Ten Years After 25 Years After Bayh-Dole, Part 7

We are working through Boettiger and Bennett on changes they would like to see in Bayh-Dole practice. Here’s the fourth change:

access to patented, publicly funded technologies for humanitarian purposes

We now reach a deeply ironic portion of our authors’ discussion. Boettiger and Bennett started with a worry about research rights and research tools–both of which require apparently some degree of compulsory licensing or limitations on exclusivity or limitation on the property right in patents on subject inventions. These are perfectly legitimate concerns, especially for research that’s conducted to advance research.

The moves proposed by Boettiger and Bennett, if done at the federal level, correspond to a substantial reversal of Bayh-Dole and a return to the Kennedy patent policy, in which a portion of the patent right is held by the government, for non-exclusive access, and any monopoly portion of the patent had better act promptly to “call forth private capital” and disseminate an invention more rapidly and with better results than could the federal government.

As our authors examine the anticommons created by university administrators, they are led to repudiate Bayh-Dole–without coming out and saying so–for a range of publicly funded projects. They decline to propose amendments to Bayh-Dole–perhaps because doing so would undo most everything that made Bayh-Dole candy to patent brokers in the first place and return things to the protocols of the Kennedy patent policy–but now without the external network of agents that had the connections with industry and could work selectively and independent of institutional requirements and conflicts of interest. You know, the system that taught the Bayh-Dole middlemen how things could be and made them try to create their own system to displace the one that was, to some degree, working.

But in Boettiger and Bennett’s fourth point, above, we hit squarely the objection that the Kennedy patent policy had to monopoly control of research results intended for “humanitarian” purposes. The “humanitarian” purpose is just that of “public welfare” as set out by Vannevar Bush. Under the Kennedy patent policy, there would be no monopoly in the results of such research. Under the IPA program, the government could step in at any moment that the work of a monopoly holder was not clearly better than the government’s own release of rights. But under the arbitrary policy of Bayh-Dole, no government purpose–indeed, no researcher’s intention–has any claim on the disposition of patent rights. Even the non-Bayh-Dole gesture at 2 CFR 200.316 regarding intangible property rights acquired or improved with federal funds provided as grants is ignored by university patent administrators. They respect no obligation to act as trustee in such rights on behalf the beneficiaries of the research. In this, they deny the purpose for which the funding has been provided and substitute a claim that monopoly rights deployed for institutional self-benefit are the means by which the public will have the benefit of federally supported research.

The argument for university monopoly rights in whatever form is corrupt in the general case and in any event it has proven to be ineffective both at stimulating development of inventions or making money for universities. It’s not that monopolies don’t have a legitimate role to play. Even a commons can be, in its way, also a monopoly. It’s rather that the particular version of monopoly deployment created by university administrators is fundamentally corrupt–corrupt with regard to the purposes of federally supported research (which are much broader than “commercialization” and only rarely focused on “commercialization”), corrupt with regard to the operation of the Bayh-Dole Act (mischaracterizing the law, refusing to comply with the law, taking personal property to which university administrators have no right), and corrupt with regard to uses and outcomes (keeping outcomes secret, misleading the public with regard to performance, persisting in monopoly claims despite nonuse and unreasonable terms).

Our authors, to their credit, have a difficult time with Bayh-Dole. Here is how they set up the problem:

The Bayh-Dole Act has a provision to require substantial domestic manufacture of products developed from publicly funded IP. This provision has probably outlived its usefulness because in today’s globalized environment, offshore product manufacture may well provide greater net economic benefit to the US economy.

Let’s be clear on a few things, first. The provision in the standard patent licensing clause required by Bayh-Dole pertains only to exclusive licenses to use or sell in the U.S. For those licenses, there must also be a requirement to make in the U.S. The bit about “publicly funded IP” is babble talk. Bayh-Dole pertains only to patentable inventions. And the restriction is limited, waivable, and in any case, actionable only by march-in procedures that have never been used in the history of the law. In other words, our authors are going to tilt at a windmill, here, because the thing they object to isn’t in the law.

Now here’s the odd thing about our authors’ objection to the U.S. manufacturing requirement: the reason they give for opposing the measure is precisely the reason given in the first place for requiring it–to place manufacturing here in the U.S. In fact restoring U.S. leadership in innovation was Senator Bayh’s leading premise in advocating for the law. This provision, directed at “today’s globalized environment,” was as pertinent (at least to Sen. Bayh) in 1980, as it was 25 years later when Boettiger and Bennett were writing, and today. Yes, the provision is “parochial” if not protectionist. Perhaps Bayh-Dole should require the same restriction for every foreign country in which a patent owner of a subject invention offers an exclusive license: if the license is exclusive to use or sell in a given country, the license must also require substantial manufacture in that same country. This would be an easy edit to Bayh-Dole. And university administrators could implement such a policy without Bayh-Dole, with a snap of their fingers. But they don’t, and won’t. Privilege without accountability has its perks.

But our authors make another point, substituting “IP” for “patent rights”:

When IP developed with public funding is licensed exclusively to private companies, the technology is typically unavailable to support product development for low-income markets, even when the commercial licensee has no intention to address those markets.

This is an important point on its own. Let’s unwrap it a bit for what it says and doesn’t say, and for its implications. First, there’s the “When”: how often are subject inventions licensed exclusively? Of course, we don’t know because Bayh-Dole now declares that all reports of use are secret. Stanford reported that it licensed about 20% of its inventions. Let’s say that’s an outer bound and most other licensing programs that are dedicated to exclusive licensing do about the same. That would mean that 80% of subject inventions are otherwise available for “low-income” markets. The thing keeping those inventions from such markets is the refusal of university patent brokers to make the inventions available. They would rather not license at all than to license non-exclusively.

Next, consider “is licensed exclusively”: here our authors use the passive voice. No agent gets named. But we know the agents–they are the university patent administrators, the licensing officers, the brokers. They could license with reservation of rights. They could license make-use rights non-exclusively and deal with the sell right territory by territory as a conditional–sell in this territory at a reasonable price within x years and you have an exclusive position in that market for y years, so long as the price stays reasonable. But patent administrators refuse to do this sort of thing. Instead, they cheat the system the other way: they assign patent rights under the guise of an exclusive license. Perhaps they are clueless, and perhaps their legal counsel is clueless. The difference is that when legal counsel is clueless, it’s called malpractice.

When an exclusive patent license includes each of the common law rights of an invention, that’s an assignment. Bayh-Dole prohibits nonprofits from assigning subject inventions without federal agency approval except to invention management organizations. Hence, Bayh-Dole’s wording on the requirements for exclusive licensing in the U.S. is carefully drafted–if an exclusive license grants the right to use or sell, it must also require U.S. manufacturing. But if an exclusive license were to grant the right to use, sell, and to make, then it wouldn’t be an exclusive license. All the ownership rights would be with the licensee. University patent administrators seal the case by also granting the right to sublicense and to enforce the patent. Enforcement of a patent is something only an owner of the patent can do. Hence, it’s clear that such documents labeled “exclusive license” are in fact assignments. There are an exclusive license to the ownership of the patent, not an exclusive promise not to enforce the patent against the licensee. What’s conveyed in such “exclusive licenses” are the substantial patent rights, not merely freedom from patent infringement granted under some patent right. But university patent administrators ignore all this. Bayh-Dole is, after all, a law of administrative convenience, and complying with the law is a matter of picking out the convenient parts and ignoring the rest.

Now, “product development”: here, the abstraction gets in the way of the point. If an exclusive licensee is developing a product in a particular ” market,” the complaint by our authors is that no one else can develop a competing product in that “market” for low-income people. That is, at reasonable price for those who aren’t wealthy. In Bayh-Dole terms, a “reasonable price for those who aren’t wealthy” might be “available to the public on reasonable terms.” In Bayh-Dole, “the public” does not mean, for convenience, “those that can afford to pay” or “the wealthy and desperate first.” The “public” is just that–the general run of folks. If the benefits of use of a subject invention aren’t available to the general run of folks at a reasonable price–no matter what country–then practical application has not been achieved and the subject invention is not “reasonably available”–even if product is being sold.

Of course, federal agency officials also don’t appear to give a rat’s ass about all this–and don’t have to because there aren’t any penalties in Bayh-Dole for refusing to enforce the law. In essence, Bayh-Dole states the maximum that a federal agency can require to “protect the public” from nonuse or unreasonable use–but apparently federal agencies are free and clear to do even less to protect the public. Perhaps this is the greatest defect designed into Bayh-Dole. Whatever the gesture about patent rights–that’s an empty gesture if there’s no need for university compliance and no interest in enforcement by federal agencies. If Bayh-Dole were enforced, 80% of subject inventions would be marched-in immediately–that is, there would be on the order of 20,000 march-ins. Universities would blow huge budgets fighting each march-in and would eventually give up. That would clear out much of that “difficult and expensive” portfolio that at present breaches Bayh-Dole’s standard patent rights clause. That would also eliminate much of the potential for university patent trolling.

But our authors are making a narrower point, too. Even when a university grants an exclusive license, it’s left up to the licensee to develop “product.” Let’s say the product is a new pen, one that writes upside down and stuff. The company can choose to develop the pen and sell it “upmarket” for a premium price, available as a status symbol as much as for utility. That might maximize brand reputation, even at the expense of profit. This approach satisfies a poorly drafted exclusive license–the university gets royalties and that’s enough. But under the same license, the company could have produced a pen that sells for next to nothing–it could be a better BIC, available to most anyone, and perhaps even with a better profit potential than the upscale version, but maybe as well with more risk and less brand value. “Tough luck” goes the university patent administrator–it has to be the company’s choice. And that’s where the discussion typically ends–with the university patent administrator taking an anti-moralist position–like Thrasymachus, arguing that what’s just is whatever those in power say is just. Bayh-Dole is not concerned merely that there’s a commercial product–it’s that the benefits of the invention are available to the public on reasonable terms. An upscale pen is only one way of practicing our pen invention–but does it meet Bayh-Dole’s expectations?

Now change “pen” for “prescription drug” and you get to the bigger problem–not just with drugs, but with anything that would directed from the get-go to helping everyone, not just the wealthy and the desperate, not just the companies pimping their maximum value at the expense of everyone else. If the purpose of the research–and even the purpose of allowing universities to step in and deal with patents–and even the purpose of the licensing is to make the benefits of subject inventions available to the “public” on “reasonable terms,” then this purpose has to move from grant proposal to funding agreement to patent rights clause to assignment to exclusive license. There’s precedence for such a thing in the grant regulations for universities–2 CFR 200.317 (previously 2 CFR 215.37). It’s just that rules are useless where university administrators disavow them and federal agencies go along with the administrators.

Our authors also reference “low-income market”: by this, perhaps they mean “developing countries” or some such. But there are wealthy people in developing countries, too. Often that’s why these countries are perpetually “developing,” because the wealthy eat up all the green grass, and everyone else has to scrape by. If federal funding to universities is not to advance science and not to develop a product for public use and not to procure something as a deliverable for government use, then about the only thing left to justify the government funding a university-based research project is public welfare.

Let’s expressly reject the proposal from university patent administrators that federal funding comes to universities so that they can produce inventions that can be patented and licensed for profit, so there can be even more research funding in an always-expanding research budget that will bloat and bloat until it encompasses everything. In whatever version, we can argue, this argument fails. Federal funding does not come to universities to ensure that technology licensing programs can be profitable. That is not the purpose of federal funding (unless it is to support profit-seeking patent licensing programs!).

If federal funding is for public welfare, it falls into category (a)(2) of the Kennedy patent policy:

Where . . .

(2) a principal purpose of the contract is for exploration into fields which directly concern the public health or public welfare  . . .

the government shall normally acquire or reserve the right to acquire the principal or exclusive rights throughout the world in and to any inventions made in the course of or under the contract.

And what’s the government to do with those rights?–not license them exclusively for speculative profit-seeking–

Government-owned patents shall be made available and the technological advances covered thereby brought into being in the shortest time possible through dedication and licensing and shall be listed in official government publications or otherwise.

The Nixon revision to this clause added “either exclusive or nonexclusive” after “licensing” to make it clear that the government could offer exclusive licenses–but with restrictions. What is interesting, here, is the continued effort to restrict the use of patents acquired by the government or arising from government funding. The underlying theme is that the patent system generally is not appropriate for government purposes–not when the government does the developing, not when the government specifies product it needs and procures it, not when the government assesses proposals for research to advance scientific knowledge or to address matters of public welfare. Perhaps it is better put this way: the patent system reserves to inventors the discretion when to use the patent system (no one is forced to file a patent application) and how to use any patent that might issue. When the federal government comes into ownership of such an invention, and obtains a patent, it has this same discretion. It may deploy the patent how it chooses, and if it allows others to deploy the patent, it does so with the understanding that those others are working on the federal government’s behalf, on its terms, to serve its interests.

And the last bit, Part 8.

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

One Response to Ten Years After 25 Years After Bayh-Dole, Part 7

  1. Pingback: Ten Years After 25 Years After Bayh-Dole, Part 6 | Research Enterprise

Comments are closed.