Bayh-Dole is thin soup when it comes to federal innovation policy

NIST wants march-in for Bayh-Dole’s section 203(a)(2) and (3) to be for “national emergencies” only. Section (a)(2) concerns health or safety needs that are not “reasonably satisfied.” Section (a)(3) concerns regulatory requirements that are not “reasonably satisfied.”  But the *price* element is in 203(a)(1). Only (a)(1) relies on the definition of practical application–use with benefits available to the public on reasonable terms.

If an invention is not used or the benefits of use are unlikely to be made available to the public on reasonable terms, no practical application has been achieved. If there’s no practical application, then a federal agency has the right under the standard patent rights clause to march in and require licensing of the subject invention. Compulsory licensing is the remedy for a failure to offer the benefits of using an invention to the public on reasonable terms. At least there’s the appearances of an apparatus in Bayh-Dole to protect the public interest. In practice, it’s not that way. In practice, no federal agency has ever “marched-in.” In practice, federal policy is not to protect the public interest with regard to practical application.

An NIH attorney–Norman Latker–drafted Bayh-Dole to reestablish a patent monopoly pipeline from the NIH to pharma. Latker had previously restarted the Institutional Patent Agreement program and drafted the master IPA. The IPA had the same effect as Bayh-Dole–to place federally supported inventions behind  patent monopolies to be served out by nonprofits to the pharmaceutical industry. The IPA also had a public interest apparatus–but in practice that apparatus was ignored and nonprofits sought exclusive licenses for the biomedical inventions they claimed to own (and the IPA, unlike Bayh-Dole, provided a legal foundation for nonprofits to assert ownership of inventions made with NIH funding).

One argument for doing so is the monopoly meme. The claim of the monopoly meme is that without a patent monopoly, no one will use or develop a research invention. That’s nonsense, but there are people who love nonsense.

But there are other policy issues at play, for which the monopoly meme is pretty much a simple political cover, a fox to run off and get all the dogs barking after it. Another policy issue is how to subsidize the U.S. biomedical industry and grow it into a dominant global economic power. The policy indicated way to do that was to push $30 billion each year into into nonprofit research that then could be captured preferentially, in the form of patents, by US biomedical companies. In this policy vision, medicines are merely company assets–the primary goal is to create a profitable, dominant “industry.” In this view, the problem now is that the U.S. biomedical companies have been dumping medicines in other countries at prices much lower than charged in the U.S. Or, put another way, consumers in other countries have not paid a fair price for their medicines. Or, yet another way, U.S. companies have failed to export medicines for the value they should have had, and as a result, the US biomedical industry has contributed to the U.S. trade deficit.

From a “fair trade” point of view, U.S. companies are dumping medicines at below the cost of their development. That in turn suppresses opportunities for local companies to develop medicines. Another way, federal subsidies for U.S. companies allow them to unfairly compete with non-U.S. companies in the development of medicines for the global market. In this, Bayh-Dole is an enabler to funnel federal dollars to selected industries by means of “research” grants. Why might other countries then adopt something like Bayh-Dole? Of course, to route government subsidies to their favored industries without WTO scrutiny.

But there’s another policy matter at work that doesn’t get much attention. Think for a moment about the pharmaceutical boycott of federally supported inventions in the early 1960s. The Public Health Service instituted a policy that merely screening members of a class of invented compounds for biological activity should not provide the company providing the screening service with a monopoly on the entire class of compounds. Pharmaceutical companies then organized a boycott of inventions supported by PHS funding. In effect, the companies wanted researchers and nonprofits to deal in patent monopolies. As an industry, the companies wanted patent monopolies, even if that meant that one company would get a monopoly for a whole class of potential therapeutics. It wouldn’t much matter, because another company could then get a monopoly for a different class of potential therapeutics. The point of the boycott was to ensure that new discoveries–again, each discovery generally involves scores to hundreds to thousands of compounds–were wrapped up as patent monopolies and either went to some one company in the industry or went to no-one at all.

In this practice of patent monopolies, pharmaceutical companies suppressed other operating models for the development of new medicines–models that would have challenged the pharmaceutical company approach based on monopolies on compounds. One such model involved pharmacies. A pharmacy compounded medicines from established recipes. If a compound were generally available, then a pharmacist might create the required doses without having to manufacture the medicine at scale for national or international distribution. Thousands of pharmacists each serving a local market could destroy a national market for a mass manufacturing of a given drug. Patent monopolies on the making of medicines have the effect of turning pharmacies into dealers for commercial products created by pharmaceutical companies.

A second model might be called the scientific-government collaborative approach. This model thrived in the development of research tools and later served as an impetus to the development of open source software and other variations of open innovation. In this approach, researchers compete for priority of discovery and do not use patents to prevent other researchers from engaging in similar research. The approach defines a “pre-commercial” period of time, from invention to the point of practical application, during which any number of researchers and their organizations might contribute to create, from the ideas that had been circulated, something that “worked” and was “safe.” Once a technology, or proto-product–a medicine, a vaccine–achieved practical application, then anyone who had the capability was free to make it, use it, and sell it. Prior to comprehensive federal regulation of new medicines, patents were used to regulate sales–for quality, for true claims, and to suppress manipulation of markets and physicians by follow-on patents that aimed to block general access to the benefits of the products that had been developed collectively to this point of practical application.

We might say that in this second approach, collective research created a platform or commons for development, spreading the cost and sharing the expertise. Once something was developed, then companies could compete with pharmacists to provide patients with the best possible formulation and delivery of the medicine or vaccine or whatever.

The federal government was ramping up the idea of providing research dollars to nonprofits (and to for-profits) for biomedical research. This horned in on pharmaceutical company interests in research. The boycott and its resolution in the form of the NIH’s IPA program, and then Bayh-Dole, addressed both the assurance of receiving a substantial federal subsidy for research and suppression of alternative approaches to the development of new medicines.

Why would the NIH then ever march-in and ruin this scheme?

Let’s chase the fox for a few minutes. If the NIH justifies its funding to nonprofits and companies by the monopoly meme, then Bayh-Dole march-in would destroy  any possibility that anyone would use or develop or sell any drug based on the federally supported invention. Without a patent monopoly, no company will develop any medicine as a mass-produced product. That statement is true in the sense that no pharmaceutical company that has a tacit agreement with other pharmaceutical companies to deal only in compounds and classes of compounds that can sit behind a patent monopoly will ever develop a product based on work that cannot be handled by a patent monopoly.

In a more general sense, perhaps you can see how the monopoly meme–no company will develop any medicine–serves as a policy choice. Recast: without patent monopolies, pharmaceutical companies would have to compete with pharmacists in the formulation of new medicines, and that would take away profits, reducing the business opportunities for pharmaceutical companies and handing those opportunities to pharmacists, making what physicians prescribed.

Similarly, the monopoly meme argues against collective development, cumulative technology, collaboration, standards, and interoperability (truly important given drug interactions). The business case for the development of a medical product by a single company, which then expects to gain the full benefit of any profits, depends on the backing of patent monopolies. The actual development of that same medical product, however, does not depend on patent monopolies. Nonprofit organizations and government agencies have a mandate to develop such medical products entirely without any profit motive or determination that a product is not worth developing if it cannot be mass manufactured for lucrative return.

Tools of collaboration have long been used in all sorts of settings to create new things–new industries, even. We are not talking about some enforced openness, nor communism–but rather the freedom of association that has allowed not only researchers but also technicians, engineers, and others to choose to collaborate, even while competing for priority of discovery, even while holding proprietary rights in aspects of what each is doing. The effort of the group is necessary to create the opportunity eventually for some of the group to compete in an open market.

The monopoly meme aims to shut down such collaborative development by arguing that it is not possible, won’t ever happen, and if it does produce a useful medicine, no company will produce it. Bayh-Dole then becomes the tool to legitimize the fragmentation of research spread around to multiple organizations in different geographic locations into organizationally claimed patent monopolies. It’s stupid, but then policy doesn’t have to be very smart to bamboozle Congress, especially if Congress often wants to be bamboozled.

Thus, NIH cannot use march-in or it ruins the lingering claim that the patent monopoly pipeline is necessary in the first place. Without the claim that patent monopolies are necessary, there is absolutely no reason for Bayh-Dole to apply to health-related inventions. Bayh-Dole was passed on the claim that the federal government did a lousy job packaging research inventions up as patent monopolies and dealing them to favored companies. Nonprofits, led by “middlemen,” would do a better job of exclusively licensing federally supported inventions to favorite companies.

It is entirely possible that nonprofits would also do a better job than the federal government licensing inventions non-exclusively, but we still haven’t addressed the basic issue that there may be no need whatsoever to do any licensing at all. Licensing is made necessary because there’s an ownership position established, and the owner is not prepared to do the work to produce product based on the invention. If there’s no ownership position, then there’s nothing to license. If the owner is prepared to make product to meet all needs based on all meaningful variations of a given claimed invention, then there’s no need for licensing. If the owner makes product of sufficient quality and quantity and at reasonable prices for public benefit, then again no need for licensing.

What we might observe,  however, is that if what is most needed as the next step following discovery is that the discovery is taught to others–the discovery, the tools used, the concepts that underlie, the data gathered, the prospects for variations and improvements and applications–if this instruction is most necessary, then neither government officials nor nonprofit administrators are likely to be best positioned to do the teaching. Even the inventors may not be the best instructors. Who then? The meaningful policy answer is whoever. But that doesn’t sound serious enough. The principal investigator then, or experts who understand the discovery and can translate it into the terminology of manufacturing or business or medicine. In any of these cases, the need is not for “technology transfer” professionals hired by institutions to represent their financial interest but rather for regular professionals who have an interest in improvement and change and have access to the discovery and the freedom to practice it and disseminate it.

If patent monopolies are not necessary for health-related inventions, then the NIH does not have to bother with “march-in” procedures under Bayh-Dole. There’s another pathway in Bayh-Dole that would work much better–though it is rarely used. The NIH could determine “exceptional circumstances” under 35 USC 202(a)(ii) and require fair, reasonable, non-exclusive licensing for federally supported biomedical subject inventions, regardless of ownership. In Bayh-Dole, “exceptional circumstances” are not “rare” circumstances–they are circumstances not addressed by the default requirements of the standard patent rights clause. The development of medicines from federally supported discoveries is one such exception. If patent monopolies are not necessary for such development, then why should a federal agency permit those monopolies in the first place? A patent monopoly–if not necessary–then disrupts collaboration, runs up development costs, creates all sorts of legal fuss and financial fuss, and in the end–if anything comes of it at all–the public must pay monopoly prices without any alternative products. The NIH should require FRAND licensing if a nonprofit contractor is going to take an ownership position in a federally supported invention.

There’s actually an argument that if Bayh-Dole were enforced, such FRAND licensing would happen anyway. That’s because Bayh-Dole has special provisions for nonprofits that own subject inventions. Here’s 35 USC 202(c)(7)(A):

In the case of a nonprofit organization, (A) a prohibition upon the assignment of rights to a subject invention in the United States without the approval of the Federal agency, except where such assignment is made to an organization which has as one of its primary functions the management of inventions (provided that such assignee shall be subject to the same provisions as the contractor);

It’s fussy and sloppy but the logic is clear. Nonprofits cannot assign subject inventions without including the same nonprofit provisions. A nonprofit subject invention remains a nonprofit subject invention even if assigned to a for-profit company. The nonprofit provisions include sharing royalties with inventors, deducting from royalties and income earned with respect to a subject invention only administrative expenses for managing subject inventions, and dedicating any remaining money to scientific research or education. One can see that the nonprofit provisions require owners of subject inventions to be involved for the public interest–if they make money from an invention, the money goes to inventors, to other expenses directly related to managing such inventions, and to scientific research or education. Not to lucrative salaries, not to new buildings, not to shareholders. Pretty interesting. Any exclusive license of all substantial rights in an invention–to make, use, and sell–assigns the invention. Thus, universities get away with labeling their contracts as “exclusive license” but in effect they assign the underlying invention. The nonprofit patent rights clause provisions apply to any assignee. If Bayh-Dole were enforced, companies would beg for non-exclusive licenses. But then that would ruin the monopoly meme argument that patent monopolies are necessary or federal research dollars are utterly wasted.

If NIH won’t march-in and won’t determine exceptional circumstances to compel non-exclusive access on reasonable terms, then it’s clear that the NIH is fine with monopoly pricing and with suppression of all other possible uses of NIH-funded health-related subject inventions. In other words, what folks don’t much like–unreasonable terms, including unreasonable price of medicines–is a consequence of NIH’s policy position that patent monopolies are essential to the use of medical research.

Other unreasonable terms that are of even greater concern than price are the refusal to license a class of compounds–a claimed invention–to others for development of compounds that an owner of a subject invention is not developing. Thus, if a company acquiring a subject invention develops one compound out of 1,000 to be a medicine, then it is unreasonable terms to refuse to license any of the other compounds the company has not chosen. No terms at all is unreasonable where a company is not using nor developing a given compound. Bayh-Dole asserts that the patent system is to be used to promote the utilization of inventions arising in federally supported research or development. Suppressing the use or development of compounds runs against Bayh-Dole’s fundamental policy.

Repealing Bayh-Dole, however, won’t alter the NIH’s position on patent monopolies and won’t change the price of medicines. Before Bayh-Dole, the NIH used the IPA program to feed patent monopolies to the pharmaceutical industry. The IPA program was shut down in 1978 as ineffective and against public policy. The NIH produced the Bayh-Dole Act the next year, and Bayh-Dole was in effect by mid-1981. The NIH policy has not been dictated by Bayh-Dole–just the opposite. The NIH has used Bayh-Dole as a vehicle for a policy the NIH already had. Repealing Bayh-Dole would just make the NIH find another way to re-establish a patent monopoly pipeline to the pharmaceutical industry.

It gets worse. Repealing Bayh-Dole would restore the Nixon executive patent policy from 1971 whenever some special statute doesn’t take precedence. But Reagan in 1987 issued an Executive Order that in effect amended the Nixon patent policy to implement Bayh-Dole. So repealing Bayh-Dole ends up back with Bayh-Dole–just now applied *after* any specialty statutes. For biomedical inventions based on federal funding, that means repealing Bayh-Dole changes essentially nothing! The NIH continues with its present contracting and licensing practices.

To get at NIH policy regarding federally supported biomedical inventions, one would have to restore the Kennedy patent policy of 1963 under which, for research directed at public health, the federal government expects to own all inventions  it supports, and the federal government makes those inventions available to all. A private party could own such inventions only if it made the case that it would do a better job than the government in making the invention available to all–and demonstrated that it had done so.

The upshot of all this is that the goal of federally funded biomedical research should be to create platforms for development, not patents for speculation. Patents are valuable in that they define, teach, and announce broadly a given discovery. But in federally supported research, patents should be used to create platforms rather than to fragment platforms into independent, institutionally owned pieces, each held behind a paywall for exclusive licensing (or assignment). The thing of greatest value to the public is that federally supported biomedical research (and inventions arising from that research) accumulates and is developed to the point of practical application–to the point that it can be put to use. The driver for such development is that patients may be better treated, if not cured. While some companies made be motivated only by profits–companies do not have to be so motivated, but that’s another discussion–there are plenty of researchers and nonprofit organizations and state and federal government agencies that are not motivated by profit-seeking or by turning the development of medicines into an excuse to support “economic development” or “research as an industry.” It’s good enough to develop medicines because doing so helps people who need those medicines. There it is, in plain language. Dispute it or mock it if you want.

In collaborative development, inventions (and other things) migrate into platforms (libraries, tool sets, ad hoc standards) and these platforms then are developed to the point of practical application by whoever shares an interest in doing so–federal, university, nonprofit, for-profit, individual. Once a biomedical asset has reached the point of practical application, then companies can compete as they choose by providing the best services or best price or building improvements and applications based on the common platform. The point, then, is that companies must also compete with people making and using and even selling product without concern for either profits or mass-production. The mass-produced commercial product has to be better than what people can do for themselves without the bother of “scaling up” or even defining a product that must be sold.

As it is, since 1968, the NIH has steadfastly sought to circumvent any such alternative to the monopoly meme and patent pipelines to the pharmaceutical industry. And more so, the NIH has sought to impose the monopoly meme on *all* federally supported research–first by attempting to expand the IPA program to all agencies (that attempt was blocked) and then by Bayh-Dole (which succeeded).

As it is, federal policy now is that public money should subsidize patent monopoly creation for for-profit companies. Bayh-Dole merely provides that nonprofit administrators can get a tiny piece of the action through licensing patents–just enough so they won’t complain. One would think that this is a World Trade Organization issue–funneling government subsidies to an industry, using universities as a cover and justifying it with the spurious claim that without monopolies discoveries won’t be used or developed and won’t benefit the public. Maybe WTO folks are just as easily bamboozled as Congress, or want to be.

Put it another way, if federal policy were not that federal money should subsidize the pharmaceutical industry (and anyone else supporting the patent pipeline–speculative investors, the biotech industry, university administrators), federal agencies would under Bayh-Dole determine exceptional circumstances for most federal grants and end university assignment (via exclusive licenses) of biomedical subject inventions to companies. Or–sigh–federal agencies would just enforce Bayh-Dole on nonprofit assignments (see 35 USC 202(c)(7)(A)) and companies would *beg* for non-exclusive licenses.

Again, federal policy is not only to use federal money to subsidize the biomedical “industry” but also not to enforce Bayh-Dole’s public protections. Bayh-Dole does not require federal agencies to enforce its standard patent rights clauses. In fact it gives them all sorts of enticements to waive or ignore enforcement. And no one enforces Bayh-Dole on federal agencies. And the public has no right of appeal.

The problem is not Bayh-Dole. The problem is the monopoly meme that fronts for a substantial flow of public money to serve the interests of private industry. That industry then exploits public suffering to harvest even more money from governments and nonprofits aiming to help people suffering from illness and injury. Nonprofits could break the cycle, but they are addicted. Federal agencies could break the cycle, but they are the ones that created it. Researchers could break the cycle, but it takes someone special to do that–someone with enough status and financial backing to make university administrators and federal agency officials back down. Until that happens, it is up to NGOs and private individuals to press for change.

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