One of the common description of the Bayh-Dole Act is that it established “uniform” federal patent policy:
Enacted on December 12, 1980, the Bayh-Dole Act (P.L. 96-517, Patent and Trademark Act Amendments of 1980) created a uniform patent policy among the many federal agencies that fund research, enabling small businesses and non-profit organizations, including universities, to retain title to inventions made under federally-funded research programs. (AUTM)
To replace these conflicting policies, the Bayh–Dole legislation was introduced to establish, by statutory mandate, a uniform patent transfer system across all federal agencies. (Senator Bayh)
Certainty of title to inventions made under federal funding is perhaps the most important incentive for commercialization. Implementation of uniform patenting and licensing procedures, however, combined with the ability of universities to grant exclusive licenses, are also significant ingredients for success. (COGR)
The University and Small Business Patent Procedures Act of 1980 (commonly referred to as “Bayh-Dole”) created the uniform framework that facilitates orderly and efficient technology transfer from universities and other institutions receiving government research funding to the private sector. (PhRMA)
It’s sort of funny that these descriptions of Bayh-Dole agree that something was made “uniform” but can’t quite agree on just what that is. The emphasis on “uniform” goes back to the earliest arguments for Bayh-Dole–and before that, to the arguments for a template Institutional Patent Agreement that could be used by all federal agencies. The IPA approach was essentially a federal master agreement that worked around certain requirements of the Kennedy statement on government patent policy, which with some later modifications, governed executive branch federal agencies in their treatment of patentable inventions until Bayh-Dole displaced it.
The Kennedy patent policy was based on diversity and flexibility. Different federal agencies had different missions. The military wanted research on weapons systems. The public health services wanted research on public health. NASA wanted research on rocket engines. The National Science Foundation wanted research to advance science. The Department of Agriculture (which was the first agency authorized to sponsor research) wanted to improve farming techniques. Within this diversity of missions, federal agencies also made a distinction between contracts and grants. A contract was a purchase, a procurement of goods and services for use by the federal government. With a contract, as with any contractor, the federal government could state the conditions on which it would fund work, including a requirement that the contractor deliver to the government title to any patentable invention. A grant, however, was made to advance a public capability–to develop a workforce of capable scientists, to advance the frontiers of science, and the like. A grant was a contract with fewer requirements. It was not quite a gift or “giveaway” of public money.
University administrators and faculty alike were strong advocates for diversity and flexibility. They opposed the idea that a single federal policy or agency should have control over research conducted at universities. They argued that distributing university research funding among the various federal agencies reduced the influence of government over university research and allowed for different kinds of research with different requirements to be conducted as necessary. Classified research could be done in federally owned, university-operated research centers; basic science could be done in university laboratories, federal agency mission support could be done with contract research. Diversity and flexibility were the keys.
It was within this context that Vannevar Bush proposed what would become the NSF, an agency that would focus on funding basic research to extend the frontiers of science. He proposed that the agency would take only a government license and leave all other matters concerning patents to the contractors and inventors. At the time, most universities did not make an ownership claim in faculty research unless a contract required it–that is, unless a sponsor of research required it. Where there was an interest in patents, universities directed inventors to Research Corporation or to a university-affiliated research foundation. These invention management organizations were set up to manage patent rights and distribute these to industry, mostly on a non-exclusive basis. The idea was to use the patent to teach industry something to its advantage while returning some royalty income to the inventors and their institutions, rather as a passive windfall than an active grubbing for money. When universities did claim ownership of patents in inventions made by their faculty, the policy reason for it was to protect the public and research from companies that might patent improvements and therefore block use or development of the invention. University patents were to be used to defeat certain uses of patents–monopoly uses, blocking uses, nonuse uses, trolling uses, patent pooling uses, anti-competitive uses.
The idea that a patent was an unqualified public good was not heard of in university policy. Among medical faculty, patenting was contrary to public service. Harvard went so far as to offer to pay legal costs of anyone challenging a biomedical patent. Even now, no one can enforce patents on surgical methods. One wonders why disease assays should be any different.
Both federal contracts and grants then reflected forms of “the public interest.” A contract reflected a public agency need, to be met by private contractors. A grant reflected a public agency’s determination that the public would be served with the information or products arising from the activities supported by the grant. In both cases, the question then arose, What to do about patents?
The federal government, as the party supplying the money, could set the terms regarding patents. Why, if the federal government, was funding the research, should it not also have the benefit of any inventions created? When the funding instrument was a contract–procurement–then the government could require delivery of any patentable inventions. This is the case today in contract research organizations. They exist to do the work required by others, and deliver what they do to their sponsors. It’s up to the sponsor to decide whether to file patent applications and what to do with any resulting patents. But when the funding was a grant, what then? The work to be done was often basic research, and the proposal was to investigate some area of the unknown, not to jump as high as requested, on time and with the specified deliverables.
When the federal government obtained patent rights, it could take one of three general directions. The first was to preclude a private marketplace from forming. For weapons systems, atomic (and nuclear) energy, and space vehicles, there was a certain logic in this. Big bombs! get your big bombs here! Now delivered by space rockets! Strike anywhere! Take out your enemies! (Offer not available in all countries; some restrictions apply; see our special offer on radiation suits). No, that wouldn’t do.
Similarly, if the government was funding essentially all of the research using contractors, then leaving patent rights with any one contractor set up conditions in which that contractor then had a future advantage relative to any other contractor competing for federal research dollars. It was not that the contractor with patents was necessarily best suited to further research, but that the contractor could prevent other contractors from doing that research, even if the other contractors were better suited to the work. If nothing was done about patents, then awarding a contract to a given contractor locked that contractor in–once patents were obtained–as the only contractor available to the government for work that depended on the inventions made. So the federal government reasoned it was better to take title and keep procurement competitive on some basis other than patent rights, other than creation of private monopolies on research contracting with the government.
The second general direction was to add the patent to a federal commons. No one used the words “federal patent commons” (as far as I know) but that’s what it was. Inventions made in federally sponsored research were acquired by the federal government and were either patented or allowed to enter the public domain. As for patents, there were two options, as the Kennedy patent policy (1963) put it, “dedication or licensing”:
“Dedication” meant publishing the invention into the public domain. Anyone could make and use the invention–for research purpose or otherwise. No one needed to ask or negotiate or license or pay. “Licensing” meant, by contrast, that there was some agreement in play, with requirements on use (such as limited to a particular area of practice) and perhaps with payment (though government treatment often includes an expectation of royalty-free licensing–the requirements on use were often depicted as being more important. Typically, government licensing was non-exclusive, not exclusive. After all, the government did not particularly need royalties and had no mandate to create monopolies. The federal government purpose in holding patents was not to make money off the public–it had taxation for that. If making money is not the basis for holding a patent, then some other public good has to be involved–documenting an invention in the patent literature, say, or stimulating competition, or limiting the control over an industry by a patent pool or dominant manufacturer.
The third general direction, then, was found in federal licensing. If a license was to be exclusive for a given area of practice, then the reasoning was that a grant of exclusivity was a measured exchange for the “calling forth of private risk capital” (to use the Kennedy patent policy wording). If an invention could not be used without further funding, and the federal government was not going to provide that funding, and that funding involved some risk–“risk capital” not just any capital–then an exclusive license might be indicated as being in the public interest. Even here, federal policy does not make particular claims to a share of money involved in the exploitation of an exclusive license. The federal government did not have a reason to enjoy a share of monopoly pricing; there was no discussion of how an exclusive license could create a private monopoly that would jack up prices and thereby increase the federal government’s income from the scheme. The purpose of the exclusive license was to recognize private funding at risk, and provide sufficient time for that funding–call it investment–to be recovered with some profit.
But even here, the idea was not to use provide the federally owned patent as a full-term private monopoly. In the Kennedy patent policy, exclusive licenses were limited to three years from the date of patent issue. That was a uniform requirement on all exclusive patent licenses offered by the federal government. The purpose was to give a risk-taker a head start on a new product or market and to motivate the risk-taker to get product done sooner rather than later. Somewhere between 4% and 5% of federally owned patents followed this direction–about the same rate for commercial use via licensing as U.S. patents generally. What’s not pointed out by others telling this same history, however, is that exclusive licenses were limited by executive branch policy for just three years from the date of the patent. If a three-year lead in development was not sufficient, so the implication went, then it was better that a patent be dedicated rather than licensed exclusively.
I don’t know of any data on the usage rates for federally owned patents that were dedicated. The implication by the advocates for Bayh-Dole is that only exclusively licensed inventions are ever used. On the face of it, that appears untrue. The present university dabbling in patent trolling argues against the proposition. I don’t know why anyone would adopt the position that exclusive licenses are necessary for use without some backing information. It may be that there are some industries in which the dominant companies refuse to use inventions unless one of them gets an exclusive license–an odd sort of industry agreement–but that does not prove any general case, or even a case within that industry, where smaller companies might take on development of an invention without the need for an initial exclusive position–a head start, or other proprietary positions, may be all that’s needed. Similarly, the willingness of companies in other countries to develop products that are the subject of monopoly rights in the U.S. but are public domain in other places points out that the monopoly isn’t necessary–even if it is desired, even if it is insisted upon by some companies as a matter of their business decisions.
By the time we move from the mid-1940s t0 1963, we have a number of federal agencies authorized to fund basic research (among them, in rough order Agriculture–1880s, HEW, Army, NSF, Atomic Energy Commission, NASA). Universities are fine with this outcome. Various laws are passed that authorize federal funding of research and in some cases provide for the federal agency to assert patent rights even outside the conditions of federal research contracts–as was the case with the AEC and NASA. You make an atomic rocket fitted with a nuclear warhead, and the government might not let you practice your invention. Oh, the uncertainty! the uncertainty!
What was not addressed in the legislation was the procedures under which a federal agency could release whatever claim it might have to title to inventions, or dedicate those inventions, or license those inventions. The one recurrent theme was that the federal government was to act in the “public interest” as determined by whatever federal agency was involved in the research. The “public interest” was (and is) a matter of judgment. There is no good guide to what might be in the “public interest” other than reason. The res publica of scientific research is a matter of finding out ways ahead, not operating by arbitrary rules. The challenge for federal agencies (and for Congress) was to find ways to permit diversity and flexibility while reducing needless agency conflicts over requirements and administrative overhead. Also at stake were agency decisions that were capricious or arbitrary–and which did not reflect any insight into the public interest.
If there was one thing clear in federal contracting, it was that the public interest was more important than government royalty income, was more important than creating private monopolies to gain an advantage in federal contracting, was more important than creating private monopolies to run up the price of products offered to the public. This was especially true where the purpose of the research was to advance the public interest–that is, by means of grants in areas such as public health and safety. Why should anyone–inventor, university, patent broker, company–gain a private monopoly by which to maximize income in an area of public health? Would that not be predatory, even if legal? Why would the federal government be complicit in enabling such a thing? These are the questions at the root of the patent problem faced with the expansion of federal funding for research, especially by means of grants.