Senator Nelson on the problem of “public interest” in federal patent policy, 2

The federal public policy for inventions made in federally funded work then becomes “whatever the contractor that hosts the work chooses to do, so long as the contractor files a patent application.” In Bayh-Dole, there’s no federal review of a contractor’s policies or licensing practices. The contractor has sole discretion whether to license or not, license exclusively or not, assign (including by exclusive license of all substantial rights), or release the invention open access. Bayh-Dole makes a government secret all the contractor’s reports on its efforts to achieve practical application, and the terms of its exclusive licenses. Courts have even found reason to prevent the disclosure of the terms of exclusive licenses granted by federal agencies. In Public Citizen Health Research Group v. National Institutes of Health (2002), the court ruled

While the Court is extremely cognizant of the mandate underlying the Freedom of Information Act for public disclosure, in conducting the balancing of private and public interests, the Court determines that the private interests favoring withholding the information dominate the balancing. The licensees would likely suffer substantial competitive harm if this information was released.

The Court found that if the NIH had to release the terms of the exclusive licenses (and assignments) that it granted, “the effectiveness of Defendant’s [NIH’s] licensing program would be impaired.” The Court bases its opinion largely on the claims made by Maria Freire, that “the NIH would cease to be an attractive or viable licensor of patented technology.” While the Court tosses at every turn Public Citizen’s contention that licensees would not refuse to adopt NIH-owned inventions if the terms were made public, the Court accepts uncritically Friere’s claims. And Friere’s claims are pretty much the same ones made by most every university licensing office in the country–that if the public knew what the deals looked like, or even that there were not that many deals anyway–then the companies that the universities were dealing with would refuse to take the exclusive licenses (and assignments) on offer.

What does all this say about how “public interest” is getting used? It appears that the most important aspect of public interest, according to the NIH and university licensing officials, is that they get to control what the public thinks about their licensing efforts, and for that everything must be kept secret as a matter of federal law. The public interest is that patent licensing programs are made to appear successful, even when they aren’t. The public interest is that federal agencies and nonprofit organizations deal patents to companies in secret on an exclusive or assignment basis, without the need for any public protections or mitigation of patent exclusionary power. The public policy for inventions made in federally supported work–work presumably in the “public interest” in some other sense than the version we’ve just mapped out–should be delegated to individual companies, exclusive of all other companies (and research organizations) to be acted upon in whatever ways these chosen companies decide–nonuse, unreasonable use, exclusionary use, patent trolling, speculative pyramid schemes, monopoly pricing, deliberate scarcity, whatever. In effect, the policy delegation means there is no federal policy, or that federal agencies may abdicate their policy responsibilities–or, as it were, sell them off to whomever happens by willing to pay.

In a sense, Bayh-Dole authorizes federal agencies and nonprofits to sell out their public responsibilities with regard to key findings of federally supported work. Bayh-Dole covers such sell outs as “in the public interest.” Here’s the Bayh-Dole requirement on federal exclusive licensing, in 35 USC 209(a):

A Federal agency may grant an exclusive or partially exclusive license on a federally owned invention under section 207(a)(2) only if—

(1) granting the license is a reasonable and necessary incentive to—

(A) call forth the investment capital and expenditures needed to bring the invention to practical application; or

(B) otherwise promote the invention’s utilization by the public;

In this construction, (B) states the general condition and (A) states a specific instance of the general condition. If the exclusive license will “promote the invention’s utilization by the public,” then the federal agency is authorized to grant the license. This condition is pretty much an empty limitation. If the standard were that of the Kennedy patent policy, then it would not be that the exclusive license might promote utilization (whatever that ends up meaning–something like a sincere intention), but rather that the exclusive license will do it better than open access. Nothing here about the licensee’s “competitive harm” or how much money the licensee might give up by not price gouging or holding an invention. back in favor of other products that are selling well. 35 USC 209(a) continues with more about the public:

the Federal agency finds that the public will be served by the granting of the license, as indicated by the applicant’s intentions, plans, and ability to bring the invention to practical application or otherwise promote the invention’s utilization by the public, and that the proposed scope of exclusivity is not greater than reasonably necessary to provide the incentive for bringing the invention to practical application, as proposed by the applicant, or otherwise to promote the invention’s utilization by the public;

Here the conditions of (1) are repeated with the addition that “the public will be served” by the license because the licensee has submitted plans–kept secret–to promote use of the invention by the public. It’s not that the licensee has to actually follow its plan and achieve that public use–it’s just that the federal agency, in its secret deliberation, decides that the plan looks good. It is, in effect, authorization to sell out federal invention policy. The implied argument is that only by selling out federal policy-making with regard to inventions made in public research can this Latker exclusive licensing approach to the administrative idea of technology transfer be made to work. Only by companies excluding all others, charging outrageous prices if they ever develop any product, and otherwise not being held accountable for what they do and don’t do will the NIH be able to grant exclusive licenses to companies that insist on such practices. In its way, this is policy desperation.

Senator Nelson identified two flaws in the IPA program. The first we have worked through–ceding public policymaking responsibilities to companies using deals on secret terms without accountability. The invention deals carry more than “rights” to inventions–they carry with them a sale of the federal responsibility to act in the “public interest.” The rejoinder that a Latker enthusiast might give is that exploitation of the exclusionary rights of the patent system by companies and nonprofits, without federal guidance or oversight, best serves the public interest. The rejoinder is evidence-free, but at least we can see the shape of the disagreement, and perhaps the emptiness of the term “public interest.”

The second flaw identified by Senator Nelson has to do with making an exception–granting exclusive licenses–the standard practice. Here’s Nelson pointing out that the proposed “standard” IPA and HEW’s version are pretty much the same, and despite the appearances, default non-exclusive licenses don’t happen:

As one might have guessed, exclusive licenses are the rule and no the exception under the patent rights awarded by HEW pursuant to the IPA . . . .

Under the IPA, a contractor (Grantee) could grant an exclusive license if either of two conditions were met:

The Grantee may license a subject invention on an exclusive basis if it determines that nonexclusive licensing will not be effective in bringing such inventions to the commercial market in a satisfactory manner.

There’s no guidance on what might be “a satisfactory manner.” If the standard is “the public interest”–as required by the previous paragraph of the IPA–then the question becomes, who decides the public interest, and we are back to Senator Nelson’s critique that the federal government should decide, not some contractor or the contractor’s exclusive licensee. The IPA continues:

[A] Exclusive licenses should be issued only after reasonable efforts have been made to license on a nonexclusive basis,


[B] where the grantee has determined that an exclusive license is necessary as an incentive for development of the invention or where market conditions are such as to require licensing on an exclusive basis.

Here we have an either-or. A Grantee might offer nonexclusive licenses, and if there are no takers, then withdraw the offer of a nonexclusive license and offer instead an exclusive license. That might make sense as a standard, but the practice can be skipped by following the second condition. The Grantee need only “determine” that an exclusive license is “necessary as an incentive” or that “market conditions” “require” an exclusive license.

Well, it is obvious that exclusivity might operate as an “incentive.” That much is a given. But the “incentive” that’s set out here is in the context of the “public interest”–that the incentive need comes about because providing nonexclusive access does not result in anyone adopting the invention–even with “reasonable efforts” to get companies to adopt. One might think, then, that a “reasonable effort” would involve announcing the nonexclusive license and its terms–should be reasonable terms, FRAND terms–along with an offer to assist in the transfer of the invention and related information regarding its use and development. Let that offer stand for two years. If nothing happens, consider exclusivity–but also, consider other “reasonable” efforts–improving the non-exclusive terms, announcing the offer in other forums, and the like.

The second condition, however, is merely a mind-experiment wrapped in the form of a “determination”–if the Grantee “determines” that exclusivity is “necessary” or “required,” well, then, the Grantee can skip even trying the nonexclusive path. The IPA gives two options, with the (B) option being the general condition and (A) being one specific instance. We saw this at work in Bayh-Dole–both the products of Latker’s approach to drafting. So Grantees take (B), assert exclusivity is necessary–just as Dr. Friere did in the Public Citizen case–and there’s your federal policy on inventions flowed down to nonprofits, to do what they want, and what they want, apparently, is to deal in exclusionary patent rights for a share of the upside in the ensuing speculation over the value of such rights. That, apparently, best serves the public interest.

Nelson argues otherwise, a simple change:

The grounds (A) and (B) for allowing an exclusive license should be conjunctive instead of disjunctive–connected by “and,” not “or”–to require both tests to be met before an exclusive license may be issued.

As we have seen, Latker ignores Nelson’s stipulations in drafting Bayh-Dole. There are no constraints on contractors with regard to whether to grant exclusive licenses, and federal agencies are authorized to grant exclusive licenses on essentially the same (A) or more general (B) drafting cleverness used in the IPA. Well, actually in Bayh-Dole there are multiple constraints on the granting of exclusive licenses by contractors–it’s just that these constraints are not enforced because federal agencies are not required to enforce the default patent rights clause, are given opportunities everywhere to waive requirements, and the control over the regulations rests with NIST, which plays dumb with regard to understanding Bayh-Dole.

Constraints on exclusive licensing in Bayh-Dole:

35 USC 200 — the patent system should be used to promote free competition and enterprise. Difficult to do that with exclusive licensing.

35 USC 202(c)(7)(A)–nonprofits must flow down the nonprofit patent rights clause in any assignment of a subject invention, and that includes exclusive licenses that grant all substantial rights in a subject invention, placing a limitation on the scope of such licenses.

35 USC 202(c)(7)(old B)–nonprofits prohibited to grant an exclusive license to other than small businesses for longer than five years from first commercial sale or eight years from date of the license (with waffling language to allow extensions). This provision was removed in 1984 before it had a chance to operate.

35 USC 203–the “march-in” provisions that allow federal agencies to break up patent exclusivity to address nonuse, unreasonable use, noncompliant use (in the case of US manufacture under an exclusive license to use or sell), or regulatory or public health needs.

35 USC 209(a)–a host of requirements on federal agencies, all apparently empty, waivable, optional, or easily circumvented by the federal agencies.

Public interest suggests that inventions made in federally supported work should be available to all–both the U.S. public and the foreign public. Even the IPA and Bayh-Dole make their gestures toward this idea, while taking up the problem that so many inventions made in federally supported work were not being used.

And there’s the rub–why should an invention be used by others merely because it has been made? Just because some discovery or invention is “new, useful, and non-obvious” does not mean that it is *better* than some other practice, inventive or otherwise. One could come up with hundreds of inventions that meet the standard for patentability and are worse to use than what is otherwise available–expensive to produce, less effective, more difficult to maintain, more difficult to train people to use, less durable. There’s just no reason that some thing, because it was discovered in publicly funded research, must be used, or must be used right away, or must be used as a commercial product, or must be the object of speculation regarding its future value. 

At least, if “public interest” is going to have some meaning–other than a Humpty-Dumpty style assertion that the words mean whatever someone says they mean–then we ought to expect that results of publicly funded research ought to be made available in a manner that reflects public concerns rather than the sole discretion of private concerns, even private concerns held by nonprofits or federal agencies, such as whether their research or licensing programs appear to be “successful” or “profitable.” Yes, we might argue that “what’s good for some exclusive licensee is good for the public” or “whatever a nonprofit or federal agency decides to do with an invention to try to make money is good for the public”–but it’s a strange sort of argument, turning government into a handmaiden of patent administrators doing whatever they please, in secrecy, without accountability.

We aren’t talking here about laissez faire, or about compelled open access, or about a free market for inventors. We are talking instead about the use of federal law to enfranchise federal contractors and federal agencies to take up the inventive output of federally supported research and hold it back or grant rights exclusively to favorite companies or “investors” on secret terms. Inventors are cut out–so they have no free market, and cannot choose who they might work with. Industry is cut out–one company or none is the default practice. That destroys cumulative technology, pre-competitive collaboration, the development of technology platforms, shared research tools, technology libraries, and standards. That motivates companies to exclude and avoid federally supported inventions, to block their practice, to undermine their relevance, to design around, to ignore, and to move their development work to jurisdictions in which the patents cannot be enforced–such as out of the United States, if a contractor or federal agency doesn’t bother with foreign patent rights–or just to infringe anyway and make those nonprofits and federal agencies come after them for the money.

If we cannot have a good sense of “public interest,” then we cannot have a discussion about whether federal policy ought to retain the responsibility for deciding whether an invention should be made openly available–and with assistance in understanding and using it–or packed down with patent-related secrecy and secret dealing for exclusive rights. If “public interest” means whatever those in control of patent rights says it means, then Bayh-Dole is a law of private whims–and that alone is grounds to repeal it.

Now, it may be that private whims are the best thing possible for the public when it comes to the outcomes of federally supported research work–but in that case, we are better off without Bayh-Dole and its wasteful administrative costs and its deceptive, inoperative gestures at public protection from invention nonuse and unreasonable use. Perhaps we revisit Vannevar Bush’s recommendation, that federal agencies obtain licenses for government purposes and nothing more. Or perhaps we return to the Kennedy patent policy, and carve out special treatment for nonprofit contractors that lack market capabilities, and require them to be neutral–not taking an ownership interest in inventions made in basic research–and leaving inventions with their inventors, but for a federal license.

If we started from this Kennedy position, then we could ask what sort of programs might be worth funding to assist inventors who chose to pursue what they had invented. If an inventor wants help, what sort of thing would that be? Surely not confiscation by a university with impunity for a future minority share of any financial upside with no commitment by the university to do much of anything. Universities have put hundreds of millions of dollars into their patent licensing programs in the Bayh-Dole era. Universities have acquired 50,000+ U.S. patents citing federal funding (and over 120,000 patents including non-federal work). At $10K per patent–a conservative estimate–we are talking $50M just for patenting the inventions where a patent issues. Add in the patent applications that fail, the PCT applications, the foreign national phase applications, and patent maintenance fees. Add in the costs to operate a patent licensing program–the staffing, the risk management, the legal oversight, the database costs, the financial reporting, the litigation–and do that for even 100 universities–and we are talking on the order of $1M/year per university for 40 years. That’s somewhere around $4B on top of the patenting costs.

Is it in the public interest that universities divert $4B+ to patent licensing programs rather than to, say, support for principal investigators to provide assistance to companies wishing to explore the use of what a research program is discovering and creating? Sure, perhaps it doesn’t have to be either-or–but at present, it’s mostly “either” and all other approaches are suppressed in favor of private interests, kept secret, at their sole discretion. Somehow, that does not look at all like “public interest.”

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