So far we have discussed President Reagan’s Executive Order 12591 mandating the promotion of commercialization (something not in Bayh-Dole) by having federal agencies grant title to patents (something else not in Bayh-Dole) to all contractors–not just to large companies that otherwise operated still under the Nixon patent policy of 1971:
(4) promote the commercialization, in accord with my Memorandum to the Heads of Executive Departments and Agencies of February 18, 1983, of patentable results of federally funded research by granting to all contractors, regardless of size, the title to patents made in whole or in part with Federal funds, in exchange for royalty-free use by or on behalf of the government;
Reagan’s executive order of course cannot unilaterally amend Bayh-Dole, nor can it ignore other statutes pertaining to ownership of inventions made in federally supported work. It can, however, unilaterally amend the Nixon patent policy to make the defaults under that policy for large companies be the same defaults that small companies and nonprofits have under Bayh-Dole: if a large company gains ownership of an invention made under contract, that company too can elect to retain title, subject to the public interest provisions of the patent rights clause (which are not enforced).
What the Reagan executive order doesn’t make obvious is that under the Nixon patent policy there are two sorts of company contractor–(1) companies, large and small alike, that have established nongovernmental commercial positions; and (2) other companies. A company in category (1) makes and sells products other than to the government. A company in category (2) either does not make and sell product or does so only to the government. Such category (2) companies include contract research organizations that exist to get contracts and deliver back research results, including any inventions that may have been made in the course of the research. Reagan’s executive order mandates that these category (2) companies are “granted title to patents” (or, perhaps Reagan really meant inventions, and perhaps meant allowed to retain title if they have got title) regardless of whether the companies have commercial expertise or market position. That’s a strange way to get about commercialization. Add in as well the companies that don’t have any expertise in making product. The Harbridge House report (1968) found that organizations without expertise were half as likely as those with expertise to succeed at commercialization of federally supported inventions. Shocking!
The Reagan executive order provides a further revision of the Nixon patent policy. The Nixon patent policy stated a default of federal ownership in four areas, regardless of what the market status of any contractor might be. The federal government should own when (i) a federal agency intended to develop products to the point of practical application, or (ii) when the federal government was the primary user and contractor ownership might set the contractor up as dominant, or (iii) in research directly concerning public health, safety, or welfare); or (iv) when the contractor was operating a government-owned lab or directing the work of others. The Reagan executive order preempts these conditions for all contractors that would otherwise be subject to the Nixon patent policy requirements.
The effect is somewhat consternating. For agencies aiming to develop products for subsequent release to the public (and to manufacturers), the Reagan revision to Nixon means that each contractor involved in a given agency effort can claim keep ownership of the inventions it acquires. If an agency spread grant work around to ten contractors, then the resulting technology fragments into patent positions for these ten contractors. The federal agency has nothing to release to the public, though it has a license to use the technology in its combined. To avoid such a result, under Bayh-Dole, a federal agency has to justify a second time its public purpose in sponsoring research, but in this second justification, the agency has to show that any deviation from Bayh-Dole will better meet the objectives not of the federal agency, nor support the public welfare, but rather better meet the objectives of Bayh-Dole–presumably the utilization of each independent invention, not the inventions taken together forming part of a technology prepared for public use.
Of course, each contractor acquiring ownership might be induced to license its share of the technology–but the overhead, cost, and risk of negotiating ten or more patent licenses just to gain access to what a federal agency has developed represents a tremendous barrier to utilization of the combination. But Bayh-Dole does not care. It treats each invention independently, as if there are no more inventions to be had in federally funded work that might promote utilization of a set of inventions made in contracts with different contractors.
A similar observation may be made about inventions made by contractors in an area of research in which the federal government is the primary developer and user of the technology. If there is no private market for such technology, then the primary purpose of such patents is to prevent other companies from competing to provide research services to the government. The company obtaining such a patent cannot have an established commercial position in a nongovernmental market–there is, within the premise of the requirement that there is no such nongovernmental market. One is left with the alternative that the use of such a patent by a company that has no market expertise for a market that does not exist, would be to offer the patent for license to companies that would, somehow, create such a market, where obtaining the inventions at no charge and even without licensing formalities from the federal government would not also do so, and more readily.
One is left with the idea that the company without expertise would license to just one other company–the monopoly meme–and that this one other company, which also must necessarily lack expertise in the invention, will invest sufficient private money not only to “develop” the invention that the federal government expects to use under its own license but to create commercial products in a market that doesn’t exist. There are examples, such as the emergence of private space technology, but those examples have not depended on a single company obtaining exclusive control of an invention from a contractor that otherwise lacked expertise and a commercial position in a nongovernmental market. In short, it’s an empty gesture that anticipates no particularly good thing other than to provide grist for speculators and disrupt federal contract bidding processes.
Finally, we may consider the effect of the Reagan executive order on research directed at public health, and here, in particular medicinal chemistry. Under Nixon, inventions in such research were to be owned by the federal government and made available to the public. Nixon revised Kennedy and made explicit that the federal government could grant exclusive licenses, but the Public Health Service and Health, Education, and Welfare had made clear that they were not about to grant exclusive licenses, while at the same time the NIH had worked to end run this prohibition by reviving the Institutional Patent Agreement program, under which the NIH signed a meta-research contract without any research in it under which contractors were required to take ownership of any invention made under subsequent NIH contracts that the contractor desired to patent. The IPA program was shut down in 1978 as ineffective and against public policy. Norman Latker, patent counsel at the NIH who had revived the IPA program, then drafted Bayh-Dole and later, after the bill had passed into law, worked on its implementing regulations.
The Reagan executive order expands the IPA work around for any contractor conducting federally funded research in areas of public health. Just as we saw in the first two cases, here inventions in related areas by different contractors can then be owned by those contractors, breaking apart common technology into different bits of ownership while preventing any other companies from having access, even if they want that access and would be willing to obtain that access non-exclusively from the federal government where they would be unwilling to do a licensing deal with a competitor under which they would most likely have to give over business and financial information, submit to audits, and pay not only for use, but merely to gain access before the benefits of use could be demonstrated.
No matter–the Reagan executive order declares that each contractor that obtains ownership of any given invention involving public health can keep ownership, and if a federal agency does not want that result, it has to justify a deviation from Bayh-Dole in a manner that shows an improvement in meeting Bayh-Dole’s objectives, not its own public purposes. Here the monopoly patent right proves itself all the more pernicious. Advocates for Bayh-Dole insist that the law was never intended to control the price of medicines–the public should be happy enough that medicines have been developed from raw inventions, no matter that a number of companies might have pursued development and even combined their efforts to speed development and lower costs.
Thus, we are left with the argument that Bayh-Dole’s purpose is to bolster the advantage of single companies claiming inventions and establishing patent monopolies relative to patients in need of medicines that it was in the public interest for the federal government to sponsor research directed at finding. Reagan’s executive order makes this outcome also the case for large companies working on federal research contracts, other than when federal statutes apply, for which, presently, there are none.
The Nixon patent policy had two ways agencies could relax the default for these four areas of government interest–they could relax the default at the time of contracting or they could relax the default invention by invention when inventions had been made. The Reagan executive order also does away with these relaxations of the default–because they are moot if the default is that contractors are “granted title to patents” (meaning, perhaps, that contractors are granted the right to retain title to inventions made under contract and which they own upon notice to the federal agency that they choose to retain title). If all contractors have this right, then large swaths of the Nixon patent policy are superseded.
All this is fine under 15 USC 2218(d), which expressly references the Nixon patent policy for fire prevention and control research:
All property rights with respect to inventions and discoveries, which are made in the course of or under contract with any government agency pursuant to this chapter, shall be subject to the basic policies set forth in the President’s Statement of Government Patent Policy issued August 23, 1971, or such revisions of that statement of the policy as may subsequently be promulgated and published in the Federal Register.
So whatever the Nixon patent policy might have had to say about inventions made under federal contract that aren’t subject inventions, the Reagan revision stipulates that those inventions are to be treated as if they were subject inventions. The questions to get at is who the contractors are, and how they are “granted title” to patents. In effect, Reagan replaced the Nixon patent policy with his own executive order, and that order applies whenever Bayh-Dole does not apply and specialty federal statutes do not intervene.
Thus, if an invention made with federal support is not owned by a contractor, then the Reagan patent policy provides that federal agencies will grant title to a patent on that invention to the contractor unless federal law prevents doing so.
NIST has recently revised the implementing regulations for Bayh-Dole to drop references to small businesses in favor of references without regard to size, arguing that this is the import of the Reagan executive order. Of course, NIST gets this just as wrong as it does other things about Bayh-Dole, since unlike Bayh-Dole, the Reagan executive order does not preempt federal law–it only revises Nixon patent policy, which is itself preempted by Bayh-Dole whenever a small business or nonprofit contractor acquires title to an invention made under contract.
All this points out the great vulnerability of a system that places the control of billions of dollars of research annually spread out over multiple federal agencies for multiple purposes in the hands of minor patent counsel tucked away in a nether region of the Department of Commerce. Any cock up cocks a long, long way. Jared Diamond in Guns, Germs, and Steel has a comment on the apparent limitations of central control in Asia relative to the constantly battling European states. NIST, however, does not appear much to care what it cocks up.