[I have made revisions and additions and placed the second half of this article in part 4.]
We are still reflecting on reflections on Bayh-Dole by “leaders” hoping that you will follow them. More:
prior to the Act, the government often funded research to spark innovation, but then put the research in the public domain for non-exclusive licensing,…
This is fake history. It is true that some federal agencies did take ownership of inventions and place inventions in the public domain or patented them and released them non-exclusively and royalty-free. And it is true that in some cases Congress attached a requirement of government ownership of inventions to specific funding authorizations, such as in the Space Act that established NASA. But it is not true in general and not true in general for the nonprofits receiving most of the government’s basic research funding. It’s fake, but you are supposed to believe it because it is stated as a general fact and made to sound true.
Company contractors with non-governmental markets could own and exploit inventions made under federal contract. This was long-standing practice for the defense agencies. Most of the 26,000 patents held by the federal government–vastly most–were defense-related inventions that the defense contractors had waived their interest in patenting. The government then patented these inventions to publish them (and perhaps to show that their research support produced patentable inventions) and (if it matters) define some control over the domestic market for the invention. The government, as it were, could by means of patents control imports of products practicing these inventions, thereby protecting in some way the interests of U.S.-based companies in the U.S. market (if not also internationally).
The claim that the federal government always took ownership of inventions is not true for nonprofits in NIH and NSF IPA programs–where most universities got their federal funding. In those programs, nonprofits were required to take ownership of any invention made in work supported by the agency that the nonprofit chose to patent. Otherwise, outside IPA programs, nonprofits could request a determination of greater rights and get approval to hold patent rights. The determination process could be slow–at least by patent administrator standards–and the nonprofit had to make a compelling case that exclusive patent control was better for the public than was open access. The determination of greater rights process, though having its own bureaucratic problems, made sense. Why should a nonprofit, which has no particular industrial use for inventions it acquires, deal in patent exclusivities, especially if companies are ready to use or develop such inventions and don’t need the burden of a patent or a nonprofit demanding payment as a precondition? The general policy idea was that nonprofits should have to make the case that their exploitation of monopoly patent rights better served the public interest.
The ghastly thing about the IPA programs was that these programs allowed nonprofits to deal in patents whenever they chose to–by notice to the inventors that the nonprofit was taking ownership because it wanted to file a patent application. No justification required. In place of the justification the IPA placed conditions on the nonprofit’s exclusive licensing. These conditions–that the nonprofit has tried open access and it didn’t work, or the nonprofit considered open access and decided that it wouldn’t work, and for no more than eight years–drop public accountability and leave nonprofits to do whatever they want.
Here’s the IPA master agreement, VI(d):
The Grantee may license a subject invention on an exclusive basis if it determines that nonexclusive licensing will not be effective in bringing such inventions to the commercial market in a satisfactory manner.
The IPA ignores entirely use of inventions that does not require a “commercial market.” Research uses, to start. Uses by professionals–think of injury rehabilitation methods or analytical techniques. Uses in standards. TCP/IP, a fundamental internet protocol. And the use of “effective” leaves plenty more room for a nonprofit to choose exclusive licensing because the nonprofit “determines” that exclusive licensing is “effective” as a general truth. The IPA in essence excludes in its conditions the very things that open innovation values–reasonable, non-discriminatory access, freedom of practice, freedom to adapt and vary, freedom from preconditions. Instead, nonprofits are told that they can suppress all other use of any invention whenever they determine it is effective to do so to “bring inventions to the commercial market.”
Exclusive licenses should be issued only after reasonable efforts have been made to license on a nonexclusive basis, or where the grantee has determined that an exclusive license is necessary as an incentive for development of the invention or where market conditions are such as to require licensing on an exclusive basis.
Here we have two prongs. The first says, try non-exclusive licensing. The second says, don’t bother trying. Consider alternative language, not used by the IPA:
Exclusive licenses should be issued only after reasonable efforts have been made to license on a nonexclusive basis and the invention has not been brought to the commercial market in a satisfactory manner, and that the scope of exclusivity is not greater than reasonably necessary to provide the incentive for bringing the invention to the commercial market.
In this alternative, nonprofits would have to try non-exclusive licensing first, and if that failed, then they could move to exclusive licensing, but only with the scope necessary to attract a company to do the commercializing. Bayh-Dole, according to Norman Latker, who drafted Bayh-Dole from the ruins of the Thornton bill, is based on the IPA program. Any nonprofit–and there were over 70 universities and other nonprofit research organizations in the NIH IPA program–could take ownership of inventions made in IPA covered work and do mostly whatever the nonprofit wanted to do, such as license exclusively public health inventions to pharmaceutical companies.
Our industry leader reflects that federal funding was often intended to “spark innovation.” That’s made-up, too. Sure, in a very broad argument, Vannevar Bush proposed a National Research Foundation to extend the frontiers of science. New knowledge at scientific frontiers then might be appropriated to develop new technology to serve public needs, starting with health care. This is the “unexpected model of innovation.” New science–the first step–is then combined with a keen perception about what a given practice area needs beyond what it is trying to do for itself combined with a keen knowledge of what’s available in industry to produce something new can “spark innovation.” That’s the second step. To get to the innovation bit, though, the unexpected model of innovation requires a third step–a focused skunk works with rapid prototyping capabilities and partnerships with key companies to supply manufacturing platforms and components to spec. The driver for their participation is the prospect of government procurement contracts and sales to the public.
Health care industry would be like the defense industry. Federal agencies would identify public health needs and companies would bid for contracts to create the new technology enabled by new science to make the “weapons” in this new war on disease. Companies would make their money from their government contracts, not by selling at monopoly prices to the public. Bush argued that his NRF would be focused on science, not on any federal agency mission and not on any corporate agenda. Bush also argued that the federal government should not operate its own research laboratories but should leave science to an independent scientific community. Bush did not win out. Federal agencies insisted on being authorized to conduct their own extramural research programs. University administrators piled on, insisting that research work be spread around among federal agencies.
Instead of the NRF, then, we have the NSF. But we also have many other federal agencies sponsoring research, and over 600 federal labs funded to do research, a number of those operated by contractors. There are no second steps of bringing together ace new science with industry technologists to work out things that governments and companies would not dare propose for themselves. There are no third steps of creating skunk works with focused development, supported by only the top companies. Instead, work is spread around like government pork funding, across agencies, across many universities and nonprofits, in overlapping, interconnected, cross-dependent work. Each university then claims it bits of ownership, holds out for exclusive licenses which mostly never happen and are mostly all the worse for innovation when they do.
Once the university administrators saw what they had advocated for–many federal agencies offering grants–they complained that the agencies had different rules and on top of that Congress had required federal ownership for some funding and not for other funding. And you see how in a bureaucrat’s mind Bayh-Dole then becomes the hack solution to a bad funding architecture. It doesn’t fix the architecture but rather shifts the defects to other parts of the system. The role of these “industry leaders” reflecting on Bayh-Dole is to dissuade anyone from looking at the new defects or if they do blaming those defects on the Bayh-Dole hackery that has produced them. It is a bit of absurdity to hold out that Bayh-Dole, a law to fix up the needs of patent administrators at nonprofits by enabling exclusive licensing of inventions made in publicly supported work without so much as an explanation of how such exclusive licensing would benefit the public more than would open access, would somehow spark innovation, or would be the thing that finally permitted federal agencies, dully funding nonprofit research, to show that their funding really was about “sparking innovation” and not about whatever else goes on in the politics of awarding grants.
Most federal agency funding, as Bush pointed out, was mission-directed. Whatever innovation might happen in that mission-directed work was the innovation that institutions could imagine for themselves and justify approving and funding. That’s the dullest sort of innovation, if one can even call it that. “Sparking innovation,” no. And even if one did want to “spark innovation” with random government payouts, wouldn’t you think that open access would be a lot more sparkier than putting everything significant in fragments behind patent paywalls managed by nonprofit bureaucrats looking to swing exclusive deals with monopoly pricing primarily to harvest money from the public via higher prices to fund their research and expand their research administration?
You might see, then, what would happen if the federal government adopted the policy of holding back all such inventions from open access in favor of choosing one favored company to take a patent monopoly position relative to everyone else in the industry, including the contractor in whose work the invention had been made. Doing so would force contractors to seek those patents rather than waive an interest–to defend the contractor from the government’s machinations to find an exclusive patent licensee.
Bayh-Dole’s implementing regulations then add a provision copped from the now-obsolete Federal Procurement Regulation that provides that even if a contractor waives its interest in a given subject invention, so long as it has properly disclosed the invention, the contractor retains a royalty-free license to the invention. That’s not anything authorized by Bayh-Dole, but there it is. That does not do a thing for access to an invention by the inventors or the research team or anyone else that might want to do research on or with the invention–the nonprofit becomes a sort of prison or reservation for the practice of the invention, but no one can practice the invention outside of that prison but for the permission of the federal government, and that permission is reserved for exclusive licensees with a plan to “commercialize” the invention. And there–on the federal side of Bayh-Dole’s provisions, the plans required really do focus on commercialization, even though that’s neither the stated policy and objective of Bayh-Dole and not even the stated policy and objective of the federal regulation subpart that implements Bayh-Dole. Here’s 35 USC 209(f):
No Federal agency shall grant any license under a patent or patent application on a federally owned invention unless the person requesting the license has supplied the agency with a plan for development or marketing of the invention . . . .
This pernicious little requirement ends federal low-overhead royalty-free licensing without formalities. No-one can get access to any federally owned invention without committing to “develop” or “market” the invention. No one can just use the invention. Furthermore, equally perniciously, the government demands that the potential user of any such invention provide a plan–one cannot get something and play with it, work it, and go from there. No, there has to be a plan, like there have to be clowns. Oh, wait, they’re already here. If anyone wanted to start wiping out Bayh-Dole’s disasters, start with repealing 35 USC 209 and remove “exclusive, or partially exclusive” from 35 USC 207, along with “including the grant to the licensee of the right of enforcement pursuant to the provisions of chapter 29 as determined appropriate in the public interest.” That would be a good start.