Defense Contracting, Bayh-Dole Fantasies, and Workarounds–2

We are working through a guide for Department of Defense contract officers that has been left up at a government website for nearly twenty years. Much has changed, but the guide hasn’t. From it we can gain some insights into how federal contract officers think about Bayh-Dole–or at least how they were mentored to think.

We have worked through some Bayh-Dole clauses, the anticipated industry objections, and proposed workarounds. Now we get to the guide’s account of the history of Bayh-Dole. Here we get much garble along with some good.

Prior to 1980, various statutes and regulations concerning patents established the Government’s right to take title to Federally funded patents and freely distribute the information to the general public.

The various statutes and regulations were directed at the government’s right to own inventions. “Federally funded patents” is just silly. The government supports research that results in inventions. The government does not pay contractors for their patent work. The government has the right to require the assignment of patentable inventions. Picky? Yes, but the pickiness points out that the guide is not in control of the details here. The point is not that the government had the right to “freely distribute the information” but rather that the government’s practice was not to enforce patent rights, permitting the general public to practice–make, use, and sell–products and services based on the inventions. A patent freely distributes information. That’s totally not the point.

This position has its foundation in the belief that the Government’s funds (i.e., taxpayers’ funds) were being used to conduct the research; therefore, the results–including patentable results–should be made available to the public (i.e., the taxpayers).

It was not so much a belief as it was a requirement. The Attorney General report of 1947 reviewed non-exclusive availability of inventions made in government funded work and concluded that exclusivity just wasn’t needed:

While opinions vary, the weight of experience is that Government-owned technology can for the most part be exploited to a satisfactory extent under a system of nonexclusive licensing or public dedication. In the occasional situation where the commercial use and exploitation of worth-while inventions are discouraged by the need for a substantial investment in promotional, developmental, and experimental work, with the attendant risk of loss, the Government should finance such operations, in whole or in part, to demonstrate or prove the commercial value of the invention. This method of encouraging the use of the invention is preferable to the grant of an exclusive license. (114)

And the practice was not uniform. The Department of Defense–the military departments–generally permitted contractors to retain ownership of inventions made under federal contract, but for the most part, contractors declined to do so. The NIH and NSF ran IPA programs–NIH’s began in the mid 1950s lapsed, and then was revived from 1968 to 1978. The IPA programs permitted nonprofit contractors to take ownership of inventions made in work receiving federal agency support.

While the Government took title to the patent, it provided the contractor who conducted the research a nonexclusive license. The patent that resulted from these sponsored projects was typically freely published or provided to any person who requested access to it.

Yes–everyone got a non-exclusive license. But there’s more garble. The patent of course was freely published–all patents are. The government’s licensing practice was to approve licenses at no charge, but sometimes with conditions on use.

Now the guide starts making up fantasy:

This free and open access policy to patents presented many problems for contractors. Envisioning commercial applications, inventors of new technology wanted to keep for themselves any economic benefits resulting from their research.

Maybe this stuff just sounds good. But even the Harbridge House report did not find such a problem. The pharmaceutical companies were not concerned with open access so much as with contamination of their own work whenever they agreed to work with compounds that had received federal support. The idea that inventors wanted economic benefits suggests “inventors” is being used here for “contractors.” In the universities, inventors did not uniformly want personal “economic benefits” from federally supported work. No doubt some did. But many did not, and some university faculties went so far as to create policies forbidding patenting. Even then, if faculty are going to request federal support for their research, and know going in that the federal government will request rights to make openly available–but subject to an inventor’s request to retain exclusive rights on a showing that doing so better serves the public interest–there’s no point to imagining that inventors wanted to financially profit from patent positions on inventions made in federally supported work.

Commercial companies depend heavily on the proper protection of their research to recoup any prior investments.

The Harbridge House report found industry practices, and within any given industry, company practices varied greatly. In general, it was not true, and is not true, that companies “depend heavily” on “proper protection of their research” to “recoup any prior investments” (whatever that means). Companies engage in research for all sorts of reasons, not merely to invent new products. And when they do research, they often publish that research or contribute it to standards. When companies wish to maintain control over research, they tend to use trade secret rather than patent (which publishes the invention). The general claim, then, that companies depend on patenting just doesn’t hold up. Here, it is set out as a general fact, almost as if it had to be concocted this way to justify Bayh-Dole (and even that is confusion!).

The thought that the Government could distribute their research results to whomever might ask for them became extremely unattractive to many contractors, universities, and research centers.

Now for full on fantasy: “extremely unattractive” to “many”–this is made up. There’s little evidence that government open access policies were at all objectionable to universities and research centers–especially those research centers operating as contract research organizations that routinely gave up patent rights to the clients that paid for the research. Most defense contractors did not care about patent rights. In many other industries, having access to results outweighed anyone holding exclusive rights. Think about it–if the government was funding ten or twenty contractors to work in a given area, and each comes up with inventions that the others will need, how does any one of them exploit its exclusivity? No, they will have to cross license or go open access. And cross licensing takes way more work–each company dealing with each other company–than does open access. If every company holds onto its inventions with white knuckles, then the use of the cumulative, collective technology is delayed for two decades.

What the university-affiliated patent development firms wanted was a faster turnaround on federal agency determinations of “greater rights” that allowed a patent development firm to retain ownership of inventions that it had acquired. Those firms, in turn, were highly selective, accepting only 10% to 15% of inventions for management–and of those, perhaps only half or less involved federal funding.

The guide descends into pure garble:

As a result, technologies that were potentially commercially viable were never fully available to the Government.

This does not follow from the guide’s fantasy premises–that inventors taking federal money became selfish about making money from patent positions–and whatever it is that does follow is garble. The idea–I think–is that because the government provided everyone with open access, “technologies” that were only “commercially viable” if a single company held exclusive rights would not be made available to the government as commercial products (if that’s what “never fully available” means). It appears that we are to believe that nothing invented and provided openly then gets used, and the government is helpless to continue to support research once something has been invented. It makes no sense. It is garble logic predicated on fantasy history.

In response to this situation, Congress passed the Bayh-Dole Act in 1980.

According to Senator Bayh’s introduction of the bill, the problem was that the US was falling behind other countries in technological innovation. Somehow, Bayh-Dole was to address this apparent problem. The situation was not that federally supported inventors at universities wanted a bigger bite of profits based on patent positions. But Bayh-Dole did get passed in 1980.

Focused on promoting the development of Government-sponsored inventions into commercial products,

There is no focus in Bayh-Dole on developing inventions into commercial products. The focus throughout is on use of inventions, not even “development.” Commercial use is anticipated, but not promoted, mandated, or required. The definition of “practical application” focuses on use of inventions–so that the benefits of use are available to the public on reasonable terms. Nothing about those benefits being commercial products.

the Act allowed small businesses and nonprofit organizations to retain title to the inventions they developed while working on a Government-sponsored program,

This part is mealy mouthed. Bayh-Dole allows small businesses and nonprofits to retain title to inventions that they acquire and which were made in federally funded work. The organizations don’t retain title to inventions because they develop them. Those inventions are not their inventions until they acquire them. They are the inventors’ inventions. The point of the law–such as it is–was to eliminate the delay in federal agencies making determinations of greater rights. Now a contractor, having acquired ownership of an invention, could retain title upon notice, without demonstrating any capability to manage patenting or development or licensing, and without having to make any case that dealing in patent monopolies would better serve the public interest than would open access.

apply for and receive patents on those inventions, and pursue options to commercialize the discoveries.

Nothing in Bayh-Dole allows or requires anyone to “pursue options to commercialize” whatever the guide wants to call things–inventions, patents, technologies, discoveries. A contractor could pursue options to “commercialize” an invention with open access–and might be in a better position to do so because it would have open access to all the other inventions made by other contractors working in the same area. Consider how carbon nanotube commercialization might have gone had not scores of universities patented every conceivable bit of invention around carbon nanotubes and created gridlock for two decades. Even if Bayh-Dole does allow “options to commercialize” (so long as public protections are ignored, say), there’s nothing–no focus–that requires commercialization, or for that matter use of inventions. Contractors can gain ownership and sit on inventions. The government remedy is to march-in, which it has never done, even in the face of nonuse, price gouging, lack of availability, and pandemic-like needs.

In 1983, President Reagan significantly broadened the scope of the Act when he issued a memorandum requiring the application of the Act’s provisions to contractual arrangements with any contractor, regardless of status.

Nonsense. Presidents do not change laws by memo. Reagan’s 1983 memorandum was followed by a 1987 executive order. That executive order–12591–does insist on promoting commercialization of “patentable results.” The 1983 memo, by contrast, changes executive branch contracting policy to follow Bayh-Dole practices–except that Bayh-Dole takes precedence over (most) federal statutes and a presidential memo cannot do that. That’s why both the memo and EO are prefaced with “to the extent permitted by law.” The scope of Bayh-Dole was not broadened. Executive branch patent policy was changed.

By doing this, President Reagan was hoping to attract more for-profit entities into the Government research and development (R&D) arena with the incentives that they would be able to retain title to the inventions they might develop and could exploit this IP in their commercial products and sales.

I feel I’m reading a high school term paper written the night before it was due. Nothing in Reagan’s memo or EO is concerned with not having enough companies involved in government research or that the reason for this putative lack of companies was that really good companies wanted to own any inventions they made and hold these inventions back from everyone else. In fact, the memo has just the opposite. Federal agencies are directed to waive or omit Bayh-Dole requirements if “this is necessary to obtain a uniquely or highly qualified performer” or the performer is “making a substantial contribution of funds, facilities or equipment to the work performed under the award.” Opposite to the guide’s fantasy account.

EO 12591 focuses on technology transfer–from federal labs, universities, and small companies–that is, from organizations that don’t have any commercial operations to those that do. Nothing here about broadening participation in federally supported research by companies possessed of a need to hold everything they do as patent monopolies. The “broadening” is of the “technology base” by “moving new knowledge from the research laboratory into the development of new products and processes.” Now, a “technology base” implies platforms, standards, cumulative technology–a base is a foundation from which to build. Sure, one could find in that wording the idea that to broaden a technology base means that a federal lab or university exclusive licenses an invention (i.e., assigns it) to a company. But that is hardly “broadening our technology base.” That’s exploiting an exclusive position in something that otherwise might become part of a technology base. Nothing works in the guide’s phantasmagorical account of Bayh-Dole.

The Bayh-Dole Act and the related executive order called for the identification, protection, and use of IP developed or modified in R&D contractual arrangements.

The related executive order–12591–is not the memo. The guide doesn’t see the difference. Bayh-Dole does not address IP–just patentable inventions, and those inventions must be made under contract. “Developing” an invention is outside the scope of Bayh-Dole’s claims–unless developing involves more inventing or testing necessary to demonstrate reduction to practice. And there’s nothing in Bayh-Dole that requires “protection” of inventions. If a contractor does not acquire ownership of an invention, it does not have to be disclosed under Bayh-Dole, and no one has to file a patent application on it. If a contractor is fool enough to acquire ownership of an invention and then not want it, then the federal government may take ownership of the invention–but also does not have to file a patent application. And with the amendments to Bayh-Dole in 2011 in response to the AIA first-to-file changes in patent law, contractors don’t even have to disclose inventions to the federal government unless they publicly disclose them.

And more nonsense logic:

These arrangements resulted in IP that could be worth millions of dollars in future revenue, in the same or similar applications, to the contractor who invented and developed it. Because contractors may have previously invested millions of dollars in their IP, they depend heavily on their retention of exclusivity to
recoup those prior investments in the future.

IP that “could be worth millions of dollars” (and could be worth nothing, or may only be worth something to people looking to exclude others)–

Contractors “may have previously invested millions of dollars” (without apparently inventing anything of value and too stupid to identify background rights in navigating federal procurement regulations)–

Contractors that may have spent millions must “depend heavily” on patents to bail them out (because they produced nothing, and have no other way to turn a profit except to prevent others from competing with them via a patent on an invention that they somehow weren’t able to make on their own and went out and sought federal funding for).

Despite all this fantasy, the guide pivots away from Bayh-Dole:

As industry takes the lead in technology and the Government needs to collaborate with industry on research projects, some of the provisions of the Bayh-Dole Act get in the way.

After Bayh-Dole is set up to promote technology development, apparently it is so successful that it becomes a drag on development? One would think that if “industry” leads government research, then the government does not need to “collaborate” so much as “procure.” As for research and development–the focus of Bayh-Dole–the government ought to get out more and find things to research and develop that it needs and industry isn’t providing. Wouldn’t you think?

The guide cites march-in rights as the problem:

For example, under certain circumstances, the Act permits the Government to directly license the technology developed under a Government research contract if the contractor does not take diligent steps to commercialize it.

This is reasoning that would bite its own tail if it could reach it. The guide has beaten us up with the idea that the government is missing out on company collaborations because inventors want to make money from their valuable IP in which they may have invested millions and depend heavily on patents to recoup their losses, and now we find out that contractors don’t like the idea that if they don’t bother to depend on patents and don’t bother to be diligent in commercializing these precious inventions, the government could march-in and require them to license (and receive royalties for doing so) or license directly if the contractor refuses to do so. Makes no sense.


In today’s world, technology is moving so fast that some companies may choose, for business reasons, to maintain their technology as a trade secret instead of patenting it. This practice is not recognized under the current regulations.

In “today’s world” of even 2001, if a technology is “moving so fast,” then a patent is not relevant because the technology will be obsolete before a patent can issue–and that means that people are using technology and developing technology without patent positions–and that means that all the guide’s blather about depending on patent positions is nonsense. In “today’s world” fast moving technology means patents are not meaningful except to trip up competition not worth working with, trolling industry to extract rents from everyone, and to have a seat at the table when it comes to mapping industry directions, standards, and interoperabilities.

It is true that trade secret runs against patents. Patents, indeed, were created as a remedy for trade secret, enticing practitioners in on the secrets in one city to move to another and reveal those secrets, in exchange for exclusive control. Later, patents became a way for a government to pick favorites for exclusive positions, even when practices were already known. And it is true–the guide is onto something here, but doesn’t quite recognize it–that Bayh-Dole is senseless when it comes to trade secrets. But as the federal government sponsors research and at the same time agrees to let contractors keep all the good parts secret, we have to grapple with how the public benefits from research the government pays for  that isn’t published and isn’t published as a patent. It’s one thing for public research to be open access–that makes some sense–but the reason, perhaps, that even Bayh-Dole does not acknowledge trade secrets (but for patent applications, reports of utilization, and terms of licenses) is that there’s little justification for public money to go to support research that isn’t published, other than for national security work. Certainly not for scientific or health research. Not for agriculture or mining research. What then, for?

This entry was posted in Bayh-Dole and tagged , . Bookmark the permalink.