A lawyer staffing service and web site, UpCounsel, has a friendly page that offers “everything you need to know” about the Bayh-Dole Act. They promise you can learn what you need with an “11 min read.” I think it’s something of a bait page, slapped together to get search engine hits rather than to provide anything close to accurate information about Bayh-Dole. Let’s see.
What Is the Bayh-Dole Act?
The Bayh-Dole Act gave universities, non-profits, and other small businesses the ability to earn patents to inventions. This law settled a longstanding issue about the patenting of federally-funded projects.
Bayh-Dole gives no-one any “ability” to “earn patents.” Nor did it settle any issue regarding patenting of “projects.” There’s no way you would learn anything from these two sentences, and if you already know what they were “intended” to mean, then you could not possibly learn anything new from them, as they merely remind you of what you already know.
The Bayh-Dole Act establishes patent property rights pertaining to inventions arising in federally supported projects (35 USC 200) and which are owned by a party to the funding agreement. We might call that a public covenant that runs with such inventions. Patents on inventions arising in such projects are not ordinary patents. There are restrictions on how a patent owner may use such a patent. Bayh-Dole also authorizes federal agencies to grant exclusive licenses to inventions owned by the federal government, but only after an agency makes a determination that an exclusive license is necessary. Finally, Bayh-Dole sets limits on the interest a federal agency can have in an invention arising in a federally supported project after a nonprofit or small business contractor has acquired ownership of such invention. Such inventions are called “subject inventions” and Bayh-Dole’s contracting provisions apply only to subject inventions, and not to inventions before they become subject inventions.
What’s the History of the Bayh-Dole Act?
The P.L. 96-517, formally known as the Patent and Trademark Act Amendments of 1980, added a new official policy for the granting of patents in the United States.
The policy does not have to do with granting of patents. Nor is it restricted to United States patents. Bayh-Dole replaced official policy–but only for patentable inventions that were made in performance of work under a federal funding agreement and have been acquired by a party to that funding agreement. That new policy has nothing to do with the granting of patents. It has to do with the scope of federal interest in inventions once contractors have acquired them.
Birch Bayh, a Democrat from Indiana, and Bob Dole, a Republican from Kansas, crossed party lines to work together to write this legislation. The Economist deems the law so important that the magazine famously called it “innovation’s golden goose.”
Bayh and Dole did not cross party lines. Their senate bill, which failed to pass originally in the house, had over 50 senate co-sponsors. Bayh and Dole were the title co-sponsors of the legislation, the poster-boys. The legislation was written by Norman Latker, patent counsel at the NIH.
The Economist a few years later backtracked on Bayh-Dole in an article titled “Bayhing for blood or Doling out cash?” That article was not so optimistic about the law or its effects. Money quote:
Many scientists, economists and lawyers believe the act distorts the mission of universities, diverting them from the pursuit of basic knowledge, which is freely disseminated, to a focused search for results that have practical and industrial purposes.
Back to UpCounsel’s “history” of Bayh-Dole.
Congress ratified this law due to a perceived need for a uniform patent policy throughout federally-funded research facilities. The belief was that the lack of reliable technology transfers had slowed down the pace of innovations in the United States.
According to Senator Long, Bayh-Dole was the worst bill he had ever seen, but he agreed to make it a parting gift to Senator Bayh. As for the political talking points, Senator Bayh did argue that America was losing its technological leadership and somehow more patents licensed exclusively on the $10b in federal grants to universities for long-term basic research would change this situation. The Institutional Patent Agreement program had allowed over 50 nonprofits to manage inventions made in projects receiving federal support. In the biomedical area it appears that the federal government’s licensing efforts were substantially better than the universities’ efforts (and more specifically, the universities’ patent licensing agents’ efforts). If the argument had been lack of reliable technology transfer, then the federal government would have placed ownership of such inventions directly with companies. University officials admitted as much in Congressional hearings, but argued that such a position could not gain the votes needed to pass the bill, so inventions had to be laundered through nonprofits to make private ownership of such inventions politically palatable.
Anyway, the argument was not about “reliable technology transfers.” It was about the “certainty of title”–that once a nonprofit had obtained title to an invention, no federal agency should be permitted to force the nonprofit to give up that title, short of failing to report the invention, want to keep title, and file patent applications. Bayh-Dole was about preventing federal agencies from protecting inventors, the public, and industry from predatory administrative behaviors at nonprofits. That policy was not so much “uniform” as it was “arbitrary.” The Kennedy and Nixon patent policies by contrast were “uniform.”
The federal government needed to write legislation to make a uniform group of laws to encourage technology transfer. The biggest problem was that the United States had problems transferring technologies that it owned.
The federal government’s patents were on inventions that were mostly defense related, ones that federal contractors had declined to own and patent. The government’s position was that such inventions were made available, generally, non-exclusively and royalty-free. Thus, there was no particular need to track licenses or actual use. The government was not going to sue its own citizens for infringement. The argument that was made for Bayh-Dole was that the government was ineffective in its management of the inventions it owned because it did not grant enough exclusive licenses and did not make decisions to permit nonprofit contractors to own inventions and grant exclusive licenses at will. The only area where there was an identified “problem” with the federal government approach was in the area of medicinal chemistry. The pharmaceutical industry boycotted classes of compounds identified with federal support that might have a therapeutic effect because the industry insisted on holding monopoly rights to such classes of compounds. The Kennedy and Nixon executive branch patent policies expressly forbade private monopolies on federally supported inventions in the area of public health.
Bayh-Dole, in essence, argued that it was virtuous for the public to subsidize profit-making ventures and to allow those private ventures to hold a monopoly right in the fundamental inventions–classes of compounds, research tools, basic methods–and thus prevent competition and maximize the money to be squeezed from patients. Bayh-Dole further argued that it was vital for the federal government to fragment those monopoly patent rights among nonprofits as research work was distributed to many universities with investigators working on related issues, so that no one company could possibly have access to the many patentable bits of new inventive work that those university investigators produced. If these are your policy truths, then say them and embrace them.
Prior to this act, countless innovations went unused in American businesses. That’s because the government didn’t grant licenses to the private sector.
Most inventions go unused. Most patented inventions go unused. Estimates are that less than 5% of patents see any commercial play. That 5% of federally owned patents were licensed–and presumably saw commercial play–is par. Nothing to see here. What’s more, the university agent licensing rate–cited as 25% or better–was only of the inventions the agents chose to manage. The rest of the inventions made in university research, then, also were not patented. But did those inventions go unused? Who could say, since no one bothered to gather the data. Quant, you see, can be just more political rhetoric.
The federal government did grant licenses to the “private sector.” Even Bayh-Dole’s advocates recognized that. But the federal government did not have to grant licenses, because it had no monetary interest in permitting use and had no intention to sue for infringement. Only when the federal government granted an exclusive license or managed an invention for quality control or other matter to protect the public from false claims and poor manufacturing would the government need to track a license and enforce its monopoly rights. Under the IPA program, of course, the federal government did not have to grant licenses because it required contractors to obtain ownership of inventions. Under most Department of Defense research funding, similarly, contractors were permitted to obtain ownership of federally funded inventions–again, no need for the government to grant licenses.
Any innovations that happened due to federal funding were not available to businesses.
Gibberish. An “innovation” is something that’s been adopted as new by someone. Inventions, by contrast, were available to all businesses–and to researchers. The advocates for Bayh-Dole acknowledged the point. In fact, this was what they denounced–what was available to all would be used by no-one. That, too, is nonsense, but it was said with such confidence that people who hadn’t thought much about the situation believed it. Companies use work from the public domain all the time. Companies use technology covered by standards. Companies use technology that’s cross-licensed. They don’t have a monopoly position, but that doesn’t matter. The only stuff that wasn’t “available to businesses” was the stuff that the pharmaceutical industry boycotted because it *was available to businesses*. Gawd.
One of Senator Bayh’s researchers discovered that non-profits had invented ideas that had gained 30,000 patents. These innovations had cost the government $10 billion. Only 5 percent of them were in use at the time due to federal laws.
It was a government report from 1978 that did not cite its methods or sources. Maybe one of Senator Bayh’s aides did library research. Beyond this UpCounsel is hopelessly confused. Most of the 30,000 patents were defense related, not invented at nonprofits. In the IPA program, over ten years, 96 inventions were managed. Yes, the federal government spent about $10b on basic research in 1980. But that was not the funding that had anything to do with the patents the government already owned, and those patents were not the ones that the nonprofits, generally, had anything to do with. The 5% figure represents the reported licensing rate for the federally owned patents. It has nothing to do with use. If anything, it has to do with how few patents were kept from general availability by an exclusive license. Use includes research use, in-house use, professional use–not merely mass commercialization.
These were the only technologies funded by the government that led to new or improved products. The other 95 percent went unused.
There’s nothing to support such claims. This is someone’s fantasy history. Many technologies that were government supported and not patented led to new or improved products. The digital computer, the internet, much software, vaccines, drugs. Oh, lordy, this is stupid stuff.
With the Bayh-Dole Act, Congress added a mechanism for non-profits like universities to grants [sic] licenses for such innovations to small businesses.
I know, UpCounsel produced a joke page on Bayh-Dole for The Onion, except it wasn’t funny enough. Bayh-Dole required nonprofits to give a preference in exclusive licensing to small companies, but that provision was removed three years after Bayh-Dole came into effect, leaving a stub of a requirement with no enforcement.
Bayh-Dole gave control of inventions to the universities that discovered them.
Read the Supreme Court decision in Stanford v Roche. Bayh-Dole does not give control of inventions to anyone. Bayh-Dole prevents federal agencies from contracting to obtain ownership of inventions that a contractor has acquired, unless the contractor screws up. Bayh-Dole says nothing (as statute) about how a contractor is to acquire any given invention. Bayh-Dole (in the form of the standard patent rights clause) arguably prevents a university from demanding ownership of faculty-made inventions as a condition of federal funding or use of resources provided by the university to comply with federal funding.
This law was important since universities lacked the financial means to bring these innovations to the market. A dollar’s worth of innovation isn’t cheap to sell on the open market. It costs the equivalent of $10,000 of private capital to sell. Non-profits couldn’t afford that. Businesses that could didn’t have legal access to the licenses needed to buy the products.
Basic research discoveries do not need to go “to the market.” They need to be used by other investigators, contributed to on-going platforms of research technology, evaluated for possible use in industry. Anything but being reserved for mass-market commercial distribution. Only a very few inventions would have their only public benefit as mass-market products. The rest of this paragraph is so garbled there’s little point in commenting. The Harbridge House report identified projects at federal agencies that developed technology to the “point of practical application” and those projects were nearly 100% successful in producing commercial products, apparently without the need for huge investments of private capital. The point, perhaps, that’s wallowing around here is that when a single company proposes to take a basic research invention in isolation and preclude further independent research and do all development itself, the proposition may be expensive and doomed to failure. Steven Johnson in Where Good Ideas Come From identifies many, many “innovations” that arose from “networked, non-market” activities, and very few that were the result of a monopoly position exploited by a company.
With the new system in place, innovations left the labs and entered the marketplace. The corporate world enjoyed a sudden influx of new ideas. American businesses benefitted from the ideas created at universities. In exchange, businesses built longstanding relationships with non-profits that were innovating in ways that the companies could not.
There’s no data to support the claims here. Bayh-Dole makes utilization reports a government secret. University-based research outputs were cut off by university patenting practices. Arguably, less usable information reached industry. Companies advised their scientists to stop reading academic publications. When companies sought to replicate academic inventive claims, they found that they could not do it in the majority of cases. Cite one independent study–not by the university patent industry or the pharmaceutical lobbying groups–that shows the benefits of Bayh-Dole to any segment of industry, as an industry, to “American businesses.” The primary new relationships built by companies with nonprofits were in the form of exclusive licenses that amounted to invention assignments, and each exclusive license cut off that university from many potential collaborations with all other companies in the same area of work.
From 1980 through 2002, universities increased the rate of patents by a factor of 10. From 1996 until 2013, the revenue earned from university patents was $518 billion. The belief is that this patents created 3,824,000 American jobs.
Now UpCounsel switches from inventions in federally supported projects to university patenting in general, without regard to the source of funding. Look, before Bayh-Dole, most universities didn’t patent because they recommended that inventors assign their inventions to invention management organizations, and especially Research Corporation. Thus, a university didn’t generally have patents issued in its name.
In 1980, universities, institutes, and foundations received about 520 US utility patents, of which 117 recited federal funding (most of these, no doubt, under the IPA programs that ended in 1978). Research Corporation had 117 more patents, with only 7 reciting federal funding. In 2002, universities received nearly 4,100 US utility patents, of which just over 1,600 cited federal funding. Research Corporation had 256 more, with only 20 citing federal funding. So, roughly speaking, UpCounsel here is correct–university patenting did increase 10x. But what does that tell us about Bayh-Dole? Next to nothing, since we are looking at a shift at universities from using highly selective invention management agents to doing things themselves, and we are conflating university patenting in general with patenting associated with federally funded research.
According to the NSF, federal research funding for university-based research went from $3.4b in 1980 to $19b in 2002, a 5x growth. One might argue that university-based research became twice as productive–or that universities have been much less selective about what to patent. And still, there’s nothing to say about Bayh-Dole.
Specific industries thrived thanks to the Bayh-Dole Act. Smart innovations such as the Cohen-Boyer patent from Stanford University’s lab led to a $35 billion dollar industry.
Har-har-har. The first Cohen-Boyer patent application was filed in 1974, well before Bayh-Dole. Bayh-Dole has nothing to do with Cohen-Boyer. For that matter, the Cohen-Boyer patents were licensed non-exclusively as research and development tools–on the federal model that existed before Bayh-Dole and outside of the IPA program. That is, Cohen-Boyer is exactly what the advocates of Bayh-Dole said couldn’t happen–what was available to all would be used by none. Except for the signature basic invention of biotech.
Perhaps the best example is an innovation that came in 1983. Historians often mention the Axel patents as a direct result of the Bayh-Dole Act. This technology introduced foreign proteins into human cells. Columbia University has earned more than $790 million in royalties from these discoveries.
Har-har-har. The first Axel patents were filed in February 1980. Bayh-Dole also had nothing to do with them. They could not possibly have been a “direct result of the Bayh-Dole Act”–unless of course UpCounsel is living backwards in time and has become more confused than ever. It’s true that Columbia University has earned a shitload of money from exploiting its patent rights. Nothing to do with Bayh-Dole. And for that matter, why is a university earning a shitload of money a measure of the success of federal policy? Where in Bayh-Dole is there a statement of federal policy that a purpose of Bayh-Dole is to help nonprofits make a shitload of money exploiting patent monopolies.