Georgia on my mind

At the University of Georgia, the Office of the Vice President for Research has a bizarre reading of the Bayh-Dole Act:

The Bayh-Dole Act, passed in 1980, makes it possible for the federal government to assign its patent rights to universities, which then have the obligation to pursue intellectual property protection and commercialization. Prior to Bayh-Dole, only a handful of universities had technology commercialization offices.

I have highlighted some text to consider.

federal government to assign. Yes, it’s in Bayh-Dole, but only with regard to inventions made jointly by contractor personnel and federal employees. Here’s 35 USC 202(e):

(e)In any case when a Federal employee is a coinventor of any invention made with a nonprofit organization, a small business firm, or a non-Federal inventor, the Federal agency employing such coinventor may, for the purpose of consolidating rights in the invention and if it finds that it would expedite the development of the invention—

1) license or assign whatever rights it may acquire in the subject invention to the nonprofit organization, small business firm, or non-Federal inventor in accordance with the provisions of this chapter;

The government can also acquire the contractor’s personnels’ rights, but only in a voluntary agreement:

(2) acquire any rights in the subject invention from the nonprofit organization, small business firm, or non-Federal inventor, but only to the extent the party from whom the rights are acquired voluntarily enters into the transaction and no other transaction under this chapter is conditioned on such acquisition.

Notice–the rights may come from the contractor or the inventor. How could that happen if the rights vested with the contractor automatically? And why would the federal government need to acquire from the inventor if the only pathway for an inventor to obtain rights was for the university to waive its rights back to the federal government, which then allowed (in some magical way) the rights to land with the inventor, who then voluntarily agreed to assign them back to the federal agency. No, that’s not it.

But the Georgia text does not appear to be about this isolated condition of joint ownership. It makes Bayh-Dole appear to operate generally by the federal government obtaining outright ownership of all inventions made with federal support, and then assigning this ownership to universities. That is, in the Georgia version of faux Bayh-Dole, title to inventions vests with the federal government, which then assigns title to universities, apparently through the operation of the university notifying the government that the university elects to retain title. Of course, “retain title” here means something more like “retain counsel”–acquire and keep, have and hold, not keep something one already has.

The impression left with a non-specialist reader is that Bayh-Dole gives title to inventions to the universities that host federally supported work. And this is, of course, not true. While I expect Georgia to correct their text immediately, you may expect otherwise.

the obligation. It is true that if a university, having acquired ownership of a subject invention–one made with federal support within the scope of a funding agreement for research–elects to retain title, then the university does have an obligation to file a patent application on the invention. But the obligation does not arise because there’s an invention, and certainly not because the federal government has assigned its ownership interest (or presumption of title, or whatever) to the university. The obligation arises because the university has made a choice to own the invention, and that choice precludes (by law) the federal agency from making a choice whether to acquire the invention (under its regulations) and if so, whether to seek a patent or dedicate the invention to the public domain.

In the IPA scheme, the situation was much clearer. A subject invention was any invention within scope of the funding agreement. *If* the university decided to file a patent application, *then* it must obtain assignment of the invention, and after that it must abide by the public covenant established by the master IPA.

Bayh-Dole effaces the conditional of the obligation by inserting “of the contractor” as a restriction in the definition of “subject invention”–so for an invention to be “subject” it must already be owned by a contractor. That would mean for any invention made with federal support that it was not subject to the standard patent rights clause until a contractor found a way to own it. This presented a huge puzzle to university administrators and their attorneys who didn’t bother to read carefully the implementing regulations or the standard patent rights clause.

They decided that Bayh-Dole must be a vesting statute, and that every invention made with federal support was owned by the contractor, and that’s what “of the contractor” meant–it was just a confirmation of what the law presumed. But that meant “of the contractor” was superfluous–it added nothing to the definition. The Supreme Court ruled that as a matter of good interpretation of statutes, one should read each part of the law to have meaning, and so long as that meaning was reasonable, it should stand. Words should mean something. It’s basic, I know.

The government resolved the puzzle of “of the contractor” by placing in the standard patent rights clause the (f)(2) requirement. (f)(2) requires the contractor (the university) to require that its employees make a written agreement to protect the government’s (not the university’s) interest. That requirement works like a conditional substitution of parties, parallel with the requirement that follows subcontracting, under which the subcontractor stands in as a new, independent contractor and the contractor is prohibited from claiming any invention made under the subcontract as the condition of the subcontract.

In a similar way, (f)(2) requires the university to delegate to research personnel a conditional standing under the funding agreement if they should make an invention. (f)(2) does not permit the university to make any claim to own inventions as a condition of the funding agreement–whatever arrangements that might be made must take place outside the fact of federal funding. Indeed, the (f)(2) agreement takes precedence over any other agreement that the inventors might have with the university, both because the university requires the (f)(2) agreement (and therefore cannot simultaneously require something else) and because the (f)(2) agreement is part of a federal contract (which supersedes any state-enforced contract).

The result of (f)(2)–provided it is complied with–is that research personnel working within the scope of a federal funding agreement become parties to the patent rights clause–and therefore parties to the funding agreement itself. Under the definitions of funding agreement and contractor, that makes these research personnel also contractors. Since inventors own their inventions when they make them, and research personnel making the (f)(2) agreement are contractors, then any such invention they make is “of the contractor” and is necessarily a subject invention. It’s just that there are multiple contractors–inventors and the university. What Bayh-Dole and the standard patent rights clause do not address is how a university-contractor acquires title from inventor-contractors, other than that it appears that the university-contractor cannot make the fact of federal funding the condition of assignment.

The government’s interest–and what Bayh-Dole can address–is what rights the government gets. Bayh-Dole is directed at government agencies and at restricting the scope of property rights under patent law in subject inventions. Bayh-Dole is not directed at the universities. Whatever gets to universities comes through the patent rights clause in each funding agreement, as a matter of federal contract. Thus, the (f)(2) agreement is not merely a superfluous restatement of the operation of law. If a university does not comply with (f)(2), then the government’s rights are not properly protected. If a university uses a different method to attempt to protect the government’s interest, such as a comprehensive present assignment to claim ownership of all inventions upfront, then the university abuses its position as prime contractor. I know, I know. <sarc> No one cares whether a university abuses its position because institutional control of inventions for profit-seeking is clearly the best way for innovation to come about. </sarc>. It’s as if the standard patent rights clause says, “Do this or do anything else that’s convenient and self-serving.” Sure.

Under the IPA, all this too was much clearer. University patent policies and practices were reviewed and approved for their procedures, agreements with Research Corporation and other invention management organizations, and performance. Universities were required to have all research personnel promise to assign subject inventions to the university *if* the university decided to patent their inventions.

Bayh-Dole, as federal patent law applied to federal agencies rather than operating as a master federal contract at the discretion of a federal agency within the constraints of an executive order, is not positioned to tell inventors that they don’t own their inventions because federal contractors do. So Howard Bremer and others had to create a story about how it all worked: that the federal government owned those inventions, but released its interest to contractors when the contractors wanted to own, and if the contractors didn’t want to own, then the federal government could release its interest to the poor, silly, goofus inventors, as long as it was okay with the contractors. Except it doesn’t work this way. Even the definition of subject invention fails. The federal government does not have the authority to reserve for limited times the exclusive rights in inventions to federal contractors. Just to inventors. And it’s not possible to make a contractor into an inventor and the inventor into just someone who builds a “best mode.”

So Bayh-Dole couldn’t do in the form of federal patent law what the IPA could do as federal contract. That was one of the tradeoffs in trying to replace executive policy with federal patent law. The best they could do was to postpone the actual patent rights clause to the regulations and make it as ambiguous as possible what happens to title to inventions made with federal support. Clarity is not a design feature of Bayh-Dole.

If there is an obligation under the standard patent rights clause, it is that once a university has decided to own a subject invention, then it must follow through on that decision and make the effort to achieve practical application. That is, the obligation is to run with the choice. The obligation is a consequence of institutional choice, not of either law or federal contract.

pursue intellectual property. “Pursue” is a funny verb here, but no matter. The problem is “intellectual property.” Yes, patent rights are a form of intellectual property. But Georgia properly should write “pursue patents” or “file patent applications.” Bayh-Dole is part of federal patent law. It does not cover other intellectual property except for the plant variety protection that the Bayh-Dolists slipped in alongside plant patents (amazing that they didn’t include mask works, say, too). The Georgia presentation, however, makes it appear that Bayh-Dole claims intellectual property generally–not just patents and plant variety protection–and not just statutory intellectual property (patents, copyrights, and trademarks), but pretty much anything that anyone might call intellectual property when they mean intangible or tangible asset arising in research, regardless of whether it can be owned, and hence “invention, whether or not patentable.”

Here’s Georgia’s Intellectual Property Policy on the matter:

Such creative and scholarly works and inventions that have commercial potential may be protected under the laws of various countries that establish rights called “Intellectual Property,” a term that includes patents, copyrights, trade secrets, trademarks, plant variety protection, and other rights.

The “other rights” is useful in a definition that can’t bring itself to define anything. Here’s their definition:

“Intellectual Property” shall be deemed to refer to patentable materials, copyrighted materials, trademarks, software, and trade secrets, whether or not formal protection is sought.

Note the “whether or not” bit. Here’s their definition of “software”:

“Software” shall include one or more computer programs existing in any form, or any associated operational procedures, manuals or other documentation, whether or not protectable or protected by patent or copyright. The term “computer program” shall mean a set of instructions, statements, or related data that, in actual or modified form, is capable of causing a computer or computer system to perform specified functions.

That is, software–whether or not protectable–and trade secrets, which again may not be patentable or protected by copyright–and everything else “whether or not formal protection is sought.” What “informal” protection is, I have no idea. But it’s clear that Georgia uses “intellectual property” to mean something much broader than patentable inventions and novel plant varieties, and broader even than statutory intellectual property. But Georgia ascribes its ownership of all such stuff to the work of Bayh-Dole. Deceptive? Sloppy? Ignorant? Self-servingly clever? Your call. No matter, pick any one option and you’ll be right!

commercialization. The mandate in Bayh-Dole is practical application. That’s what gets a definition:

The term “practical application” means to manufacture in the case of a composition or product, to practice in the case of a process or method, or to operate in the case of a machine or system; and, in each case, under such conditions as [1] to establish that the invention is [2] being utilized and that [3] its benefits are to the extent permitted by law or Government regulations [4] available to the public [5] on reasonable terms.

I have added some numbers to make clear the parts of this definition. The goal of Bayh-Dole is met when inventions are used in a non-secret way with demonstrable public benefits on reasonable terms. No mention here of commercialization. Despite papers to the contrary, there is no implied obligation to commercialize, just as there is no presumption of ownership. Certainly commercialization can meet the objective of practical application, but is no where near to the only way to do so, and is not an obligation of Bayh-Dole or the standard patent rights clause. Even the march-in procedures do not consider commercialization.

In any event, commercialization on its own is not good enough. Practical application–even in the form of commercial products–must be such that the benefits of the use of the invention are available to the public on reasonable terms. Reasonable terms are not just any terms that a licensor or seller thinks are reasonable. There would be no need to state a restriction to reasonable terms in the law if that were the case. So reasonable must mean something that would limit what a licensor or seller–in the case of commercial activity–could otherwise have the power (and rationalizations) to do.

There’s no definition of “reasonable” in Bayh-Dole. That’s perhaps by design. In the IPA, “reasonable” is given an operational definition. A university deciding to file a patent application must

license non-exclusively OR

license exclusively only after demonstrating that non-exclusive hasn’t or won’t work AND limit exclusive licenses to three years from first commercial sale or eight years from the date of the license, whichever comes soonest


achieve practical application by three years from patent issue

OR ELSE the government can intervene

AND the government can intervene anyway on 30 day’s notice with a hearing but no appeal whenever it believes the public welfare is better served, when use of an invention is necessary to comply with federal regulations, or to meet a public health or safety need.

That’s “reasonable.” The Bayh-Dole march-in provisions back federal agencies away from intervention. Bayh-Dole retains the general categories, but ties them to availability of products, not simply to meeting a need under a patent right (35 USC 203(a)):

(1) action is necessary because the contractor or assignee has not taken, or is not expected to take within a reasonable time, effective steps to achieve practical application of the subject invention in such field of use;

Notice the waffle around “reasonable time…steps to achieve” rather than simply whether there is practical application or not. “Reasonable time” is now directed at the steps rather than the outcome. That is, the cause of action is that the contractor did not in a reasonable time do something effective, not that there’s nothing meeting the definition of practical application.  And there’s now no time limit. Reasonable is whatever. One could be 18 years into the term of a patent on a subject invention, and having immediately put the invention on a “tech available for licensing” list, one may have taken an effective step in a reasonable time. Who would know?

(2) action is necessary to alleviate health or safety needs which are not reasonably satisfied by the contractor, assignee, or their licensees;

Here’s another use of “reasonable” that’s not directed at “reasonable terms.” Here reasonable has to do with a determination of whether the “needs” are “satisfied.” There is no consideration of, say, the terms (such as, say, cost). Desperate folk will pay most anything to meet their health or safety needs. And those needs may therefore be “reasonably” satisfied, but that’s not the point of “on reasonable terms.” March-in has been so skewed that agencies cannot address issues of pricing or tying or discriminatory licensing or excessive royalty requirements–only whether people can obtain access to the invention. You see how they screwed over the public covenant using almost the same wording?

(3) action is necessary to meet requirements for public use specified by Federal regulations and such requirements are not reasonably satisfied by the contractor, assignee, or licensees; or

Again, reasonable is displaced from terms to availability, regardless of the patenting practices at work–price, tying, discrimination, royalty demands. March-in cannot address gouging or monopoly abuse. Perhaps anti-trust law can, but that’s not via agency march-in. Thus, when there’s an immediate need, there’s no way for the government to intervene. Any march-in procedure will be so long that the issue will be academic long before it is resolved in favor of public access on reasonable terms.

(4) action is necessary because the agreement required by section 204 has not been obtained or waived or because a licensee of the exclusive right to use or sell any subject invention in the United States is in breach of its agreement obtained pursuant to section 204.

And a bit to allow march-in if there’s a failure to be properly protectionist. The funny thing here is that federal agencies are given the right to waive dear section 204 requirements, so there’s no particularly reason why an action would be “necessary.” Clever folks, those Bayh-Dolians. And even if there’s non-compliance here, there’s nothing that considers whether the benefits of the subject invention are available to the public on reasonable terms. If a company manufactures abroad, that’s not a concern for practical application, even if it is for a protectionist view of American workers. With H-1B visas and other immigration programs, it is entirely possible that workers doing the manufacturing in the U.S. aren’t even U.S. citizens. So much for protectionism. It’s a strange clause anyway, relative to what used to be the federal patent commons, which had to do with control of *imports* rather than a demand for domestic manufacturing.

The Georgia statement makes it appear that commercialization is the only route, even when it is not. Even if a university decides to attempt “commercialization,” company involvement in making product for sale does not preclude other ways of achieving practical application. Commercialization may come about in parallel with local use, for instance. It may be that any capable person can use the method without the need for a commercial product version. I’ve invented a saw. You can make a saw. It might not be as nice as my product saw. But then I have to offer my product saw to you on terms that make it reasonable for you to buy my saw rather than go to the bother of making one yourself. That’s reasonable terms talk. That’s different from the normal monopoly patent that allows for any terms other than those exploiting market power. Sell to the rich and desperate and loyal first. Then work down to everyone else when the patent expires.

Commercialization may in fact depend on widespread local use happening first. Everyone is using this method of assembling genetic sequences–we could sell them a product that standardizes the method and makes everyone more efficient. First the use–practical application–and then the commercial version. Similarly, commercialization may depend on the formation of standards–ad hoc or formal. The internet’s technologies became standards, and vast commercial opportunities were created as a result. It is a lesson that comes hard, apparently, in the university patenting community. Patents, for them, are weapons which must be used as weapons, or at least waggled threateningly as weapons. What a lousy way to run a restaurant.

One more thing about university use of “commercialization” as a substitute for “practical application.” For universities, “commercialization” means “attempt to license for a price or other financial upside.” Or, “start a chain of speculation from which to profit from.” All those metrics about inventions, patents, licenses, startups, investment in startups, and income have nothing to do with practical application or commercialization–that is, universities do not report the number of products for sale, the extent of availability of those products, or the cost to acquire and use those products relative to alternatives–i.e., establishing that the products are available on reasonable terms.

None of that. Just success stories–and you can assume that the success stories reported in university annual reports on technology transfer are *all they have*. They do not stand for a vast number of similar untold success stories. Further, if you read those success stories carefully, almost all of them have to do with the fact of a license and potential for public good, rather than establishing practical application as expected by Bayh-Dole. The reality is, commercialization in the form of products available to the public on reasonable terms is on the order of 1 in 1,000–and maybe much worse than that. It’s a sad story that universities refuse to report directly. And Bayh-Dole excludes reports of use from federal public disclosure law. So we won’t know.

If universities were *truly* committed to commercialization on reasonable terms, they would not accept any payment until that payment was based on a royalty from sales. That would make university administrators take an interest in what happens after an exclusive license is granted, and might even get them to rethink the idea of limited the term of exclusivity so that if the first licensee doesn’t get after it, others will. Or, even better, allow local use immediately and license only for the sale of commercial versions. Then one gets to practical application as quickly as people can learn and do. And of course if learning and doing is too complicated and expensive for anyone to accomplish on their own, then we will all await the commercial version as the only viable version. But university administrators are not *truly* committed to commercialization. Speculation is good enough. If some investors can weedle money from other investors–or better a company purchases their startup to get it out of the way or to harvest its engineers–then the university and first investors make money, regardless of whether there is every practical application. Nice. But not what Bayh-Dole is about. Bayh-Dole is not a moral mandate for universities to seek profits from monopoly positions in patents. Except that is what it has become, and that’s what the Georgia statement turns into a virtue.

technology commercialization offices. The Georgia folks are lying by telling a truth. It’s a clever technique. Before Bayh-Dole, most universities did not have “technology commercialization offices.” That’s because nearly all universities designated one or more invention management organizations to deal with patents. The universities did not have to have staff, finance, and operate such offices. A network of affiliated agents took on that work. By the time of Bayh-Dole, over 70 universities had IPAs with the federal government–and so were involved in dealing in patents, if only to require their inventors to assign to a designated invention agent. A number more had “technology transfer offices” set up to help faculty identify inventions and get these sent off to Research Corporation (or another agent) for review. And the offices that did exist–notably at MIT, UC, and Stanford–were technology licensing offices, not “commercialization” offices.

Bayh-Dole does not require that a university have a “commercialization” office, or the capability to license patents, or a patent policy, or a patent agreement with employees, nothing. Just designate someone to receive invention disclosures and pass these to the government, educate employees about the value of timely reporting inventions, and require them to make an agreement to protect the government’s interest. Pretty much done. A university can check the box “not interested” on every invention report and leave it to inventors to work things out with the federal agency that funded their work–that is, if the inventors can’t find an invention management agent to partner with.

The Georgia statement makes it appear that Bayh-Dole brought into existence a necessary resource to bridge university research and economic activity–commercial products as the means by which the public benefits from university work. Sure, that’s one way among others. But Bayh-Dole was not responsible for this new university activity. Bayh-Dole destroyed the IPA system. And faux Bayh-Dole destroyed the private system of invention management. The new bureaucratic approach is about 100x less effective than the system it destroyed. But it creates lots of opportunities for university administrators to make a good living.

There are plenty of reasons for a university to provide services to faculty and students who have discovered or invented something and want to make that something available beyond announcing it in publications and showing it off in labs or on web sites. That part makes sense, regardless of Bayh-Dole. It’s part of the conscience of a university community, if nothing else. People will help because they want to, not because they are required to by federal law or because they will only act if they have a money incentive offered in recognition of commercial products but in reality mostly paid by speculators buying their way into a potential revenue scheme that may be speculators all the way down. So having resources–call them technology transfer, industry liaison, technology licensing, economic development, early stage ventures, even commercialization–is important. But to make it appear that Bayh-Dole is the reason, or that commercialization is the only thing that matters–that’s just deceptive.

When Georgia IP policy guidance is on your mind, think “deception.”

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