Initial inventors, cumulative development, and the public covenant in federally supported inventions, 2

We are using Steven Anderman’s article “Overplaying the innovation card: The stronger intellectual property rights and competition law” to work through ideas about invention and follow-on development in the context of federal funding for university research and the effect of the Bayh-Dole Act. Anderman argues that both IP law and competition law work to strike a balance between the rights of initial inventors and authors (especially of technology) and what a company can do in an competitive environment with regard to markets, consumers, and the like.

While Anderman’s emphasis is on EU competition law, his discussion raises points about how federal law attempts to regulate the competitive behaviors of owners of patents on subject inventions–posi–inventions made with federal support and owned by a contractor, that is, by a party to a federal funding agreement for research. Anderman makes a distinction between initial invention and follow on work, which he calls cumulative invention. Cumulative invention may be critical for innovation, for products to reach consumers, and for those products to be reasonably priced. Law that creates an initial monopoly position must be tempered so that monopoly position does not become despotic on the broader effort to provide the public with new and beneficial goods and services on reasonable terms. At stake is not merely pricing but also competition, opportunity, improvements, applications, and quality.

Anderman points out that a broad patent may create adverse conditions for innovation and public use that outweigh the private advantage of the patent if given its full play. Implicit in Anderman’s argument is the idea that a government grants a patent for a public purpose, not merely to create a dominant personal property right that may block innovation and public use. In the U.S., this idea that a patent is a personal right that is bounded only by the outer reaches of anti-trust law, whatever its merits, fails utterly in the case of posi. Despite its limitations–many by design–Bayh-Dole remains a form of competition law built into federal patent law to place restrictions on the property rights of owners of inventions made with federal support. 

Here’s Anderman:

A second issue has to do with the particular placement of subvention research in the broader mix of support. A claim frequently made is that the special role of university research is that of “basic” science–pursuing the development of science without a specific commercial application in mind. But if Bayh-Dole stimulates institutions involved in basic science to make sweeping claims in patent applications, so that all future follow-on practice is caught up in those claims, then follow-on work is all but precluded, other than for “science” in other universities. But none of that follow-on work can be used outside of academia, because the patent on the fundamental work excludes it–but for a license, which institutions by default insist must be exclusive, if not an assignment of all substantial rights in the invention.

We might argue that Bayh-Dole enables practices that far from promoting the use of inventions made with federal support on the contrary invite efforts to design around or undermine or ignore such inventions–inaccessible, offered only for monetization, offered only exclusively, offered with the substantial overhead of risk-wary institutional managers. Bayh-Dole does have in its statement of policy and objectives, which set boundaries for the patent property right defined by a patent on a subject invention, a policy and objective pertaining to competition:

to ensure that inventions made by nonprofit organizations and small business firms are used in a manner to promote free competition and enterprise without unduly encumbering future research and discovery

The latter part of this provision was added by amendment in 1984–the “without unduly encumbering” bit. In the IPA program, on the ruins of which Bayh-Dole is built, “free competition and enterprise” was encouraged by requiring non-exclusive licensing before exclusive licensing and by placing term limits on exclusive licenses, if such were eventually required. Fragments of this practice made their way into Bayh-Dole’s provisions for the standard patent rights clause, but even there, amendments effaced specific guidance–though those amendments did not do away with the policy and objective. It’s just that the policy and objective regarding competition must be interpreted with regard to the management of the patent property right in a posi generally, without any “safe harbors” for clearly approved practice.

We might ask, how may the patent system be used to promote “free competition and enterprise”? There would be no need for this policy and objective if any use of the patent system did so. There also would be no need for this policy and objective merely to indicate that Congress did not intend for Bayh-Dole to forbid what anti-trust law already made illegal. No, something else must be going on. This policy and objective must mean that there are limits on the property right in posi–that institutional owners of posi cannot use the patent property right in just any way, and cannot trade in that patent property right to permit others to use a posi in just any way.

It is here, in this provision on free competition, that we see the faint glow of a concern for competition. That concern is, in its way, a concern with regard to follow on developments. We might ask what the difference in meaning is between “free competition” and “enterprise” (or “free enterprise”). These are not synonyms. Competition requires at least to competitors. But free enterprise is not defined in terms of competition. It is defined in terms of rights. A free enterprise may be started without approval from a patent owner. An enterprise is an undertaking, a project, a business that involves some effort, some commitment. Like, say a “research enterprise.” There’s not much guidance in federal law for the definition of “enterprise,” but here’s a start.

21 USC 848 deals with “continuing criminal enterprise.” Apart from regular felonious activity, such enterprises are defined to involve activities

(A) which are undertaken by such person in concert with five or more other persons with respect to whom such person occupies a position of organizer, a supervisory position, or any other position of management, and

(B) from which such person obtains substantial income or resources.

That is, an enterprise (A) involves five or more people with a manager and (B) makes money. If the patent system is to be used to promote such “free” enterprise in the context of “free competition,” then it must be the case that each posi is to be used to encourage the formation of enterprises–not constrained enterprises, not just a single enterprise, but enterprise writ broadly, free enterprise. Free of what? we might ask. Free of monopoly patent claims. Free to do follow on work. Free to utilize. Free to achieve practical application. Such things run against the monetization of patents in the ordinary sense, and certainly run against excluding all practice of an invention while waiting (often forever) for a monopoly licensing opportunity to show up. These are not thoughts that apply to any ordinary patent, but it is clear that Bayh-Dole is not addressing ordinary patents and is not attempting to produce ordinary patents from subvention research inventions.

But it is manifestly clear that university patent brokers are trying to shape Bayh-Dole to do just that–to undo the idea that subvention research funding is provided to support research conducted in the public interest, and that inventions made in that research should be deployed with limitations on any patent property rights to limit the use of the patent system to specific practices and purposes. That was the purpose of the IPA program. That has been the agenda behind Bayh-Dole, its implementing regulations, its standard patent rights clause, and the train of amendments that have been slipped through–removing overt restrictions on exclusive licenses (while leaving in place language that requires such restrictions); removing expectations that some reports of invention use will be public (by changing a “may” to a “shall”); making march-in procedures convoluted (so to never be used); making the grant of a federal license to practice and have practiced unclear as to scope (though that scope was spelled out in executive branch patent policy and even in the IPA); removing restrictions on small business preferences (leaving a general expectation of such preference, but without meaningful guidance or enforcement).

Bayh-Dole in its most basic form has been used as an instrument to disrupt and by-pass federal policy that requires posi to be something other than ordinary patents. It is this use of the law, and not the flawed architecture or amateurish drafting or naive notion of the role of patents in research and public welfare, that makes Bayh-Dole such a stink-bomb. Bayh-Dole is the gateway drug for the corruption of university administrators and research faculty. Bayh-Dole appears to turn posi into ordinary patents on ordinary inventions, as if there was no public purpose involved in undertaking the research, as if the justification for using public funds to support the work did not exist, or that such justification was meant, really, to be a subsidy for the benefit of private speculators, as if the subvention was directed to those speculators and not to the public generally.

Do you see how the workaround operates? The basis for federal research support at universities is subvention. Inventions made in subvention should be made broadly available for public use and follow on development. Patents on these inventions–posi–should be used only when doing so enhances that availability or creates useful products where otherwise there would be none. But never at the expense of availability, or to diminish free competition or to preclude enterprises forming.

The policy and objective is to use the patent system to promote use, promote free competition, to promote free enterprise. These are restrictions on the ordinary patent. These reflect the public covenant that justifies subvention funding. To assert that the key advances of subvention research are ordinary patents is to repudiate the foundation of subvention funding. The government uses public money, so the repudiation goes, to subsidize the risk-taking of speculators seeking to monetize patents on subject inventions. The purpose, so the repudiation goes, of federal support for research at universities is to provide such speculators ready access to patents that they otherwise would not readily obtain–work would enter the public domain, work would be done in companies, work would be licensed non-exclusively. In each circumstance, the opportunity to monetize the patent rights through an exclusive position–to exclude use, to exclude follow on research, to exclude variations, to exclude competitors, to exclude enterprise–is lost.

Bayh-Dole, so the argument continues, provides a means of by-passing the public covenant. Or, put another way, Bayh-Dole represents an apocryphal re-interpretation of the public covenant. The new public covenant is for research advances made at universities to be encumbered with patent rights, as broadly and tangled and fragmented as possible, and these rights to be put on offer for monopoly exploitation–whether sold as monopoly products at monopoly prices or to be litigated for infringement if anyone happens to practice the invention, or traded on by speculators in the hope that someday there may be monopoly products or litigation or other speculators who see even greater future value. This is what university administrators mean by “commercialization.” And in pursing “commercialization” they destroy the environment for follow-on, cumulative development.

Anderman points out that under EU rules, it’s entirely acceptable for a company with a dominant position to use its patent rights to exclude others. Further, in a note, Anderman observes that even non-dominant actors can do the same:

But in the matter of subvention research–following both federal funding regulations and the misshapen Bayh-Dole Act–owners of posi do not have these same privileges. Their property rights in posi are restricted. Their behaviors with regard to licensing and assignment and use of funds are restricted. They are tasked with the public covenant that runs with subvention funding. They must use the patent system to promote free competition, not to eject competitors from the market. The purpose of posi is not to achieve a dominant position a market, but rather to encourage participate in that market.

Anderman notes that in response to EU claims that patents had been used inappropriately to harm competition and public good, companies countered with claims that patents represented “property rights.” This may be true. The US patent law stipulates that patents “have the attributes of personal property.” But patent law also prefaces this right with a restriction (35 USC 261):

Subject to the provisions of this title, patents shall have the attributes of personal property.

And for posi, those provisions are spelled out in 35 USC 200. Whatever attributes of personal property a posi has is limited by Bayh-Dole–all of it, including 35 USC 200, the statement of policy and objectives, the implementing regulations, the standard patent rights clause. Posi are not ordinary patents with ordinary property rights. Their purpose is to enable follow on work, not to preclude it or to paywall access. The rights in a posi are less than those that might be permitted by anti-trust law. Or, put another way, the standard for market abusive behaviors is lower for posi than for ordinary patents. It’s just that there’s a huge, well funded effort to argue that only the standard for ordinary patents should apply, that universities lack market power and therefore may do pretty much what they want with posi. Such an argument has no merit. It’s not just that the reasoning fails or that there’s little evidence to support it as a matter of pragmatic consequence–it’s more simple than these things. The argument repudiates the public covenant that runs with posi, and rejects the fundamental justification for the federal government providing public money to university-based research in the first place.

Posi may have the attributes of personal property, but those attributes are subject to the provisions of federal patent law, and “subject inventions” are a special category of invention carved out by the Bayh-Dole Act, for which the patents carry the public covenant established by the Act, starting with the restrictions set forth in the statement of policy and objective at 35 USC 200. We might construe those restrictions as mere limitations on the rights a patent broker really needs to exploit a patent right successfully, eventually to be overturned by further amendments to Bayh-Dole. And if we did so, we would be readily accepted by university patent brokers for evenings of social entertainment trading war stories of licensing exploits.

But we may also construe these restrictions as an effort to balance the rights of an initial patent owner with those of downstream researchers, users, and developers. These rights are especially important where the initial invention or discovery is in basic research, where the patent right may be so broad as to claim a new continent to be explored; and where the initial invention or discovery has been made in a project proposed to be in the public interest, in which the university is to act as a trustee not a self-interested owner; and where the government expects to fund other, related research that may build on published results, both to advance the frontiers of science as well as to promote new capabilities which the public may use, industry may use, and which government may use.

In this perspective, the distinctive role for university-based subvention research is to expand frontiers of science, not to claim those frontiers on behalf of would-be speculators, not to exclude others from following and also exploring. The public covenant is to promote utilization, to promote free competition and enterprise. To do these things, patent behaviors must be anything but ordinary. The university role–whether institutionally led or led by the principal investigators and inventors–is to seed the opportunity, not to fragment it and reserve it for monopoly exploitation. Necessarily, then, the public covenant that runs with posi property rights requires patent owners to break up the patent property right so that everyone has access to common platforms of technology. Rather than fragment ownership, such practices allow the formation of commons, and within commons for the development of platforms, standards, and shared enterprise.

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

Leave a Reply