Federal agency patent enforcement under Bayh-Dole, 5

Let’s work through the situation on federal ownership of inventions outside of Bayh-Dole protocols for acquiring inventions from contractors that screw up their patenting of subject inventions. Again, we are looking for any indication that federal agencies, in acquiring patents, or having the federal government issue patents to itself, has any right to enforce those patents other than as provided by Bayh-Dole–that is, by assigning the patent to a non-federal party, and only then when compliant with the conditions for exclusive licenses in Bayh-Dole.

Authority for a federal agency to acquire ownership of inventions made by federal employees is not within the scope of Bayh-Dole, just as Bayh-Dole gives no such authority to federal agencies to pass along to contractors, as the Supreme Court made clear in Stanford v Roche–Bayh-Dole is not a vesting statute for federal agencies or for contractors.

For federal ownership of inventions, we turn to an Executive Order, federal law, and executive branch patent policy. The outline is this: the Dubilier decision in 1933 confirmed that the federal government could own inventions but that Congress must provide the authorization to do so. The Attorney General (first Biddle, then Wolf, but really Sonnett but in practice probably Kreeger) produced a report in 1947 that assumes the federal government has the right to enforce patents it holds, but represents that the federal government has never done so, should not do so, and should adopt uniform practices that make doing so impossible, but for to support other controls on the use of inventions to protect the public. Truman’s executive order (1950) then takes up these issues to address federal ownership of inventions made by federal employees.

Federal ownership of inventions made by federal employees is addressed by Executive Order 10096, issued by President Truman in 1950. EO 10096 sets out the conditions under which the federal government may assert rights in inventions made by federal employees. It is codified at 37 CFR 501. While 37 CFR 501 gets as far as authorizing the filing of patent applications, it says nothing about licensing inventions or litigation to enforce federal government patent rights. That in itself is interesting. There’s nothing in the EO or codifying CFR that authorizes the federal government to enforce patent rights under 35 USC 281.

You might argue–there is no need! It’s obvious that the federal government may sue its citizens for patent infringement–to seek injunctions to suppress use of inventions, to demand compensation for “damage” caused to the federal government by its citizens using federally owned inventions without paying the government for the right to do so. It’s obvious too, you might say, that the federal government might choose favorite companies that it grants less than all substantial rights to, and then provides “air cover” for those companies by using the power of the federal government to sue infringers–beating up the unfavored competition, and offloading the cost and bother of patent enforcement to the federal government, to the benefit of the owners of the favored company or companies. All this threat to suppress use and demand payment for governmental “damage,” you might argue, is in the public interest. You think?

There is authorization in EO 10096 for the federal government to take ownership of inventions, as there is in various federal statutes that allocate money for federal research and development. The EO also provides for the filing of patent applications. So far, so good. There’s just nothing in EO 10096 that authorizes federal agencies to enforce patents on those inventions. Isn’t that strange? Here’s President Truman dealing with the aftermath of the Dubilier decision, under which federal employees using federal resources with federal permission to do so but working outside the scope of their assigned duties, invent something useful (in radio communications) and the Supreme Court finds that they own their inventions, that nothing in federal employment somehow voids their right to do so, the government gets a shop right, and if folks don’t like the outcome, then they should do something about it.

Tucked into the Dubilier decision–as if assumed–is that the federal government wanted ownership of the invention so that the invention would be available to all. The federal government did not need ownership to have a free right to the invention for government purposes. The whole point of the dispute, if you will, was whether federal employees could deal in patents on the side while being federal employees, while the federal government then had nothing to say about it. The government case was not that the federal employees had chosen a bad commercialization partner, or that the federal government could do better patent licensing deals than could employee-inventors. Here’s how the Supreme Court put the government’s argument:

The bills aver that the inventions and patents are held in trust for the United States, and that the court should so declare. It is claimed that, as the work of the Bureau, including all that Dunmore and Lowell did, was in the public interest, these public servants had dedicated the offspring of their brains to the public, and so held their patents in trust for the common weal, represented here in a corporate capacity by the United States. The patentees, we are told, should surrender the patents for cancellation, and the respondent must also give up its rights under the patents.

The government argument was that the federal employees must hold their patents “in trust for the common weal.” The Court rejected that argument. Where would the government draw the line on what could be the inventor’s and what must be the public’s? The Court concludes that it is not up to the federal government–it is up to Congress:

The courts ought not to declare any such policy; its formulation belongs solely to the Congress.

The Court then considers two Acts under which the government permitted patenting but required the invention to be released to the public royalty-free. In both cases, courts found that a patent without a right to exclude was not rightly a patent:

If the proviso must be construed literally we would have a situation wherein all the patents taken out under the act would be nullified by the very terms of the act under which they were granted, for the reason that a patent which does not carry with it the limited monopoly referred to in the Constitution is in reality not a patent at all. The only value that a patent has is the right that it extends to the patentee to exclude all others from making, using, or selling the invention for a certain period of years. A patent that is dedicated to the public is virtually the same as a patent that has expired.

Anything else would be merely “a publication or prior reference.” The Court then ends its decision by repeating its assertion that federal administrators lack the authority to make a claim for federal ownership of inventions:

Hitherto both the executive and the legislative branches of the government have concurred in what we consider the correct view,—that any such declaration of policy must come from Congress and that no power to declare it is vested in administrative officers.

Now, there’s one paragraph that the Supreme Court later deleted from its published opinion (the Solicitor General requested the deletion). That paragraph is useful for understanding the expectation of the federal government in obtaining patents:

Moreover no court could, however clear the proof of such a contract, order the execution of an assignment. No act of Congress has been called to our attention authorizing the United States to take a patent or to hold one by assignment. No statutory authority exists for the transfer of a patent to any department or officer of the Government, or for the administration of patents, or the issuance of licenses on behalf of the United States. In these circumstances no public policy requires us to deprive the inventor of his exclusive rights as respects the general public and to lodge them in a dead hand incapable of turning the patent to account for the benefit of the public.

It appears that Dubilier lays the groundwork for the scattershot of federal laws that followed and that do provide, in specific circumstances, that inventions made with the allocated funding be assigned to the federal government. Dubilier, then, too lays the groundwork for Bayh-Dole to intervene in the operation of those laws, following complaints from university patent management firms that the government takes too long to decide whether to allow a firm to retain exclusive rights in any given invention made in federally supported work.

As for the federal government’s ownership of patents, it is depicted as “a dead hand.” There is no conception here that the federal government would take ownership of inventions in order to sue its citizens for infringement, to suppress use, to demand payment for use. Any patent held by the federal government would be no patent at all–it would be a non-patent, a mere publication, a patent with no term of exclusivity, a dead hand. It is this image that then informs the criticism of federal patenting in the run-up to Bayh-Dole, that the government has all these patents–26,000, 28,000, 30,000–and most aren’t licensed. Well, those patents are, in the dead hand, as it were, not rightly patents. One can see in the Dubilier court reasoning not that the federal government would do a worser job of licensing than might inventors, but that the federal government was “incapable of turning the patent to account for the benefit of the public.” The federal government lacked the authority to enforce the patents. There was no exclusive right that the federal government could exploit when it took an invention from an inventor. The government might issue patents to itself, but it could not manufacture from that a right of the federal government to then sue its own citizens for using what the government had taken from inventors to better serve the public “weal.”

David Kreeger, most recently the former Special Assistant to the Attorney General and primary editor of the Attorney General’s report on government patenting (1947), writing in Law and Contemporary Problems (1947), argued that the federal government should take ownership of the inventions of federal employees. A government shop right in such inventions is not sufficient:

Clearly, a nonexclusive Government license alone cannot guarantee the fullest public benefit from the invention, nor prevent a private toll for the use of publicly funded technology, nor assure against its complete suppression; it is only Government ownership or control of the invention that can place the public interest first, and either open to the public the technology which public funds have financed, or, if conditions upon its use are desirable, frame those conditions in the public interest.

In all of this, it is difficult to square the problem of federal patent enforcement with the idea that when the federal government took ownership of an invention and obtained a patent, the entire purpose of the patent is wrapped up in the exploitation of its exclusivity. Kreeger’s argument is that the patent provides the basis for the government to place conditions on the use of an invention made in work receiving federal support. Implicit is the idea that behind any such license is the government’s right to sue for infringement if those conditions are not accepted, license refused, and use proceeds. But no one addresses such patent enforcement.

A patent license is a promise not to sue for infringement. If the government has no standing to sue for infringement anyway, what is the point of the license? There may be other advantages involved in a contract with the federal government involving the transfer of technology–technical advice, for instance–but there’s no point really in having a license.

But once Bayh-Dole comes along and authorizes federal agencies to grant exclusive licenses, then there is a point to a non-exclusive license–a non-exclusive license becomes a promise by the federal government not to grant an exclusive license that would deny access to all others. And if the exclusive licensee has the right to enforce the patent, then the non-exclusive license acts as a safe harbor against the threat that the federal government would choose a favorite company to wield the federal patent as an exclusive–and now private–right.

For the Dubilier court, a patent without the right to exclude is no patent at all. The Attorney General’s Report finds in federal patents a limited right to exclude only those practices that would harm the public and for which no other federal remedy exists. Thus, at least in 1947, we can see that the federal government expected to have a right to sue for infringement but that it would not exercise that right except in non-exclusive licensing to protect the public from misuse of inventions. This was the practice recognized by the Kennedy executive branch patent policy (1963)–a right to sue for infringement but not to exclude all practice or to extract payment but only to protect the public.



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