The NIH’s complicity in faux Bayh-Dole and high drug prices

Here’s “A ’20-20′ View of Invention Reporting to the National Institutes of Health”–published by the NIH in 1995.


The Bayh-Dole Act encourages researchers to patent and market their
inventions by guaranteeing patent rights.  This Act automatically
grants first rights to a patent for an invention fully or partially
funded by a Federal agency to the awardee organization.  To obtain
these benefits, however, the inventor and the organization have
several reporting requirements that protect the rights of the

What’s wrong with this answer? First Bayh-Dole says nothing whatsoever about whether researchers should “patent and market” their inventions. Bayh-Dole applies to federal agencies, requiring them to use a standard patent rights clause created (now) by the Department of Commerce (delegated to NIST). And Bayh-Dole creates “subject” inventions, a new category of patentable invention in federal patent law, and states a policy that governs subject inventions. There’s nothing in Bayhj-Dole about researchers, for that matter. The basic definition is “contractor” and the reference made to anything close to “researchers” is the “employee/inventor” stuff of 35 USC 202(d)–when inventors can request “retention of rights.”

The fun stuff comes next. The researchers might be encouraged to patent and market their inventions, but according to the NIH, Bayh-Dole doesn’t even do that–it takes inventions from researchers before they ever have any rights in them:

This Act automatically grants first rights to a patent for an invention fully or partially funded by a Federal agency to the awardee organization.

Well, that’s totally wrong. The Supreme Court made that clear in Stanford v Roche in 2011. Not that it wasn’t blazingly obvious to anyone who took the time to read the statute, even in 1995. But advocates for Bayh-Dole told a different story–the faux Bayh-Dole–and in that story, Bayh-Dole changed federal patent law so that inventions are owned not by their inventors (as provided by the U.S. Constitution) and not even by their employers, but by the institution that receives and manages the federal award on their behalf.

Supreme Court ruling:

Nowhere in the Act is title expressly vested in contractors or anyone else; nowhere in the Act are inventors expressly deprived of their interest in federally funded inventions.

The Bayh-Dole Act does not confer title to federally funded inventions on contractors or authorize contractors to unilaterally take title to those inventions; it simply assures contractors that they may keep title to whatever it is they already have

Only when an invention belongs to the contractor does the Bayh-Dole Act come into play. The Act’s disposition of rights—like much of the rest of the Bayh-Dole Act—serves to clarify the order of priority of rights between the Federal Government and a federal contractor in a federally funded invention that already belongs to the contractor. Nothing more.

Yet the idea that somehow the purpose of Bayh-Dole was to disenfranchise inventors from their inventions persists–that neither the government nor inventors should have a claim on inventions unless university patent brokers decided they didn’t want to keep ownership. It’s inventor loathing and open innovation loathing all wrapped up in a ball of patent monopoly happiness. Universities, so the argument goes, should embrace the assertion of monopoly rights and trade on that assertion to make as much money as they can–money so needed for research, for research to help the public, for research that will save dying children. Except universities don’t even bother to track their royalty income, let alone report what they’ve done with it, other than to from time to time note that it amounts to a tiny percentage of the overall expenditures on research each year.

Once you have strung these threads together, it’s easy to see how the mindset develops:

Universities should trade on patent monopolies to ensure commercial development of inventions.

Commercialization links federal research dollars with public benefit.

Universities should trade on patent monopolies to generate money for more research.

All this is in the public interest.

Patent brokers should have first dibs on inventions made with federal funding.

They will do a better job than either federal agencies or inventors.

Any other approach means that inventions will not be developed, the public won’t benefit, universities won’t have more money for research, and the federal research dollars will have been wasted.

If we put this in the euphemistic language of faux Bayh-Dole, it looks like this:

Universities have a mandate use the patent system to promote the commercial development of inventions made with federal research support. [not from Bayh-Dole]

University technology licensing officers identify inventions, file patent applications, and market inventions to industry for commercial development. [only after they strip inventors of their rights via policy and adhesion contracts]

Any income from commercialization is shared with the inventors and dedicated to the support of more research. [but first used to expand the patenting program and then used as administrative slush funds]

This approach has been a wild success, creating many new and beneficial products and companies, and especially creating the biotech industry. [actually, there’s no public evidence that the approach has been anything other than a wild disaster for most research findings; even the few successes mostly come despite this approach, not because of it]

Or, in the words of the NIH:


Patent protection gives the owner of the patent the right to exclude
others from making, using, offering to sell, selling, or importing
into the United States the invention during the lifetime of the
patent, thus protecting the incentive for commercial development of
the invention.  However, it does not give the public the right to use
the invention if it is claimed by another's patent.  A company will
be more willing to make the investment needed to commercialize an
invention if it can eliminate or decrease competition.  When a patent
is licensed and successfully commercialized, it can lead to royalties
for the organization and the inventor, economic development for the
Nation, and improvements in the public health.

Patent protection is the key component of technology transfer.  Of
the legal options available, including trademarks, trade names,
copyrights, and licensing, patenting is probably the most crucial to
commercializing research results.

While it is true that some companies–pharmaceutical companies, say–“will be more willing” to commercialize an invention if they can “eliminate or decrease competition,” the statement is not at all true in general. Many companies are fine with making investments in competitive environments. Indeed, the whole idea of competition in innovation is based on the idea that others will invest to copy a new product–and without a monopoly position. The narrow worry is that if the expense to develop is huge, and the cost to copy is low, then it’s not “fair” that a private investor bears the cost but does not reap the entire benefit. And even here, the fallacy is that a private investor has to bear that huge cost entirely. It’s almost as if there are investors who *insist* on taking on the entire cost and *insist* on controlling the entire benefit, even when the entire cost is only a tiny portion of the entire benefit, and even when the entire benefit means two decades of monopoly pricing rather than competitive pricing.

The truly awkward thing here, though, is the NIH’s implicit endorsement that the purpose of Bayh-Dole is to allow companies to “eliminate or decrease competition.” That’s basically antitrust stuff. And it runs directly against the policy stated by Bayh-Dole (35 USC 200):

to promote free competition and enterprise without unduly encumbering future research and discovery

But none of this commercialization-by-monopoly apparatus is necessary, and none of it arises in Bayh-Dole. Bayh-Dole is noncommittal with regard to university ownership, commercialization, income, and the like. Its concerns are practical application, free competition, and US manufacturing, with a broad government license to practice and have practiced any invention. But you won’t read about such things in most any “guides” to Bayh-Dole. But for the Supreme Court–you will have to look long and hard to find any account of Bayh-Dole that doesn’t repeat the faux version of the law, the love monster version, the inventor-loathing/commons-loathing version of the law. Even most criticisms of Bayh-Dole start with an acceptance of the faux version and then argue that people aren’t being smart enough about how to turn research into public benefit and economic vitality.


The first step, if the inventor believes he/she has an
invention/discovery, is to report it promptly to the organization's
technology transfer office, the office of sponsored research, or the
institutional administrative official responsible for technology
issues.  The employee/investigator is required to report any
invention in accordance with the terms of the employee agreement
he/she signed.  (Note: Organizations are required, as a condition of
Federal funding, to enter into employee agreements with all
appropriate staff.)

This advice is also wrong. Bayh-Dole does not require an “employee agreements.” The standard patent rights clause introduces a written agreement requirement (that’s the (f)(2) agreement)–but that agreement is not in Bayh-Dole and isn’t an “employee agreement.” Instead, it is part of the funding agreement, a flow-down of responsibilities that only an inventor can have to protect the federal government’s interest in inventions made with federal support.


The invention belongs to the awardee organization.  The Bayh-Dole Act
requires that there be employee agreements in place at the awardee
organizations that obligate inventors to assign title to Federally-
supported inventions to the organization.

Here the NIH repeats its advice and makes it clear that Bayh-Dole requires employee agreements. This is simply false. Bayh-Dole does not require employee agreements. The Supreme Court ruled, as well, that Bayh-Dole applies only after a contractor acquires rights in an invention made with federal support. The NIH ought to issue a retraction for this guide and for anything like it that the NIH has published since. But since the NIH is complicit in the faux Bayh-Dole from the start, don’t expect the NIH to do anything of the sort.

Following the “20 questions” are “words for the wise,” including:

1.  ASSIGNMENT.  Transfer of title or ownership in patent rights in
the form of a written assignment document.  By law, an inventor has
initial ownership of an invention.  However, awardee organizations
are required by the Bayh-Dole Act to have in place employee
agreements requiring an inventor to "assign" or give ownership of an
invention to the organization upon acceptance of Federal funds.

Again, so false. Assignment in Bayh-Dole is directed at subject inventions, not patents (35 USC 202(c)(7), enumerating requirements for a standard patent rights clause:

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);

Assignment of inventions may happen by formal transfer or by conveying all substantial rights in the invention, such as by means of an exclusive license to make, use, and sell. One sign that an assignment has been made is that the licensee has the right to enforce the patent or settle claims of infringement. And there’s nothing at all in Bayh-Dole about requiring an inventor to “assign” inventions. Assignment to the research host is not even a requirement of the (f)(2) agreement. The inventor agrees to sign papers to “establish the government’s rights in the subject inventions.” Not the university’s rights. The wrongness of the NIH’s “guide” is negligence, malpractice (to the extent any NIH legal counsel was involved), and even fraud, if deliberate. But what’s most dismaying is how happily casual the NIH is about presenting the idea that inventors must give up ownership of their inventions to their host institutions, simply because they accept federal support. Why should the government be so blithe about institutional control of inventions?

Why are these ideas at the heart of Bayh-Dole? Inventor-loathing. Commons-loathing. University administrators as controllers of innovation. Universities as monopoly traders and patent trolls. Why? What makes these sorts of behaviors desirable? Why would these behaviors be tolerable even if the approach was somehow a wild success?

There’s a simple answer. The pharmaceutical industry in the U.S. has been built on monopoly claims. Pharma could not permit the federal government to sponsor research in the area of medicinal chemistry and make the results available to everyone. Nor could it allow the federal government to step in and develop cures for diseases–malaria, leukemia, polio, infections–without allowing first a “market” in those diseases. The government’s commons approach took ownership of inventions to break up patent monopolies. The government dedicated inventions to the public or licensed patented inventions non-exclusively to any qualified company. The purpose of the federally managed patent was to ensure that only those manufacturers who committed to producing high-quality medicines with proper safety and dosage and training requirements would enter the market–that is, companies pretty much unlike the pharma companies of the time.

The Public Health Service patent policy contemplated companies stepping in to serve markets for medical products after the research had revealed possible therapeutic compounds (and classes of compounds) that might be considered for use. This meant that the market available was one that we would call “generic” today–companies competing on production efficiency, formulations, and availability. There would still be profits to be had, but not monopoly profits. The market for medicines would be a boring profitable market, not a volatile, crazily lucrative market. It’s that vision of monopoly markets to cash in on remedies for human suffering that pharma fixated on–or, better, the investors behind pharma. And that’s what could not be tolerated in the PHS patent policy.

When the PHS required pharma companies to sign a patent agreement that they would not obtain patent rights merely by screening PHS-discovered compounds for therapeutic effects, the pharmas refused and boycotted PHS-discovered compounds. That boycott led to the NIH mutiny to PHS policy, in the form of a revived and revised Institutional Patent Agreement program. The IPA program had all the trappings of complying with the Kennedy executive branch patent policy and with PHS expectations of non-exclusive licensing, except the IPA program didn’t comply with either and created a pathway by which patents on federally supported medical inventions could be taken out by nonprofits and licensed exclusively to pharma companies. And that’s what the nonprofits did–almost always an exclusive license, and at licensing rates 1/5th that of the federal government’s own efforts. But this, too, was called a success.

When Congress blocked the expansion of the IPA program to government-wide and the PHS examined the IPA program and shut it down, the NIH mutineers (Norman Latker, in particular) created the Bayh-Dole bill and were lucky and clever enough to get it passed by Congress, with the help of the university patent brokers, led by Howard Bremer of WARF. Universities would be good stewards of patents, managing them in the public interest, on behalf of the federal government. University patent brokers would do the public’s work, but do it better than the federal government could. That was the claim, and it was untrue at the time, and is untrue now, and in any case imagines a dichotomy between federal “commercialization” and private “commercialization” that’s meaningless, other than in the context of the pharma industry fighting to preserve its future investments in monopolies on all the compounds that might be the next wonder drug or cure.

High drug prices are the result of this effort to create Bayh-Dole, make it look like the IPA program, with the various gestures in the public interest–use the patent system, achieve practical application, benefit the public, do all this without a profit motive by dedicating all proceeds (after costs) to more research in the public interest. Except that really what was constructed was a way for pharma to get ownership of inventions made with federal support for about 1% of net revenues from future sale. Universities happily constructed patent agreements labeled “exclusive license” that were assignments. Pharma paid the university’s patenting costs (as if the university had obtained the patents on behalf of the company). Pharma got the right to enforce the patents. Pharma got all the substantial rights in each invention–to make, use, and sell (and therefore to prevent all other making and using, not for the particular compound a company happened to develop, but for all other compounds in the class of compounds claimed by the patent–hundreds to thousands of compounds kept from research and development).

To cover this arrangement, Bayh-Dole was made government-wide and called a “uniform” policy, even though it is hardly uniform in its implementation and even though the policy it displaced was “uniform.” Bayh-Dole is, rather, arbitrary. For companies contracting directly with the federal government, Bayh-Dole is merely a bother. For nonprofits, however, Bayh-Dole has been devastating, exporting to all industries the patent mentality of nonprofits complicit in exploiting the pathway to big pharma (at first) and then developing it into a pathway to biotech (which then may be acquired by big pharma). What has been consternating to university administrators is that so few companies in other industries want to e complicit in the trade in patent monopolies. For most companies in most industries, as the Harbridge House report in the mid 1960s documented, patents just aren’t that important, aren’t used to create monopolies–rather, they are used to define standards, to trade to shift competition, to mark areas for industry roadmaps, to claim priority, to counter the use of patents by interlopers and trolls and folks looking to exploit monopoly positions.

Thus, university patent brokers turn out to be like drug dealers working Christian high schools. You can get a few folks addicted and sell some drugs, but you are working against a mindset that says “what you offer is not that important.” That in essence is what the present university technology licensing agenda is all about. Behind all the talk about the public interest, commercialization, and money for research, the practice has remained remarkably stable from 1968, when the IPA program was reborn as a rough beast–create monopolies, trade on monopolies, sell the public on the necessity of monopolies. Heck, Bayh-Dole even set up the federal government to do the same thing (at it chose) by making exclusive licensing = assignments allowable by statute–the government could then issue patents to itself and assign the rights to whatever company it favored. This behavior exploits the patent system by, in essence, precluding it. Instead of inventors applying for patents and seeking to develop their work, inventors are stripped of rights and the patents–which according to the Constitution should go to inventors–go instead to the government and from there to whatever company the government chooses. That’s the patent system that operated in the states before the union, and in England before patent reform, where a patent was a royal favor and possible source of revenue by creating a monopoly enforced by decree. That’s the patent system Jefferson didn’t want, and so patents in the U.S. issue only based on inventive merit and a willingness to publish the details of the invention.

But that’s not the patent system as exploited by Bayh-Dole. Instead, the idea is to end-run the patent system, to end-run academic research as independent inquiry, to end-run federal policy to create a robust public domain and control the quality of companies competing to provide medicines to support public health. That’s what we have. It starts, strangely, in the NIH seeking to provide monopoly positions to pharma, takes in university patent brokers as the mechanism, and then uses the claim that federal law mandates university ownership, commercialization, monopolies, and money for research to persuade university administrators to sell out faculty research. Thus, we have the system. Thus, any proposal for anything different is shouted down. Thus, we have no reporting data on actual outcomes. Thus, we have a disaster for public health. Thus, we have the highest drug prices in the world. That’s the legacy of Bayh-Dole, and the legacy of the NIH.

There are many ways to approach the movement of research finding to new practices, products, and industries. Even Bayh-Dole technically permits other approaches. But really, not. Even if Bayh-Dole were repealed tomorrow, the damage would persist in awful university patent policies, changed to “comply” with the law, with university licensing operations dedicated to monopoly practices and hired for that purpose and trained for that purpose. Bayh-Dole has taken on the attributes of prophecy, of a religious sect devoted to public benefit through monopoly patent licensing for “commercialization” of research inventions. Any challenge to Bayh-Dole–reason, evidence, history–results in doubling down, repeating the faux version of the law, not conversion or moderation or diversification or competition.

But, wow, it’s time for a change.

This entry was posted in Bayh-Dole, History, Stanford v Roche. Bookmark the permalink.

Leave a Reply