity has been followed by increasingly compulsory, comprehensive claims on faculty scholarship. In a great irony, university administrators have adopted many of the same practices of the federal agencies, practices that were so roundly criticized in the run up to Bayh-Dole. As it stands now, most universities in the United States have a policy that demands ownership of all inventions and go one step further and often define “invention” to mean most anything for which an ownership claim can be made and which might be sold off for institutional profit. The “present assignment” movement is an advanced symptom of this impulse to take ownership of faculty scholarship where there is some prospect of a license for profit.

In short, present university practice is to be non-selective, compulsory, and institutional about faculty inventions, including inventions made with federal support. There is no good rationale for doing so in the history of the Bayh-Dole Act.  Everything points away from such a rationale. Yet the advocates of what I call the “faux” Bayh-Dole Act claim it was just this system of efficient title-taking that they intended all along with Bayh-Dole. The arguments about government limitations was just political cover at an opportune moment. When the university that implements a non-selective, compulsory, institutional claim on faculty scholarship is a public university, it merely substitutes one form of government claim–federal–for another form of government claim–state. It is as if the purpose of Bayh-Dole was to argue that states should have priority in disrespecting faculty scholarship and academic freedom over that of the federal government.

It is an odd dispute, put that way. Who is more effective at disrupting faculty independence, prerogative, and initiative, the federal bureaucrat or the state bureaucrat? But that is not what Bayh-Dole was and is about, despite the claims made by advocates of the existing state of affairs in university technology transfer. Bayh-Dole was and is about respect for faculty inventors, pre-approval of their efforts to develop their inventions, and a release for government agencies not to have to review case-by-case each petition for action or to take on the costs and distractions of filing patent applications on everything for which the agency could otherwise claim ownership. If anything Bayh-Dole exhibits a distrust for university claims, reserving the greatest number of requirements for universities and other non-profits, putting fewer conditions on small businesses, and putting the fewest conditions of all on faculty inventors when the agency approves their request to retain ownership of their inventions. It is altogether odd that this distrust of universities should be turned by advocates of non-selective, compulsory, comprehensive systems to be, somehow, a mandate for such an approach. It is just the opposite. It is harder to be more wrong. I suppose that’s a testimony to the moxie some folks have.

What should have happened, and what still must happen, is that states need to implement procedures for Bayh-Dole at the state level. That is, just as the federal government passed laws to limit the claims of federal agencies when they chose to support faculty research, and to relieve those agencies of having to make comprehensive, compulsory, non-selective claims, and further to relieve those agencies of the expense of managing claimed inventions, so also state governments must pass their own mini-version of Bayh-Dole, to respect the independence of faculty investigators and inventors, and to protect state instruments (such as public universities) from the expense, the distraction, and the morass of claiming everything in order to sort through it for whatever might be an appropriate match for state interests.

The appropriate thing, now, is to get state government out of the same mire that the federal government found itself in by falling into a non-selective, compulsory, comprehensive institutionalizing of research inventions made by faculty. The state claims are not made based on funding, and aren’t part of a federal contract to which investigators are joined by side agreements required by the contract–rather the state claims are made based on iffy things as employment (as if faculty are directed in their work by the state, and they do their work so that the state may own and exploit their work) and use of resources (as if the state only allows resources to be used if it may own the results). Such a practice, to follow the lead of federal research policy, should be stopped. There is no compelling reason for government control of faculty scholarship. Faculty gain nothing by becoming the day-workers of government administrators. As Paul Feyerabend once cogently argued, separation of science and state is as important as the separation of church and state.

Once the state becomes an interested owner and money-seeker of invention, it gives up its role as impartial mediator of markets and interests. In the world as understood by John Locke, doing so means that one loses the “Freedom of Men under Government” (see the Second Treatise on Government, Chapter IV) and one reverts to the “Law of Nature” where there is “no appeal but to heaven”–that is government ceases to provide the foundation for appeal because it is itself self-interested in the outcome. For a particularly painful account of what this meant for state and federal government in the United States, see Richard White’s treatment of the transcontinental railroad corporations in Railroaded. The Bayh-Dole Act has left faculty investigators and inventors exposed to predatory state interests, which have been provided with arguments that aim to rationalize the disrespect of faculty independence, even as Bayh-Dole enfranchises that independence in federal policy.

Here in the state of Washington, Senators Chase and Shin have introduced again a bill that could serve as a model for other states, a Faculty Freedom to Innovate Act. The bill, SB 5247, would prohibit the state’s public universities from claiming ownership of inventive faculty scholarship as a condition of employment or use of resources. A university could still come to own such inventions, but only under a voluntary arrangement. The University of Washington patent folks have already come out against the bill, as might be expected. They argue in a fiscal note that if they don’t get to claim all inventions then they will be unlikely to recover all the expenses they pay out for managing all the inventions they do claim from the “small subset” they actually might someday license for big bucks. I will spare the garbled and confused arguments–at least here–and point out that no one apparently asked the faculty at the university what they thought of the chance to get some bit of their independence back from the control of the state.

Should the state be an interested party in the exploitation of patent rights–interested in the money-making part, especially? Or should the state be interested in serving as advocate, mediator, and steward for patent rights, when doing so advances the use of inventions? The Bayh-Dole Act says, if you are going to trust someone to make good decisions in this society regarding inventions, we may as well start with faculty researchers able to win competitive grants from the government. Let the government serve as an impartial arbiter; let faculty find (and develop) organizations to assist them; use the patent system creatively to promote use of inventions. And that promotion can mean standards, platforms, open source, startups, cross-licensing, public licensing, make/use commons, royalty-free licensing, and even from time to time exclusive licensing.

Now it is up to the states to step in and defend their faculty in public universities from the same sort of institutional compulsion to use the power of the state to claim ownership of inventive stuff. Patent reform will not get at this. Revisions to Bayh-Dole could get at this, but it will be tricky to get done and the outcome is iffy. Trying to “fix” (as in car) university licensing operations by instituting “template” deals and postponing payment or adjusting policies for efficiency won’t do it–in such “problem-solving,” faculty are still not provided with the respect they deserve and the independence that is so closely associated with innovation.

We need “fix” as in “cat,” not “car.”  That is, we need to limit the propagation of state claims to faculty scholarship. This is what SB 5247 aims to do [it did not even get to a vote]. This is what a state-level innovation policy looks like, when directed at university faculty research. This is what Bayh-Dole does on the federal level. Now it is up to the states to follow the federal government’s lead. When they do so, they will re-ignite faculty interest and initiative, dramatically lower their own expenses and exposure to liability, encourage open collaborations, and spark creative class behaviors–in short, an innovation policy. There will be challenges, as university patent licensing officers and policy writers have sought to destroy much of the infrastructure that university affiliated personnel have created and advanced throughout the 20th century–from Research Corporation to affiliated university invention management foundations, to the funding model that underlies the National Science Foundation, to the biotech startup (Genentech, say), to open source software. It will take some effort to move administrators from dollar-hungry licensing moguls to mediators and stewards. But it has been done before, and it must be done again.

In each generation, it seems, we face fundamental questions of how we will develop our society. Often those questions don’t show up on a billboard for everyone to recognize. They appear as responses to creeping institutionalism, or complacency, or entitlement, and someone has to say, enough is enough and work up the energy to put things back on track. This is such a time for faculty-originated inventions. It is part of the de-stagnation of American research. It is part of the restoration of Bayh-Dole to its provisions as written, as affirmed by the US Supreme Court in 2011, and in defiance of an entrenched bureaucracy that is contemptuous of that affirmation and seeks to make Bayh-Dole into a law that strips university faculty of their leadership roles and responsibilities as independent investigators.

I don’t know when state control of scholarship got to be a really keen thing, but in university technology transfer offices, it has come to be a given, a public good. This could be East Germany, in the old days. Perhaps that is where the ghost of East Germany ended up, in university technology licensing shops, where to be totalitarian is just fine if it keeps one in a good paying job.  Under a Faculty Freedom to Innovate law, university licensing operations will do just fine. They will, I am sure, flourish in a way they are not doing at present. University licensing offices do not report their true metrics, so it is impossible to see what they are licensing and what is held hostage to future payments that will never happen. But everything points to the fact that they are not licensing a third of what they are claiming. Maybe it’s 4 or 5%. Which would be typical for any non-selective, compulsory, comprehensive institutional accumulator of patents. But that level of activity–often to serve the SOTs, not the practice community, and often about the money not the use (use = success, not infringement)–is no justification for state control of scholarship or to expect that such control will advance innovation. It hasn’t, and won’t.

The best way for states to fix the flaw in Bayh-Dole is to pass a Faculty Freedom to Innovate law.  The states that do so first will be the foundation and springboard for research-based innovation. Maybe Washington state will be the first to do so.






This entry was posted in Bayh-Dole, Freedom, Policy, Technology Transfer and tagged , , , . Bookmark the permalink.