It starts with Dubilier. In 1933, the Supreme Court decided in Dubilier that federal employees, just because they were employed, did not give up their personal–Constitutional–rights in inventions that they made. Considering the possible differences between private employment and federal employment, the Court placed the responsibility for claims on federal employment inventions with the Congress:
To the laws passed by the Congress, and to them alone, may we look for guidance as to the extent and the limitations of the respective rights of the inventor and the public.
And this court has held that the Constitution evinces no public policy which requires the holder of a patent to cede the use or benefit of the invention to the United States, even though the discovery concerns matters which can properly be used only by the government; as, for example, munitions of war.
Oddly, then, the Court creates the situation in which the Congress does not have a Constitutional mandate (or power delegated by the States) to create a claim on inventions made by federal employees or contractors, other than when, as with private employment, an inventor is hired expressly to invent, and does so. Thus, this outcome (quoting the opinion in Solomons v United States):
An employee, performing all the duties assigned to him in his department of service, may exercise his inventive faculties in any direction he chooses, with the assurance that whatever invention he may thus conceive, and perfect is his individual property.
The Court concludes that there is no public policy that inventors, as federal employees, give up their right to patent their inventions:
The government’s position in reality is, and must be, that a public policy, to be declared by a court, forbids one employed by the United States, for scientific research, to obtain a patent for what he invents, though neither the Constitution nor any statute so declares.
If there is to be such a public policy, it is to be found in the Constitution, where the Court has already looked and found nothing, or in a federal statute that derives from some power that the Constitution does authorize Congress to exercise. The Court says nothing about the executive branch just making something up as it goes–as far as the Dubilier Court is concerned, the federal government has two options: hire expressly to invent some defined thing or hold a shop right when some such thing has been invented using federal time or equipment. Nor should the courts be the place for developing such public policy:
The courts ought not to declare any such policy; its formulation belongs solely to the Congress.
Pretty clear guidance. The Court sums up–this is a the key paragraph, so here it is in full. Don’t skip it just because it’s long. That’ll do no good. I’ll break it up into more littler bits. Maybe that helps:
All of this legislative history emphasizes what we have stated—that the courts are incompetent to answer the difficult question whether the patentee is to be allowed his exclusive right or compelled to dedicate his invention to the public.
So, not the courts. Maybe some executive branch officer? There’s no express authority for that. And if there is authority for that, who?
It is suggested that the election rests with the authoritative officers of the government. Under what power, express or implied, may such officers, by administrative fiat, determine the nature and extent of rights exercised under a charter granted a patentee pursuant to constitutional and legislative provisions? Apart from the fact that express authority is nowhere to be found, the question arises, Who are the authoritative officers whose determination shall bind the United States and the patentee?
The Court does not warm to the idea that just anyone in government might dictate invention rights:
The government’s position comes to this—that the courts may not re-examine the exercise of an authority by some officer, not named, purporting to deprive the patentee of the rights conferred upon him by law. Nothing would be settled by such a holding, except that the determination of the reciprocal rights and obligations of the government and its employee as respects inventions are to be adjudicated, without review, by an unspecified department head or bureau chief.
So the Court arrives at the conclusion that Congress must provide the guidance–and, it goes without having to be said–within the powers that the Constitution provides:
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.
That “no power to declare it is vested in administrative officers” carries quite a kick.
Nearly 17 years after the Dubilier decision, on January 23, 1950, President Truman issued executive order 10096 claiming that the president had the authority “vested in me by the Constitution and statutes” and as President and Commander in Chief, to go ahead and issue a requirement that federal employees give up their right to patent whenever the government asks. Inventions are to be owned by the government if made during “working hours” or using government resources or personnel, or which “bear a direct relationship to or are made in consequence of the official duties of the inventor.” If the circumstances aren’t so clear as that, then administrative officers (with approval from the Government Patents Board) leaves the inventions with the inventors and obtains a shop right for the government. If it’s neither of these two circumstances, then all rights stay with the inventor.
Now EO 10096 runs against the Supreme Court guidance in Dubilier. The Court was adamant that the President did not have any express authority (urp–so maybe the prez has implied authority, legal reasoning being as it is) to have the federal government take from inventors the rights to their inventions unless the federal government hired those inventors expressly to invent the things they had indeed invented. You might not care, but Ervin Kaplan, a medical doctor working at the VA, argued that EO 10096 was indeed unconstitutional and the district court for Northern Illinois agreed:
Based upon the preceding analysis, it is the opinion of this Court that the President lacked the authority to unilaterally promulgate E.O. 10096. The provisions of the order are a violation of the separation of powers sought by the architects of our governmental system. The E. O. is, in fact, a unilateral promulgation which cannot be considered a part of the contract of employment so as to bind governmental employees under the principles of the Dubilier case. Absent this element of agreement between the parties, the order is little more than a unilateral statement of policy, in the style of legislation, treating rewards to be issued, or withheld, from inventors. The Constitution specifically reserves to Congress the authority to establish such policies and the attempts by the executive branch to make incursions on Congress’ authority are unwarranted and without force and effect. Consequently, this court is compelled to find that the E. O. is unconstitutional and of no effect on the merits of this case.
Nothing is ever finished, however, and the opinion was overturned on appeal in 1976. The appeals court found a statute (5 USC 301) that “would appear to authorize the President” to issue EO 10096. 5 USC 301 authorizes regulations for the “government” of executive branch departments, conduct of employees, distribution and performance of business, and “custody, use, and preservation of its records, papers, and property.” Weird, though, that there’s nothing here about taking personal property from employees, and that’s what EO 10096 authorizes. But the appeals court declared that it found what it was looking for, and unlike Bono, that was that. However, for over twenty years, the executive branch was busy trying to find a way to get Congress to confirm through legislation the authority of EO 10096, and it’s this little bit that leads to multiple efforts by agency officials and especially their patent lawyers to write bills that Congress might pass, under the rubric of needing “uniform” patent policy mandated by legislation. After numerous failures, Bayh-Dole.
But before 1980, the executive branch was hard at work as well on two other fronts–federal exclusive patent licensing (an Attorney General’s opinion was that Congress must approve such exclusive transactions) and allowing federal contractors and their inventors to retain ownership of inventions made in federally supported work. If you think that somehow greedy corporations somehow corrupted federal officials, you may want to put those thoughts on pause. Sure, corporations are motivated to get all they can, any legal way (or nearly legal, or maybe not legal but not enforced so everyone is doing it and is therefore de facto okay, if not necessary, to do). See Dodge v Ford Motor Co. But the motivation behind the need for “uniform” executive branch patent policy was driven by the executive branch, and in particular lawyers such as Howard Forman and R. Tenney Johnson, Norman Latker, and Jesse Lasken. Industry did not have to lobby so much as pave the way for the executive branch patent attorneys to drive their trucks.