We are working through Niels Reimers’s op/ed in the Mercury News, published last April (2021) and now being used by the Bayh-Dole Coalition, a lobbying organization backed by a number of universities and front groups, to try to prevent Bayh-Dole’s public protections from being used to, well, to protect the public from universities and their exclusive licensees exploiting public health for unreasonable profits–essentially, using patents to price-gouge, undersupply the market, and suppress competition.
Reimers claims that if Bayh-Dole is used to protect the public from such practices, then universities won’t be able to do more deals that exploit public health using patents on inventions made in federally supported research, and all hell will ensue, with science itself shuddering to a halt and with it investment in new medicines. There just won’t be any, according to Reimers, unless the public allows itself to be exploited.
This, apparently, is the genius of Bayh-Dole. Congress, so the argument implies, realized that the best way to get new medicines to the public was to create incentives for universities and companies to use patents to suppress competition, create monopoly pricing, and screw the heck out of the public any which way they could. Congress included in Bayh-Dole various policies and requirements that appear to promote competition and reasonable terms, among other things, but these were clearly only intended to be ceremonial gestures and Congress didn’t intend for these to be ever used, as doing so would limit the opportunities for universities to work exclusively with the companies most ready to screw over the public. Follow?
Reimers claims that Bayh-Dole “unlocked” the potential of the federally owned patents, those 28,000 patents owned by the federal government that had already been made available open access because the contractors that hired the employees that made the inventions under federal contract did not want to pursue patenting. That doesn’t work. Even though Bayh-Dole authorized the federal government to deal in patent monopolies–grant exclusive licenses, play favorites, demand payment, have a stake in pushing prices higher–that authorization would be forward-looking. The federal government was not going to suddenly lock up its patents, and allow only a single company to have a go at each invention that already had been released open access.
But perhaps Reimers means rather that the federal government also took ownership of inventions made at universities and made these inventions available open access, thereby “locking,” as it were, their potential. That’s a pretty twisted usage of “lock.” The idea perhaps is that people will not use something that they could not exclude all others from using. That doesn’t work, given that we use new stuff all the time that others have access to, and so do companies. Even pharmaceutical companies use the results of federally supported research when these results are useful, and also not behind a patent paywall. So that must not be what Reimers means.
Perhaps then Reimers means that a nonprofit’s inventions all must be “developed” in order to be used, must be made into commercial products, presumably mass produced, and so the cost to design such products, build out any new production facilities, and test things out could get expensive, and the only reason anyone would do all this work is that they had a patent monopoly that would prevent competition. Even here, though, Reimers glosses over all the research-based inventions that are methods and research tools–stuff that the researchers are already using and others may use with no more outlay than the researchers themselves have made. No “development” necessary, and if there is “development,” it is not lots of money, not “huge risks.”
Reimers skips over, too, all the inventions that can be used in a non-product form–companies can implement directly, such as to improve production processes. Reimers also fails to mention inventions that depend on rights held by others, which can’t be practiced without agreement with those other rights holders–and for these, a reasonable resolution often is a cross-license, which ruins the idea of a patent monopoly as the basis for development. No point in developing something that you can’t practice because you have been a cheap prig about your exclusive rights relative to the exclusive rights of others. Again, it just doesn’t seem like this is what Reimers is getting at. If it is, then he is being silly. Most inventions of any usefulness don’t need exclusive rights to be used–the exclusive rights are helpful in getting the university owners of those rights paid (for doing mostly nothing) and getting paid more because the companies they license to exclusively can price much higher–ten times higher, even–than they would if there were competition or they were required to offer “reasonable” terms.
Reimers is interesting. His most famous licensing program was for the Cohen-Boyer gene splicing inventions–before Bayh-Dole. In that program, Reimers did everything right–Stanford licensed its patents (and UC’s patents) non-exclusively at very low cost, consulted with industry at each point and made the patent applications available for inspection, declined offers for exclusive licensing, and licensed pretty much everyone in the industry, making $260M or so in the process. But what does Reimers say about this amazing licensing program?
Hughes Well, another criticism that arose in the late 1970s was that recombinant DNA technology had already been widely adopted. In fact, companies were founded on that basis.
Reimers Sure, before our patent issued [1980].
Hughes Yes. Was that a stumbling block?
Reimers Well, no. I typically like to be at the front end of technology transfer. That is, we approach companies and propose we work together in developing this technology. We’ll file the patent, we’ll exercise the right to exclude others, and we’ll develop it together. Well, when you do nonexclusive licensing, in a way you’re just applying a tax. But I had no shyness about that here.
The invention–federally funded, managed under both the NIH and NSF IPA programs– was already being used, and without the need for either a patent or exclusivity. Odd. So Reimers calls payments made for Stanford’s non-exclusive license a “tax.” There are other things that a non-exclusive licensing relationship can support that is more than a tax, but Reimers doesn’t recognize those.
There’s really only one target for Reimers’s argument, and that is the area of medicines and related biomedical inventions. Even here, there’s a difference between showing that a compound has biological activity and how that compound might be used in medicine. Certainly one use would be to consider testing the compound in various formulations ahead of mass producing a medical product. Even here, a patent monopoly may not be necessary–the cost of development might be covered by a federal agency or by a nonprofit foundation. And that’s what has taken place, historically, for many biomedical interventions, from vaccines to antibiotics to anti-cancer agents. While there are medicines and vitamins and vaccines that have been developed at private expense, there’s no public policy reason that this must be the case for every invention.
The new law unlocked those patents’ potential. Suddenly, universities could easily license their researchers’ discoveries and ideas. In my job, the number of invention disclosures submitted to our office — paperwork detailing new discoveries — immediately doubled.
Reimers turns to making the claim that Bayh-Dole, permitting federal contractors to own and exploit patents in inventions made in federally supported work without having to explain first how their use of patents (necessarily, exclusive deals) better served the public interest than would open access.
Nationwide, this single reform created an unprecedented wave of private-sector innovation. Between 1996 and 2017, more than 13,000 start-ups formed based on licensed university research.
There’s no evidence that Bayh-Dole created an “unprecedented waive of private-sector innovation.” AUTM–the lobbying group created by Latker to pitch nonprofit exclusive patent control–does not even bother to track Bayh-Dole inventions. Patenting by universities does not show any great change after Bayh-Dole came into effect. Things continued to grow at roughly the rate they had been growing at. As for startups, AUTM overcounts them, does not have any idea which ones are based on Bayh-Dole inventions, and furthermore has no interest in whether those startups end up producing a commercial product based on the licensed invention. Perhaps Reimers is right, but this is in the same general category as speculating about pigs and wings.
Reimers, having moved through a series of fallacies and fantasies about Bayh-Dole, turns to his worry of the day–state attorney generals want the government to march-in and open up access to covid treatments. “This would be an enormous mistake.” Maybe, truly, it would have been an enormous mistake–given how terrible the mRNA “vaccines” have turned out to be–don’t stop infection, don’t stop transmission, and have huge, even life-threatening side effects and unknown long-term issues. Oh, please. Just look at the data.
But Reimers goes on:
The march-in provision allows the government limited authority to license additional developers, but only if current licensees are failing to commercialize a patent.
Reimers is wrong. At least he’s consistent! March-in allows a federal agency to compel the patent holder (contractor, assignee) to grant licenses if the patent holder (i) has not timely achieved practical application; or (ii) and (iii) has failed to reasonably satisfy a public health or regulatory need; or (iv) has failed to require US manufacture in cases of exclusive licenses in the US to use or to sell product based on a Bayh-Dole invention.
There’s nothing in Bayh-Dole that gives a company a free pass if it has “commercialized” a “patent.” The standard for march-in (i) is “practical application.” Bayh-Dole defines practical application at 35 USC 201(f)–basically, an invention is being utilized and the benefits of that utilization are available to the public on reasonable terms. It’s not then that there’s commercialization, but that the terms on offer to the public are reasonable–including, obviously, reasonable pricing. As for march-in (ii), it’s not sufficient that a company is selling product–there has to be enough to meet the public need. Again, mere commercialization doesn’t matter to Bayh-Dole. Reimers wants you to believe it does, but he is wrong about the law.
No start-up will pay for the right to utilize a scientific, university insight, then invest in developing it further, if the government can dismiss intellectual property protections at will.
And why should a company have to pay for the right to use an invention made in university research, especially research that’s been pitched to the federal government as in the public interest? Reimers has the wrong question. A way better question is why universities and federal agencies don’t fund development of research findings if those inventions directed to public health are so gosh darned innovative? If a startup “invests” to “develop” an invention, then its proprietary interest lies in the developments. It doesn’t much need to own all the things on which the developments are based–but that startup most assuredly would prefer that those foundational things aren’t owned by others, ready to lock them out or make them pay. At least if other companies own rights, a startup might be able to cross-license and so get on with things. But if a university owns the rights, then it is money or nothing, and if money, then it pretty much has to be an exclusive license with a poison pill terms to prevent cross-licensing or dedication to a standard or sublicensing. The last folks that most startups want to have owning foundational patent rights is a university. That’s the practice truth. Even university inventors don’t like it, other than that universities offer to pay for the patenting–and if all you want is a patent on your academic c.v., then that looks like an okay deal. Otherwise, it’s all–I’ll use a technical term here–administrative crap.
In the last 40 years, the ingenuity of American companies has transformed almost every aspect of our lives. The Bayh-Dole Act made this possible. Undermining it could bring our technological renaissance to an end.
The first sentence is suspect. Many non-American companies have played a role. Lots of aspects of our lives have not been transformed, or have been transformed by other than companies. But the second sentence is more nonsense. Bayh-Dole did not make whatever transformations there have been “possible.” At best, Bayh-Dole has allowed contractors to hold patent monopolies and sit on them, without having to make a case that sitting on a patent monopoly better serves the public interest than open access. At best Bayh-Dole has allowed university patent administrators to gain a happy livelihood misrepresenting the law and making unsubstantiated claims for its wild success. Sure, Reimers can have whatever opinion he wants to about Bayh-Dole. Maybe the tooth fairy has made it all possible, too. There’s no evidence for either position, but maybe it feels right.
Reimers calls using Bayh-Dole march-in “undermining” the law. How very strange! Given the pressing public need, and the failure of companies holding exclusive rights to meet that need, why wouldn’t federal agencies (and the public) turn to Bayh-Dole’s public protections to address that need? That would not be “undermining” Bayh-Dole. Not using Bayh-Dole’s public protections is “undermining” Bayh-Dole. Reimers’s argument boils down to “universities won’t be able to make as much money dealing in exclusive licenses in the future if the public allows Bayh-Dole to be used to address an immediate public health crisis.” It is a petty, sniffling argument. It’s the same argument that the NIH used in Public Citizen to prevent public disclosure of its exclusive licensing terms. “Oh, please, Court, prevent disclosure of the terms we negotiated, because if we are forced to reveal these terms, then we won’t be able to do more exclusive licenses like this one, even if our licensee is screwing over the public with outrageous pricing because doing more deals that screw over the public is more important than protecting the public from unreasonable terms.” Of course, the Public Citizen court found the NIH argument persuasive, so it’s good of Reimers to try it in his op/ed.
But for anyone with their head on right, Reimers’s opinions are unfounded, deceptive, and have their moral compass–and their innovation compass–all wrong.
The core of Reimers’s argument is that if companies are exploiting patent positions on inventions made in federally funded work–charging unreasonable prices, limiting supply, suppressing alternative versions, and the like–the federal government should not use rights it has under Bayh-Dole to address those behaviors because doing so will “chill” the opportunities for other companies to take exclusive licenses with the intent of doing much the same thing.
Companies taking exclusive licenses to inventions made in federally supported work often screw over the public, but that screw over is an essential “incentive” to the effort to get companies to invest in the development of new products based on such inventions. Relying on Bayh-Dole’s public protections to introduce competition ruins the screw-over incentives and everything that’s good in the world will melt away and fail–or at least technology transfer will fail, private investment in developing inventions made in federal research will fail, the federal research itself will be wasted, and so science will stop, and we will fall into a dark age.
One does not have to run around looking for empirical evidence to critique Reimers’s basic argument. Bayh-Dole, according to Reimers, gives companies that aim to screw over the public an inside track to take exclusive licenses. The public benefit that then arises must necessarily include the screw-over incentives. The only way the public will benefit from federally supported research is if companies are free to screw over the public, and a good way to do that is to have patent positions that suppress competition, multiple suppliers, variations on product features, and the like.
Reimers’s position becomes a self-fulfilling prophecy. If university patent administrators insist on licensing only exclusively and only to companies that will license only if they have the right to screw over the public, then it must follow that if university patent administrators cannot assure companies that no one will rely on Bayh-Dole to introduce competition when the companies fail to meet public needs or offer products on unreasonable terms, then the patent administrators will not be able to those sorts of deals with those sorts of companies. One might respond–exactly. Congress did not intend for Bayh-Dole to be used to select for those sorts of companies, doing those sorts of things. Go ahead, try to find any shred of evidence in Bayh-Dole that Congress intended the thing Reimers argues for. Yup. Not there.