We have been working through FSA order 110-1, an early–pretty much the earliest–federal policy on inventions made in federally funded work. Why? The imp of this policy’s approach to inventions, rights, open access, and patent monopoly haunts subsequent policy discussions all the way through to Bayh-Dole. In a sense, 110-1 sets up the Bayh-Dole argument that patent monopolies without public protections are key to the development of inventions made in federally supported work into products that benefit the public. The Bayh-Dole argument is wrong–wrong in practice, wrong in reasoning, wrong in outcomes–but being wrong isn’t much of a big thing when it comes to the politics of “innovation.”
FSA order 110-1 on inventions made in federally funded work states a preference (open access), reserves the option to restrict access (to use patent exclusivity–but only when necessary for adequate commercial development and distribution, or if restricting access would be necessary for better development and widest distribution), but with a procedure and standards to guard against excessive royalties and unreasonable conditions. It’s just that the policy imagines a procedure without giving thought to the problem of whether any procedure is possible (of the form: procedure to win the lottery by purchasing the correct tickets) and the determination to restrict open access similarly lacks guidance. The overall effect of the design of “open access except when you go through bothersome procedures to do so” takes on the appearance of “really should restrict open access (nudge, nudge, wink, wink) except you have to go through bothersome procedures to do so.”
Only agency slowness to formalize the obvious (so it appears) stands between the public and beneficial commercial development. It is easy to see how the Bayh-Dole argument would emerge as federal agencies struggled with the impossible procedures and determinations regarding when patent monopolies might be better for the public than open access. Bayh-Dole removes a bungling bureaucracy around deciding when open access should be restricted and replaces it with a useless, wasteful bureaucracy around paperwork reporting, much of it kept secret. The point: the bureaucrats turned a question about how new things might come to be used into a general (and bungled) question about abstractions (“inventions” and “development” and “public benefit”) tied to a debate over which ineffectual bureaucratic practices then put a better public face on the policy. Free to all? or Patent monopolies?
The effect of the “uniform” demand made by the Bayh-Dole advocates is to prevent there from being any flexibility in the policy–federal agencies cannot choose open for some things and not for others without going through a procedure that is set up to prevent them from choosing open. The “uniform” demand is directed squarely at the default of FSA 110-1 as restated in the Kennedy and Nixon patent policies. On the other side is the Harbridge House report from 1968, which argues that a “uniform” federal patent policy is unworkable, does not reflect the variety of industry practices with respect to patents and even the variety of company practices with respect to patents within any given industry, and does not respect the impressive outcomes of federal agency practices based on variants of open access.
From 1952 until 1980, when Bayh-Dole was passed, federal officials fought a battle over this architecture baked into 110-1. It wasn’t a fight over innovation or public benefit or the use of inventions made in federally supported public health research. It was, rather, a fight over whether a bothersome, if not impossible, procedure ought to make restricted access difficult or make open access difficult. In either case, the working premise behind requiring a determination involves someone having to see the future and deciding in general one default bothersome procedure will be better than the other. There doesn’t appear to be a middle ground in which folks use their best judgment–that would be capricious and uncertain. Even when Harbridge House concluded in its report on government patenting that a single default policy was unworkable given the range of federal contracting purposes and contexts and industry practices.
One side aimed to enforce open access as the default. The other aimed to reverse the default preference and make patent exclusivity the default. The patent exclusivity folks, led by federal attorneys such as Howard Forman and Norman Latker, attacked the procedure by which federal agency heads were to determine whether to allow a contractor to exploit patent exclusivity. The procedure, they argued, was burdensome, slow, and uncertain. And in a way they were right. If a contractor failed to state a clear case for how their private exploitation of patent rights would be more effective at achieving adequate commercial development and distribution than open access, there would be a delay while the decision was mulled over. If many contractors all requested to retain rights at the same time, there would be a delay while the decision case load was worked through. And for all that–what sort of information was needed to make a case about what might happen in the future, with something new and apparently lacking commercial development, but thought to meet a pressing public health need? One can identify capabilities and promise to spend money, but how does one compare those claims–even if we think capabilities and money predict success–with the unstated claims of everyone else who might be interested in using or developing the invention but who haven’t been asked for their input? If asked, then what? More delays, more uncertainty. It is easy to see how the procedure to determine if patent exclusivity was necessary might be attacked.
So, instead of making the determination procedure more efficient, the opponents of open access defaults got rid of the determination procedure (to the extent they could–it still applied to non-small companies until Latker drafted Reagan’s Memorandum on inventions directed at all contractors) and created their own burdensome, if not nearly impossible, procedure by which to manage exceptions in contracting (“exceptional circumstances” under 35 USC 202(a)) and march-in after the fact (35 USC 203).
The executive branch opponents of open access also attacked the safeguards. These, they argued, might be necessary in the abstract but in any particular instance, safeguards restricted patent property rights and this worked against finding a company willing to take on the risk and expense of developing a given invention as a commercial product while excluding all others from practicing the invention. So let there be safeguards but make sure that federal agencies had the freedom to waive them, or waive enforcement of them, or to assert them in less restrictive forms, or not ask for information that would lead to a review of enforcement, or keep any such information secret. Anything but enforce safeguards that would limit exploitation of patent monopoly positions in any way patent holders wanted.
In 1963, eleven years after FSA 110-1, President Kennedy issued a government-wide patent policy. That policy acknowledged patent exclusivity for for-profit contractors with established markets–essentially all procurement contracting, with on exception. Any federal contractor with an established non-governmental market and capability to develop inventions that chose to do so, could retain patent rights to inventions made in federally supported work, subject to certain safeguards–a royalty-free government license to make, use, and sell; and a working requirement that the invention be brought to the point of practical application within three years of patent issue or make available for license on reasonable terms or the federal government could require non-exclusive, royalty-free licensing.
This default did not apply to nonprofits or to for-profits that were research only (“contract research organizations”). This default also did not apply where federal agencies were doing the development (and which they did with pretty much 100% success), and did not apply where federal agencies were the primary users of the results and allowing contractor patent exclusivity would mess with competitive bidding. Most importantly, this default did not apply–and this is the big exception–to any work addressing public health, that is, just what FSA 110-1 addressed. The Kennedy patent policy kept the FSA order 110-1 architecture in place for public health-related inventions made in federal work, but expanded the scope of the policy to include health-related funding by any federal agency, not just, say, the Public Health Service. For health-related inventions, then, the burdensome, slow, and uncertain process remained for unit heads to decide whether patent exclusivity was necessary, or could necessarily do a better job of creating adequate commercial development and distribution, or could at least do the job more adequately and quickly for widest distribution than any other approach.
The policy-prescribed decision is all but impossible to make on the criteria given. On the one hand, it is easy to see that patent exclusivity is almost never necessary. If something is really important for public health, then the government can pay for the development. If something isn’t important at all, then what’s the point of allowing patent speculators to disrupt downstream work that might rely on open access to research results? On the other hand, if one buys into the idea that patent exclusivity provides the necessary motivation (i.e., maximizing profits as the reward for taking on solely the cost of patenting, licensing, and development) for new things to be developed at all, without which inventions are suppressed, no products ever made, and public funding wasted, then it would appear that patent exclusivity is an imperative and beyond being made an option, companies should be found (or started) that will take those exclusive rights and do the development and distribution, at private expense, and in exchange for their efforts, retain a right to as much money as patent exclusivity can get them. Think of research funding as federal venture capital but without an equity interest in the endeavor and with a show of safeguards and accountability that never gets used to justify turning subsidies and subventions for public-directed work into investments with a profit motive made possible by patent exclusions. The entities attracted to the Bayh-Dole practice are those that would not spend their money unless they have the opportunity to exploit an exclusive position, including exploiting pricing, availability, applications, and variations. That is, patent speculators and investors in patent speculators.
Federal procurements and subventions for public health, in this patent monopoly argument, are turned into “investments.” But the return these investments is the availability of commercial public health products on whatever terms the patent holders choose–including no availability at all. The public should be thankful for any commercial products and accept the claim that without patent exclusivity there would be no such commercial products at all. In a bolder version of this argument, the public should be thankful just for the appearance of effort to make the patent exclusion process work.
The policy-mandated decision to permit patent exclusivity or require open access in federally funded public health research becomes a plausible preference. It seems plausible that someone could decide, based on a properly collected and vetted set of facts for any given research contracting circumstances. That decision could not be based on intuition, or whim, or a belief. Bureaucracy has no place for personal judgment, nor does operations at scale. As federal funding ramps up, and the number of contractors, both for-profit and nonprofit increases, individual decisions within federal agencies regarding the disposition of inventions becomes unacceptable. There cannot be another Vannevar Bush, wheeling and dealing to get things done, using a dozen top companies and a dozen top universities. Work has to be spread around, as pork funding, and requirements on inventions has to be decided by the manipulation of abstractions, reduced to rules and procedures.
This is where the patent exclusivity folks argued the policy was capricious and uncertain and had to be made “uniform”–that is, without involving a decision by agency officials except in “exceptional circumstances” before any research or development contract was awarded and only then by following a procedure set up to make it difficult to address exceptional circumstances. Otherwise, public health research would continue to be the primary exceptional circumstance and federal patent attorneys and their allied nonprofit patent development practitioners (eventually becoming the Association of University Technology Managers) would not be able to deal in patent exclusivities on inventions made in such research.
The premise of this private patent version is that money is what’s needed to develop inventions into profitable, mass-produced products. It is hard to argue with such a premise. But it’s not a sound premise, anyway. As Joseph Gabriel points out in Medical Monopoly, in the early history of American deployment of new medicines, the thing that was most desired was that physicians tested and evaluated new medicines first, prior to any decision to mass produce commercial versions. Money was not so important as physician access, observation, and reporting. Yes, one could buy physicians–but then we have the problem of how to obtain independent, non-bought, reporting.
Even with money as necessary to fund the “development” of a new medicine–including paying physicians to test candidate compounds–there’s still left open whether the money needed ought to come from one source only–the government, private speculators, companies competing to get their versions of a new medicine to the market, or organizations (companies, nonprofits) collaborating to fund some aspects of development (testing, say) and not others (advertising, perhaps). The patent exclusivity argument implies that a single private money source should be the default for the development of new, important public health products. The argument in favor of this approach presents such money sources as enablement–without such unique sources of private money, each directed at a new product, no new public health products will ever be made from the results of federally supported research.
We can see this enablement now in perspective. If all other sources of money for development are effectively suppressed by patent exclusivity, then only single private money sources will have an opportunity to “invest.” If federal agencies decline to fund development in favor of spending money on ever more research, the argument for patent exclusivity then becomes that the role of the federal government is research and development should be left to the “private sector.” Add default patent exclusivity, and the “private sector” becomes isolated private patent speculators each obtaining some organization’s or inventor’s patent rights in what rapidly becomes a fragmented patent landscape–and doing this less than 20% of the time, and generally failing to produce commercial product when they do obtain patent rights.
The premise is that innovation can be bought and the most effective people at buying in are capital investors who otherwise need to have anything in the game but their money. Companies that might already have made most of the investment–in talent, equipment, supply chains, distribution chains, customer connections–drop out of this view in favor of risk capital, with risks all the greater because the money does not come with the market and production intelligence of companies already involved in the area of some given invention. If all these things sound reasonable to you, then you will find an affinity for the private, speculative version of the use of the patent system.
The odd thing is that as a matter of law–or government regulation, or federal patent policy governing research contracting patent clauses–pushing patent rights from individual inventors to organizations set up to attract speculative capital to the development of inventions (on the general premise that all inventions require development, and that the development must be toward mass production of product (i.e., commercialization), and that the product must be profitable), appears to run against the Constitution. The practice runs against the idea of personal liberty, the idea that an inventor ought to have the say in how that inventor’s inventing might be used and developed. Sure, one can dislike the idea of personal liberty in inventing. Certainly university administrators have shown a general dislike in such things. Same for federal patent attorneys. Same for a lot of policy wonks. All appear to agree with Senator Bayh that inventors ought to be last in line for rights, after administrators at organizations and federal agencies had first pawed through ownerless inventions, Constitution and common law suspended until they had got their sniffing done.
The Constitution grants to the Congress the power to reserve exclusive rights in discoveries to inventors, not to organizations that might be disposed to deal patent rights to speculative investors. Even if one believed that speculative investors were, by default, the best means of realizing public benefit from federally supported research, enacting a law to force that default runs counter to the power granted to the Congress by Article I, Section 8, Clause 8. We might expect, too, that federal regulations to that effect would similarly be problematic, to the extent that they have the force of law. The Constitutional patent deal is offered exclusively to inventors, not to organizations parasitic on inventors, whatever the premise of their parasitism–even that they have in theory a better profile for exploiting for financial value the patent right to exclude all others. The Supreme Court in Stanford v Roche ruled as much–that organizations might obtain patent rights to inventions arising in federally supported work in conventional ways–by bargaining for those rights–but they sure didn’t get any special privilege to those rights via Bayh-Dole, not even leverage in their bargaining.
The Stanford v Roche decision should have been the death of Bayh-Dole practice. Universities should have repealed their patent policies that claimed inventions made in federal work. They should have made their potential inventors parties to the federal funding agreements and assisted those that invented with navigating the inventor patent rights clause at 37 CFR 401.9. But did they? No. Instead, they doubled down on university ownership–bureaucratic control–of faculty and student inventions. If there’s one thing a bureaucracy hates, it’s individual choice. Doesn’t matter if that individual is some graduate student or Vannevar Bush.