NASA, Monopolies, and the Cold War:

The Origins and Consequences of NASA Patent Policy, 1958-1996

Sylvia Katharine Kraemer

Director of Policy Development

Office of Policy and Plans

National Aeronautics and Space Administration*

Annual Meeting of the Society for the History of Technology, October 1999

The author will appreciate any comments on or suggested improvements to this paper. Readers are welcome to copy and/or quote any length portion(s) of the paper, but we request that you fully acknowledge the source.
 
 

(Outline:)

A. Patents and Incentives for Innovation

B. NASA Monopolies, and the Cold War:

Origins of NASA Patent Policy

C. Who Benefited Most From NASA’s Patent Policy?
 
 

A. Patents and Incentives for Innovation

The right to convert a novel idea into a commodity for its owner’s benefit and profit is a cornerstone of an industrialized free market economy. By securing inventors’ intellectual property rights in their inventions, patent law and policy not only govern how quickly and how profitably technology can be introduced into the marketplace; they can also divert the flow of capital from established enterprises to new industries promising greater profits. However, although the U.S. Patent and Trademark Office has been awarding patents since it was founded in 1790, many politicians and judges have condemned defenders of patents as would-be monopolists, intent on withholding important new inventions from the market place.

Since 1790 the USPTO has issued well over 5 million patents. During the past 50 years the portion of U.S. exports represented by intellectual property has grown from 10% to over 50%. Not surprisingly, discussions of U.S. policy have become complex, thanks not only to the growing volume of patenting over all, but to the increasing porosity of international boundaries in the high-technology marketplace. Magnify the patent interests of a single inventor by the intellectual property interest of entire industries, and one has the makings of global economic warfare fought in U.S. courtrooms and conference rooms of international trade agreement negotiators everywhere.

The treatment of intellectual property in U.S. law and policy reflects not only our legal heritage, but differing attitudes toward the social merit of new technology and business enterprise in general. In Great Atlantic & Pacific Tea Co. v. Supermarket Equipment Corp. (1950) Justice William O. Douglas, a Roosevelt appointee, argued that the lower courts, whose rulings the Supreme Court would reverse, had accepted a "flimsy and ... spurious" patent claim (for a device to draw groceries forward at the supermarket check-out counter) that failed to meet the Constitutional test of patentability, i.e., an invention which would "promote the Progress of Science and useful Arts..." Douglas then proceeded to ignore the second of these objectives by requiring that patentable inventions "serve the ends of science ... push back the frontiers of chemistry, physics, and the like; to make a distinctive contribution to scientific knowledge."

When Douglas wrote this opinion in 1950, university scientists, flourishing under Cold War-stimulated U.S. government research funding, argued that scientists advanced "the frontiers of chemistry, physics, and the like" as "basic" or "fundamental" science, which they pursued on university campuses for academic purposes only. "Useful" (not academic) research was applied science, performed largely in industrial laboratories. The rhetorical usefulness of their arguments was demonstrated when the successful legislation to create a National Science Foundation was first proposed in the Congress. Reinforcing the distinction between knowledge that was merely useful and knowledge that advanced science, Justice Douglas argued "It is not enough that an article is new and useful," insisted Douglas; "The Constitution never sanctioned the patenting of gadgets." (Historically, Douglas was being something of a radical: For none other than Thomas Jefferson -- who, as Secretary of State functioned as the administrator of the patent law of 1790 -- it was quite enough that an article be "new and useful;" it was mere scientific discoveries that were to be excluded from patents.)

Just why the Constitution’s standard of advancing the "useful Arts" could not be met by articles that were merely "new and useful" Douglas did not make clear. Today his view might arguably be dismissed as "elitist". Indeed, maintained Justice Robert H. Jackson, writing the opinion of the Court in this case, "The concept of invention is inherently elusive when applied to combinations of old elements. This, together with the imprecision of our language, have counseled courts and text writers to be cautious in affirmative definitions or rules on the subject."

Justice Douglas left no doubt as to the moral standing (in his view) of most "Progress ... [in the] useful arts" when, in writing his opinion, he reached back to 1882 to borrow from Justice Joseph P. Bradley’s opinion in an earlier patent case before the Supreme Court. "It was never the object of ... [patent] laws," sniffed Bradley,

to grant a monopoly for every trifling device, every shadow of a shade of an idea, which would naturally and spontaneously occur to any skilled mechanic or operator in the ordinary progress of manufactures. Such an indiscriminate creation of exclusive privileges ... creates a class of speculative schemers who make it their business to watch the advancing wave of improvement, and gather its foam in the form of patented monopolies, which enable them to lay a heavy tax upon the industry of the country, without contributing anything to the real advancement of the arts. Justice Bradley’s view had merit at least on one point: one observer has noted that "perhaps as much as 80% of the technical knowledge disclosed in patents never appears in a subsequent journal article or other scientific/technical publication."

The Supreme Court’s discussion in 1950 of the merits of Great Atlantic & Pacific Tea Co. v. Supermarket Equipment Corp. illustrates an ambivalence toward technological innovation in U.S. law and policy at the very time that the federal government began to pour unprecedented sums of money into research and development stimulated by Cold War anxieties. As the nation’s scientific and engineering research activities expanded exponentially after World War II, so also did researchers and their institutions become more dependent on public funding. But who, then, should own the fruits of that research? And, of equal importance, which sector -- public or private -- was best able to amplify those research results into profitable enterprises or increased public goods? In capitalist societies are "useful...discoveries" best purveyed to society as articles of commerce, manufactured by those who may profit from them; or as public goods, in which case, profiting no one in particular, they may not be manufactured at all? The tension between government or the marketplace as the best purveyor of new discoveries -- fundamentally an ideological tension, given the difficulty of empirically and conclusively "proving" the point -- permeated the early debates over the disposition of patent rights arising from NASA-funded inventions. Those debates, like debates about economic policy, were not only about economic efficiency, but about the allocation of the social as well as private benefits of innovation.

B. NASA, Monopolies, and the Cold War: Origins of NASA Patent Policy

NASA patent policy originated in the midst of this ideological tension. When the agency was created in 1958, it permeated all discussions of the "right" patent policy for the new space agency throughout its first decade. The extraordinary amount of increasingly sophisticated military equipment required to conduct World War II, and the continuation of those requirements to implement the policy of "containment" during what would become four decades of "Cold War", transformed the U.S. government into a powerful consumer of the goods and services produced by U.S. industry. Because of the nature of these goods -- e.g., technologically advanced weapons, aircraft, artillery, ballistic missiles, surveillance equipment, and the instruments to guide and control them -- Federal dollars became the fuel that powered a vast enterprise of post-War industrial research and development. That enterprise produced, in turn, patentable inventions by the thousands. Resolving issues surrounding the ownership of patent rights in those inventions thrust all three branches of the U.S. government into a policy thicket that was no less ideological than technical.

When the Eisenhower White House drafted legislation that would become the Space Act of 1958 (which serves to this day in an amended form as NASA’s charter), the administration had two models, ready at hand, to draw from: the patent policy of the National Advisory Committee for Aeronautics, NASA’s predecessor agency created in 1915, and the policy used by the Atomic Energy Commission, created in 1946. Like the National Bureau of Standards (1901), the Bureau of the Census (1902), and the Bureau of Mines (1901 ), the NACA had been conceived as the publicly supported research arm for industry, generating the essential data and research industry could not, or would not, generate on its own. While World War I ravaged the European landscape, the NACA was tasked to use its "good offices" to bring an end to the patent war being fought with in the nascent U.S. aircraft industry. Otherwise the United States might not be able to prepare itself to play its part in World War I, should that become necessary. Fearing that the warring aircraft manufacturers would not agree to a cross-licensing agreement similar to that which had been developed when the automobile manufacturers were in similar straits, NACA Chairman Charles D. Walcott proposed to President Wilson that he seek an amendment to the current naval or military appropriations bill to authorize the federal government to acquire whatever patents it needed "by purchase, condemnation, donation, or otherwise." The legislation was enacted as proposed. Accordingly, the NACA established a subcommittee on patents, which increased the prospect for the industry that, scrapping over the whole loaf, it might soon have none.

The aircraft manufacturers settled for half a loaf. The NACA subcommittee on patents obtained in the summer of 1917 the NACA Main Committee’s blessing on a cross-licensing agreement, to be administered by the newly formed Aircraft Manufacturers Association, the general result of which was that the "American aviation industry would operate without major patents . . . the ideas and techniques of aircraft manufacturers were to be shared openly among the members." Not everyone was pleased with this arrangement. Vocal critics complained that the cross-licensing agreement had created an aircraft patent trust. The end of World War I in 1918 brought with it the end of $100 million in government contracts for military aircraft and parts. But the airplane had proven itself as an air carrier of people and goods -- and as an instrument of war. The Army Air Corps Act of 1926 contained an additional provision having a policy significance that would become more evident as the Federal government, during the next World War, became the largest "buyer" of new technology in recent history. That provision created a "patents and design board . . . the three members of which shall be an assistant Secretary of War, an Assistant Secretary of the Navy and an Assistant Secretary of Commerce," to evaluate the patentability (by the U.S. government) of new designs "for aircraft, aircraft parts, or aeronautical accessories . . . upon the recommendation of the National Advisory Committee for Aeronautics." Should the board so advise, the U.S. government would offer to buy the patent, or a nonexclusive right to use a patentable design, for an amount not to exceed $75,000. Less than a year later the Congress amended the Army Air Corps Act of 1926 to place the burden of disposing of any patentable designs submitted to it on the NACA. Note, however, that the presumptive patent title owner was the inventor.

The White House evidently looked upon the new aerospace organization’s mission as more analogous to the Atomic Energy Commission (AEC). The AEC had been created in 1946 by an Act of Congress; its purpose was to transfer the management of the Nation’s atomic weapons development from the U.S. Army to a civilian authority. Because of the novelty of nuclear science and technologies and the critical role they had played, and were likely to continue to play, in maintaining a national security posture in the midst of the Cold War, the AEC’s patent policy presumed that the federal government would retain title to all inventions by its employees or contractors. The AEC, in short, was a "title" agency. (Department of Defense policy, on the other hand, presumed that the R&D contractor would retain patent rights to inventions made while working under a DoD contract, while reserving to the government a royalty-free, irrevocable and nonexclusive license to use those inventions.)

Many of the firms that were likely to serve as contractors to the new aerospace agency were already serving as contractors to the Defense Department, while at the same time conducting applied research and development under contract to the NACA. The NACA’s patent policy was similar to the Defense Department’s: When asked in 1949 to comment on a proposed "Executive Order Providing for a Uniform Patent Policy for the Government With Respect to Inventions Made by Government Employees," John F. Victory, Executive Secretary of the NACA, objected that the draft order ensured that normally "complete assignment of all patent rights to the Government;" such a policy "would make Government employment even less attractive to engineers and scientists than it now is." For these reasons, former NACA and current Defense Department contractors had undoubtedly expected the legislation creating NASA to contain a patent policy similar to that under which they had been performing R&D for the government already. Small wonder, then, that Washington representatives of the aerospace industry reacted strongly when they learned that the patent language in the draft legislation reported out by the House and Senate committees, after hearings during which patents were "rarely mentioned," would turn the new space agency into a "title" agency like the AEC.

Such was the bill that passed the House of Representatives on June 2, 1958, which directed that NASA retain title to "any invention or discovery made or conceived under any contract, subcontract, arrangement, or other relationship with the [NASA] Administrator." As compensation, the Administrator could make a cash award to the inventor. Alternatively, the Administrator could waive title while retaining a license to practice the invention, and offer a royalty fee to the inventor. Corresponding Senate legislation contained no patent language whatsoever.

Grumbling from the aerospace industry, as well as the legal profession, was sufficient to lead House Science and Astronautics Committee chairman John W. McCormack of Massachusetts to appoint a subcommittee to hold hearings on NASA patent policy as then proposed. After several weeks of discussion with "interested parties, both Government and private," the subcommittee offered substitute language which bound the NASA Administrator’s decision to waive title would be by two considerations: whether the inventor was a government employee, or whether the invention was made using government resources and services. If the invention or discovery, though not made by an employee of the government or a government contractor, did nonetheless relate "directly ... to the contract or to the work or duties he was employed or assigned to perform and was made during working hours, or with a contribution from the Government...," NASA would be the presumptive patent owner. In the event that the Administrator found that retention of the title by NASA was not warranted, the Administrator was bound to require from the patentee "a nonexclusive, irrevocable, royalty-free worldwide license to make, use, and dispose of the invention or discovery, or to have the invention or discovery made, used, or disposed of, for governmental purposes". The House subcommittee’s draft language also established an Inventions Review Board, which would hear appeals of title decisions made by the Administrator.

Notwithstanding the House subcommittee’s best efforts, the language that emerged from the conference committee in July, and which survived to become Section 305 of the National Aeronautics and Space Act of 1958, reverted to the NASA patent policy as initially written (by which NASA would be a "title" agency) while retaining a provision directing the Administrator to establish an Inventions and Contributions Board. Section 305 attempted to resolve the issue of public vs. private property rights in government funded inventions, in true American fashion, through a scheme of procedural "checks and balances" that distributed the authority to determine patent rights among the Administrator, the Commissioner of Patents, the U.S. Court of Appeals for the Federal Circuit, the existing Board of Patent Interferences, and an Inventions and Contributions Board to be established by the Administrator.

It is small wonder that the fledgling space agency had barely been set out to fly on its own before NASA’s lawyers complained of the administrative burdens imposed by Section 305 and everyone else (so it must have seemed to the Congress) began to complain of its content. Within a year after President Eisenhower signed the Space Act, the House Committee on Science and Astronautics launched into another round of hearings on the merits of NASA’s patent policy. As industry complaints grew, committee chairman Overton Brooks (D-La) appointed a special subcommittee on patents and scientific inventions to launch hearings on NASA’s now highly controversial patent policy. NASA’s attorneys were obliged to defend the Agency’s patent policy. Denying an inventor patent rights in his invention, argued Associate General Counsel Paul G. Dembling, did not necessarily diminish incentives for innovation, because Section 305 authorized the Administrator to make cash awards, on the recommendation of NASA’s Inventions and Contributions Board (created by the Space Act) for "any contribution of a scientific and technical nature which has significant value in the conduct of aeronautical and space activities."

Some in industry had asserted that NASA should increase the prices it was willing to pay for contracted work to compensate contractors for the losses they would suffer as a result of NASA’s patent policy. The government’s property rights are "a matter of law", replied NASA’s attorneys, and NASA could not be seen to purchase what it already owns. A few appealed to Cold War patriotism: "We would like to think that American industry will display the same patriotism in this cold war of space research," warned Ray Harris,

that it has always displayed in the shooting wars, and will not let selfish interests stand in the way of the national good. One cannot fail to get the impression, both from our officials and the public, that it is considered as much a defeat when the Russians put up the first Sputnik, or put up a bigger satellite than we do, as it is when we lose a fight on the battlefield. Accordingly, we are looking for the cooperation of industry in our programs regardless of the impact of these patent provisions.

A few days later, Gerald O’Brien chided the New York Patent Law Association:

Your campaign [against NASA’s patent policy] has alerted industry to shy away from participation in contracts with the Administration to cause even the slightest delay in the progress of space exploration or in aeronautics, or otherwise show any absence of unity of purpose of both industry and government, which for propaganda purposes may provide the back-drop to Khruschev’s acts at `missile blackmail’ or encouragement to the brandishing of arms for political effect.

Gravely maiming the body politic, or looting its treasure -- these were tough choices.

The summer and early autumn of 1959 found NASA spokesmen, along with over thirty other witnesses from government, industry, and the bar, and one university professor, passing through a hearing room on Capitol Hill as the House subcommittee, responding to "increasing criticism from industry and the professions", conducted hearings on new patent language for the Space Act proposed by NASA. The new language authorized the NASA Administrator "to prescribe by regulation the various terms and conditions governing disposition of rights to inventions made during the course of work under a NASA contract." The subcommittee’s summary of witness testimony echoed now familiar themes:

There is some similarity between the so-called patents of the Soviet system and the wording of sections 305 and 306 of the National Aeronautics and Space Act of 1958 (which are substantially the same as patent and compensation or award provisions of the Atomic Energy Act). The Government ownership provision of the NASA and AEC statutes, if carried out literally, would correspond to the `public patents’ provision of the Soviet system in that there can be no private commercialization of the inventions without Government approval. The `compensation or award’ provision of the NASA and AEC statutes correspond to the `private patent’ provisions of the Russian system in that the inventor will not be rewarded unless the Government allows commercialization or use of the invention in the interest of the public. While NASA’s and the AEC’s patent policies may have resembled those of the Soviet Union to members of the House subcommittee, the Department of Defense required only "an irrevocable, royalty-free, nonexclusive license" to use an invention made by one of its R&D contractors. Among other government agencies, patent policies ranged from ones to the AEC’s to policies similar to the Defense Department’s. Defending NASA’s proposed policy revisions giving the Administrator greater discretion over the assignment of agency patents, NASA’s witnesses cited several instances in which the agency had had to forego the use of an R&D firm’s advanced technology when the firm declined to contract with NASA because of its patent policy. Gilding the lily, House subcommittee chairman Erwin Mitchel (D-Ga) added, speaking in support of a June 1960 resolution based on his committee’s report, "the basic reason is because the present patent provision is detrimental to our space effort. With the current patent provision in effect, NASA does not get the best research for the lowest cost. Sometimes it cannot get the research it needs at all".

Lest the House subcommittee appear closed-minded in its support for rendering NASA’s patent policy more friendly to the emerging aerospace industry, it heard and reported on the testimony of three witnesses who did not favor the NASA proposal. The first was Roland A. Anderson, General Counsel of the Atomic Energy Commission. Anderson pointed out the inconsistency between the patent language now favored by NASA and Sections 203(a)(3) and 303 of the Space Act, which require the agency to disseminate scientific and technical information to the public subject only to national security restrictions. A similar argument, which treated publicly acquired knowledge as a public good, was made by Dr. Seymour Melman of Columbia University. Administrators and scientists at U.S. research universities had been arguing with considerable success at least since Vannevar Bush’s Science: The Endless Frontier (1945), that science was the source of all technological progress. On behalf of that same community, Melman argued that treating knowledge as a commodity to be owned, rather than a public good, "has the effect of placing barriers to communication and thereby restricts the pace of [scientific] inquiry. . . . free publication promotes the widest and the most rapid diffusion of knowledge, while restriction on publication has a converse effect."

The issue was put into sharper relief by Chicago patent attorney Louis Robertson. Granting NASA the latitude to award patents without statutory guidance, argued Robertson (thus parting company with many of his colleagues at the bar), would endanger the credibility of the entire patent system. Awarding a patent is tantamount to giving the patentee a "monopoly on the invention; . . . there are [not] many times when the monopoly power to the private industry on something developed by the Government or at its expense is justified. And if it isn’t justified, I think that the power to grant that [monopoly] power probably should not be in bureaucratic hands. There is too much danger that some of the administrators having that power would lack the constant desire and diligence necessary to keep it from being abused." Robertson added, when "the contractor often brings to these [NASA] contracts great skill, great collections of experts, great know-how in his own field," the contractor might "very well [be] entitled to royalty, but that doesn’t necessarily mean that he is entitled to monopoly power."

But Anderson, Melman, and Robertson were outnumbered. The House subcommittee’s report proposed that the revisions sought by NASA be adopted, with the addition of a requirement that NASA retain royalty-free, irrevocable licenses on any inventions produced under its contracts, and issue patent waivers "to cases involving contracts already in existence at the time of enactment of the new provision". The essential argument of the subcommittee’s report was that private enterprise and market economics would allocate most efficiently the fruits of federally-funded aerospace research

As the full House considered the NASA patent policy issue, the ideological confusion evident in the subcommittee’s hearings, persisted. Rep. Chet Holifield (D-Ca) fired the first volley at the proposed language, charging that it would "defraud the American taxpayer of his legitimate access to all of the fruits of the research and development which have come about as a result of the expenditure of public money." The now familiar debate rolled on from there, culminating in an announcement by Rep. James M. Quigley (D-Pa):

The most persuasive argument that has been advanced for considering patent changes in the NASA Act at this time is, in effect, that we did not know what we were doing when we passed that act 2 years ago. It strikes me that the most persuasive argument for not doing anything now is, as the debate has indicated, we are not any more sure as to what we are being asked to do at this time. Mr. Quigley was asked to yield to another gentleman.

Even though Mr. Quigley did not have the last word, the amendment was defeated. And, in fact, Mr. Quigley was right. Our understanding of the interaction of patent protection and technological innovation, much less the interaction of technological innovation and economic growth, was in its infancy. That, however, did not prevent the expenditure of a good deal of energy, ink, and paper among the contestants in this debate to persuade each other that the very survival of the country or, alternatively, the people’s hard-earned treasure, hung in the balance. NASA’s attorneys would have to soldier on.

Only one thing was certain as the issue was joined and rejoined in the Congress, the patent bar and industry associations, during the five years that passed between the creation of NASA in 1958 and President John F. Kennedy’s Statement on Government Patent Policy, issued October 10, 1963: The federal government itself could not, should not, and would not attempt to bring any of the devices to which it held patent rights to market. To do so would be to compete with the private sector, which would have been an unthinkable intrusion of government into the market place -- especially unthinkable in an era and a political culture in which virtually every R&D dollar spent by the U.S. government was justified as critical to the defeat of communism. At the end of the day, whether NASA should be allowed to become, like the Atomic Energy Commission, a surrogate owner, on behalf of the American people, of exotic and powerful technologies, would be determined by where members of Congress and their numerous and varied constituents stood on two ideological issues.

The first was rooted in competing conceptions of the "public good" in a country with an unwritten "free market" canon. One of the principles of that canon was that initiative, industry, and creativity--and by extrapolation, the economic progress of the United States--have all been powered by unfettered competition. Yes, the New Deal introduced the notion of a social safety net, limitations on bankers lest they speculate with hard-working families’ savings, and some supervision of the securities industry. The "Fair Deal" added the principle that full employment was a legitimate concern of the federal government. But not everybody had voted for Franklin D. Roosevelt and Harry S. Truman. The anti-communist fulminations of the 1950’s were still fresh in peoples’ minds, as was that residual American conviction that private property, and private enterprise, were the engines that "made America great". To the extent that NASA patent policy could be seen as a form of seizure of the actual or potential fruits of private enterprise, that policy would be subject to noisy dispute.

The second issue arose from the evident confusion over the actual purpose of the United States in Space. If NASA’s purpose -- and by the summer of 1961 President Kennedy had clearly re-articulated NASA’s purpose as a successful human mission to the Moon -- was to help "win" the Cold War, then the civilian agency was merely another sturdy recruit in the nation’s struggle against communism. Though a civilian agency, thanks largely to President Eisenhower’s foreboding over the growing power of the "military industrial complex," NASA’s connections with the armed forces extended from borrowed management experience and shared industrial contractors to shared technologies. A guided missile is a guided missile, whether it carries a warhead or a chimpanzee or an imaging instrument. Thus the question: was NASA to be a technological enterprise on its own right, alone and apart from the interests of the advanced technology industries it helped to create and support--or, for that matter, the scientists who supplied it with peaceful pursuits and the dreamers who supplied its "vision"? Or was the space agency merely a specialized battalion mobilized to fight the Cold War?

?

The aerospace industry was further aroused in the summer of 1961 by a NASA-proposed change in the agency’s licensing regulations that would have awarded the inventing contractor a royalty-free, non-exclusive license to its contractors’ inventions as soon as they were reported to NASA. But if a contractor did not exploit the invention commercially within eight years after disclosing the invention, or five years after the patent was issued, whichever came first, another contractor could receive an exclusive license from NASA to exploit it.

"It’s about time something concrete was done about the question of proprietary rights," complained the aerospace trade journal, Missiles and Rockets, in its July 24, 1961 issue. "Ours presumably is a society based on economic, democratic capitalism, not communism. We count on competition among private firms to keep us ahead of the Russians . . . . To dispossess a company of its know-how is to take away its winning weight on the shuffleboard. . . . In the competition with the Soviet Union in the decade ahead, every bit of industrial skill this nation possesses must be available to the government. Anything that threatens the enthusiasm and willingness of private firms to participate in government programs endangers our chances of survival as a nation. Missile/space industry firms are willing to gamble time and capital in programs such as Apollo. But they must be assured that they will not be denied the benefits of their efforts."

A month later, the industry’s leading trade journal, Aviation Week and Space Technology, reported on a study it had conducted that concluded that "only a fraction of the inventions made under NASA research contracts actually are reported to the space agency." More importantly, NASA’s patent policy was preventing the agency from obtaining expert industrial participation in the space program; some firms had "not dared to enter into any agreement with NASA or any of its prime or subcontractors which might be construed to invoke research and development work on [their] part;" thus "NASA patent policy [had] a ‘debilitating effect upon industrial incentive".

While the Congress stirred the broth of NASA’s patent controversy, the National Aeronautics and Space Council, a White House advisory group created by the 1958 Space Act, took up the issue. Given that the patent policy governing the rest of the U.S. Government (with the exception of the AEC on which NASA’s patent policy had been modeled) was more acceptable to industry, political inertia favored the declaration of a uniform, U.S. government-wide patent policy in which the inventing contractor was the presumptive owner of the patent rights to an invention made under a contract to the U.S. government. The White House would not have an easy task.

A New York Times headline to a story on the patent controversy declared: "industry Girds for Patent Fight". This was not an overstatement, considering the formidable Louisianan now determined to prevent what he saw as a massive theft of the people’s property. Writing in the Winter, 1961 issue of the Federal Bar Journal, Senator Russell B. Long (D-La.), chairman of the Senate Small Business Committee’s Subcommittee on Monopoly, offered one of the most extensive and detailed arguments supporting U.S. government retention of patent rights made by anyone at that time. The nub of Long’s argument was that the overriding public policy objective should be economic progress; economic progress required the rapid dissemination of the results of research, especially when the U.S. government has paid for that research. By this standard, NASA and AEC patent policies were the right ones, and should be extended throughout the government.

Not only did U.S. government retention of patent rights to contractors’ inventions favor small businesses and the economy in general, argued Long; assertions that the government would, as a result, not be able to get the best contractors, or be charged premium prices by them, were fallacious. Recalling public statements by senior executives from the Douglas Aircraft Company and Raytheon Corporation, Long insisted that Industrial contractors reaped real benefits from their government contracts--most notably experience and know-how--that could be used on other contracts. The traditional justification for the temporary monopoly on a product that constitutes patent ownership does not apply to inventions made under U.S. government contracts:

The granting of patent privileges is justified only insofar as it serves as an incentive to take risks.... But where are the risks in Government-financed research and development contracts? There really are none. Practically all R&D contracts let by federal departments and agencies are on a cost-plus basis. No matter how expensive a project turns out to be, costs are covered by the Government. Moreover, there is no risk in finding a market for the new product. The one concession Long was willing to make was that the government was poorly suited to commercialize any invention itself. But rather than hand an invention’s promise of profit over to private interests, Long recommended that the United States establish an entity similar to that established by the British and the Canadians, viz. "a ‘Federal Inventions Authority’ to administer all Government-owned patents", charged with "protecting the public interest in scientific and technological developments achieved through the activities of departments and agencies of the U.S. Government", as well as "with the dissemination of knowledge so developed in order to stimulate invention and innovation which will cut costs, produce new products, and increase per capital industrial production through efficiency and new technology." "Generous monetary awards as well as public recognition" could be substituted for the incentives to invention that might otherwise exist with patent ownership.

In the midst of the debate, in October of 1962, NASA asked for public comment on the new licensing regulations that had so aroused the ire of the aerospace industry. NASA announced that it would "offer revocable, non-exclusive, royalty-free licenses under its patents for two years. If, after the [NASA] patent is two years old, no one has developed the invention commercially, NASA will make the invention available for the grant of an exclusive license. . . . In no case, however, will a contractor responsible for an invention be denied his right to use it." Furthermore, the agency would favor issuing waivers of exclusive rights to practice inventions made under NASA contracts to its industrial contractors.

This made no one happy. "Public Rights" advocates, led by Senators Long and Estes Kefauver (D-Tenn) persisted in their arguments, maintaining that consumers were taxed twice when they paid for government-financed research, and then had to pay monopolistic prices for the products of that research. Long’s and Kefauver’s opponents, the Electronics and Aerospace Industries associations and the National Association of Manufacturers, persisted in their argument: government retention of patent rights to inventions made by federal contractors was tantamount to theft, and would discourage the best companies from contracting with the government.

Senators Long, Kefauver, and Morse were talking as if they meant what they said, and they happened to belong to the same party as the incumbent in the White House. The time had come for the White House to report on the results of its own foray "through the tangled underbrush of Government patenting policies." At last, on October 12, 1963, without proposing any change in existing law, the Kennedy Administration took its position "between a rock and a hard place." The Administration’s 1963 guidelines "normally" reserved to the government "principal or exclusive rights throughout the world" to any of its contractor’s inventions if the "products, processes, or methods" involved were "intended for commercial use," or if "a principal purpose of the contract is for exploration into fields which directly concern the public health or public welfare," or "where the government has been the principal developer of the field, and the acquisition of exclusive rights at the time of contracting might confer on the contractor a preferred or dominant position." In other words, if the "public good" required unlimited use of the invention, the government would retain title. On the other hand, the contractor would "normally acquire the principal or exclusive rights throughout the world" to any of its inventions if "the work called for by the contract is in a field of technology in which the contractor has acquired technical competence" or has otherwise established a "non-governmental position."

Since NASA was unlikely to award a contract to a firm that was not technically competent to carry out the work required, the new guidelines seemed to ensure that the contractor would be allowed to claim rights to any invention made under a government contract. However, if two or more potential contractors offered equally meritorious proposals, the "willingness to grant the government principal or exclusive rights in resulting inventions will be an additional factor in the evaluation of proposals." The policy also gave the Government "march-in rights" to require a contractor who had failed to bring an invention "to practical application" within three years to grant a non-exclusive, royalty-free license to another applicant. Government agencies were directed to publicize their patented technologies, making them available for licensing. Finally, the policy statement tacitly conceded what Mr. Quigley had observed three years before, and the President’s own science advisor Jerome B. Wiesner later confirmed: no one really knew whether any one policy would work as intended. The Federal Council for Science and Technology (FCST), a White House interagency coordinating group, was tasked with finding out.

Senators Long and Morse were not happy with this new policy either. They were joined by Senator Philip Hart (D-Mich), who introduced legislation requiring public ownership of federally financed inventions. Democrats in the Senate were divided on the issue; some, like Senator John McClellan (D-Ark.) supported the Kennedy guidelines. As a result, debating bills and amendments to bills, arguing over whether NASA should be authorized to grant patent waivers to contractors and grantees, appears to have become something of a Senatorial pastime for much of the 1960s’. The Department of Justice, which had maintained since 1947 that patent rights to an invention should be assigned to the government if the government paid for the research producing the invention, now supported liberalization. Nonetheless, Attorney General Nicholas Katzenbach admitted to the Senate Small Business committee that he knew of no "data, studies, or facts of any kind at all, which could possibly support" the notion that giving patent rights to contractors would "foster `the prompt working of inventions’."

The trend in U.S. government patent policy toward relying on the private sector to exploit government-funded innovation gained strength with the arrival at the White House in early 1970 of the first Republican president in a decade. Richard M. Nixon was unlikely to reverse this trend. In August of 1971 Nixon released a Statement on Government Patent Policy which, he explained, reflected changes called for as a result of "studies and experience over the past seven years". While much of the revised "Statement" repeated language from the Kennedy patent policy of 1963, the Nixon policy allowed the heads of government agencies to grant exclusive licenses to commercial developers of inventions made under government contracts, subject to a loose array of criteria likely to be met by anyone seeking a license. NASA followed suit with the first (for any government agency) publication of its semi-annual NASA Patent Abstracts Bibliography, which contained abstracts of over 1,892 NASA-owned inventions available for exclusive licensing. By the beginning of January, 1974, the Agency announced the availability of 2,600 of its patents for exclusive and non-exclusive licensing.

NASA patent policy became the law of the land in 1980, with the passage of amendments to the Patent and Trademark Laws (The Bayh-Dole Act, Public Law 96-517). "Bayh-Dole" authorized federal agencies to grant "exclusive or partially exclusive licenses in any invention covered by a federally owned domestic patent or patent application". However, before a federal agency could issue an exclusive or partially exclusive license, it would have to determine that a "practical application" of the invention in question had not been achieved under any nonexclusive license. The agency would also have to determine that the proposed license would not "tend substantially to lessen competition or result in undue concentration in any section of the country in any line of commerce to which the technology to be licensed relates, or to create or maintain other situations inconsistent with antitrust laws." Thirdly, "preference" in issuing exclusive or partially exclusive licenses to Federally owned inventions would have to be given to small business firms that had demonstrated their ability to successfully develop and market the invention. The government retained its right to a royalty-free license to practice any invention ‘based on and in inspired by’ information received by the contractor under the work of the contract.

Small businesses and non-profit organizations were winners under the Bayh-Dole amendments to the U.S. patent and trademark statutes, which allowed "each nonprofit organization [including universities] or small business firm" to "elect to retain title" to any invention made under a "funding agreement" (i.e., a contract or grant) with the Federal government. Exceptions were limited to instances when retaining title might impede U.S. foreign intelligence or counter-intelligence activities, and "when the funding agreement includes the operation of a Government-owned, contractor-operated facility of the Department of Energy primarily dedicated to that Department’s naval nuclear propulsion or weapons related programs." The Federal agency under whose funding agreement the invention had been made was given "march-in" rights to issue an exclusive license to a third party, "upon terms that are reasonable under the circumstances," if the original contractor or assignee failed to take, "or is not expected to take within a reasonable time, effective steps to achieve practical application of the subject invention . . . ."

In 1983 President Ronald Reagan followed suit with his own Memorandum on Government Patent Policy, extending to any federal research and development contractor the same right to elect to patent inventions made under their funding agreements with the Federal government as Bayh-Dole extended to non-profit organizations and small businesses. In the ideological battle between government and the private sector over who could best exploit innovations funded by the government, the private sector had won.

In the case of NASA, the appearance of public ownership was preserved by the retention of the original "Space Act" provision that the agency would be the presumptive owner of any intellectual property created by its contractors and grantees. But the agency’s patent waiver policy, viz., any request for waiver to a contractor or grantee of NASA patent rights would be readily granted, enabled the private sector to take out of the back door what it was not offered at the front.

C. Who Benefited Most From NASA’s Patent Policy?

As an accompaniment to the growth of spending for scientific and engineering research in both public and private sectors, policymakers and economists sought to devise ways of gauging the returns on R&D investments. While the effort had begun in the 1930’s, the Cold War era saw the most sustained attempts to measure the social and economic benefits of R&D spending. And because patenting rates are one of the few quantifiable, direct "outputs" of innovation, patents became an important item in the social scientist’s tool box. However, counting patents were and remain a controversial means of ascertaining institutional inventiveness. A single patent--for example, U.S. Patent No. 821,393 granted in 1906 to Orville and Wilbur Wright for a "flying machine"--might document an invention of revolutionary importance and immeasurable profits for its eventual developers. On the other hand, diminishing numbers of patents could indicate simply that a product (and its supporting technology) has matured, which does not necessarily mean that it has become unprofitable. Generally, however, multiple patents are widely regarded as the better indicator of "high interest and a growing technology." To the extent that patents, or rights to manufacture patented products, are themselves commodities that can be bought and sold, increasing patents do indicate increasing returns on investments in research and development.

As required by President Kennedy’s 1963 Statement of Government Patent Policy, the Federal Council for Science and Technology (FCST) issued its first "Annual Report on Government Patent Policy" in 1965. Including statistics describing quantitatively the patent operations of the 13 Federal R&D agencies for fiscal years 1963 and 1964, the FCST report noted inconsistencies across agencies in collecting their own patenting and licensing data­some agencies made actual counts, while others reported only estimates; some docketed contractor invention disclosures, others did not. Otherwise the report was brief. Its most notable conclusions were that (1) it was too early to tell whether the Kennedy policy had been effective, (2) not all agencies had submitted their own patent regulations to FCST’s Patent Advisory Panel for review, and (3) preliminary indications were that "the granting of limited exclusive rights to the [government-owned] inventions within the United States might serve to draw forth the private risk capital needed to obtain public promotion and utilization of the inventions in this country."

Meanwhile, studies conducted by Mary A. Holman and others in the early 1960’s of the commercialization of government owned patents, as well as NASA patents, concluded that the contribution of federally funded R&D to commercially successful inventions was insignificant. The limitation of these studies, however, was the short period of time they covered, and the difficulty of documenting and tracing the dissemination of Federal innovations throughout the marketplace.

NASA’s Office of General Council also issued reports during the mid-1960’s of the volume of the Agency’s patenting and licensing activity, and of anecdotal case-studies in which NASA contractor inventions had produced commercially valuable products. Without extensive and corresponding non-NASA data for comparison, and without knowing how representative the case studies reported were of waived NASA patents in general, one could not conclude much of anything about these reports except that NASA’s Office of General Council had reported what it could.

Economists’ and other social scientists’ studies of technology dissemination in the private sector yielded more useful information, possibly because larger data sources enabled more methodologically rigorous use of systematically collected data. Prior to World War II Simon Kuznets and Robert Merton were among the more notable social scientists to attempt to trace the rate and dispersion of technological change through patent data. Their work produced the observation that technological improvements in particular industries reach points of diminishing returns. After World War II further work on patents was done by Jacob Schmookler, who found that the stimulus to the inventions he surveyed could be identified in very few cases, and

...For almost all of these that stimulus is a technical problem or opportunity conceived by the inventor largely in economic terms.... When the inventions themselves are examined in their historical context, in most instances either the inventions contain no identifiable scientific component, or the science that they embody is at least twenty years old. Schmookler was decoupling the linear causal link between science and technology which

had become an ideological axiom among many in the natural sciences, ever since Vannevar Bush asserted that generously funded "basic" science would assure us an "endless frontier." In so doing, he reasserted the primacy of economic incentive as a source of technological innovation -- or the importance of "market pull" vs. scientific "push" in accounting for technological advances. Schmookler’s work skillfully combined quantitative data (e.g. patents) with qualitative information (e.g. assessments in trade and technological literature) to answer not only the question, "How Much?", but the related and more important question for policy-makers, "How, and to what affect?"

Complete patent data for any organization allows us not only to aggregate patents, but can serve as a comprehensive catalogue of its patentable inventions which, in turn, permits random sampling and the beginnings of systematic -- and thus generalizable -- qualitative analysis of what those patents might tell us about who invents, under what circumstances, and with what results. Thanks to the U.S. Patent and Trademark Office, we now have access to an on-line treasure trove of patent data, allowing us to randomly sample individuals’ or organizations’ patented inventions since 1976. Combined with licensing and commercialization information, these generalizations can help us locate enterprises, whether large or small, having the greatest potential for R&D investment return.

Using the USPTO’s database, NASA’s Office of Policy and Plans has developed a secondary database of all patents assigned to NASA from 1976 to 1996, a 20 year period that averages local variations, brings us as close to the present as possible, and covers the period between the post-Apollo let-down in NASA’s budget and the ramp-up for the Shuttle program. With the help of NASA’s patent counsels, we added to this database the licenses issued by NASA to various firms to use patents assigned to the agency during this period, along with the classifications to which the USPTO’s examiners had assigned each and every NASA patent. By analyzing the patent classifications represented by the more than 2600 patents assigned to NASA between 1976 and 1996, one can begin to make some interesting observations about what happened to federal tax dollars dedicated to aerospace R&D in NASA.

To recall the federal R&D context, consider the portion of national spending for R&D represented by investments in NASA R&D. From 1960 to 1995, both public and private U.S. spending for R&D grew from $13.5 billion to $171 billion, or by a multiple of 13. The federal government’s share in 1960, including money passed through industrial and academic institutions, was $8.7 billion, or about 65%. By 1995, the federal government’s share of national R&D spending declined to 35%, about half of what it had been 35 years earlier. The most significant shift in relative funding for R&D by the federal government began in the early 1980’s, having dropped below 50% by the time Ronald Reagan was elected to office (1980).

From 1976 to 1996, the federal government’s share of all patents issued by the U.S. Patent and Trademark Office averaged around 3%. This seems a low figure, until we realize that R&D represents only a small fraction of federal activities and outlays. For example, in 1960 the net outlays of the federal government exceeded $92 billion, of which only about 10% went to R&D. By 1980, the portion of the federal budget going to R&D had dropped to 5%, and in 1995 that portion had slipped to less than 5%. Secondly, with few exceptions (NASA being one of them) federal agencies do not automatically retain property rights in the inventions made by their contractors or grantees. So, the 3% of all U.S. patents assigned to all agencies of the federal government is probably as good a baseline as any.

Meanwhile, NASA’s R&D budget, which was about 13% of the federal R&D budget in 1960, dropped to less than 5% of federal R&D expenditures in the early 1980’s. However, in the last ten years it has recovered steadily, and in 1996 was a little over 10% of the federal R&D budget. What level of patenting did NASA have over this period? Two decades of NASA invention and patenting is long enough to discern trends which, in turn, can serve as another baseline. Between 1976 and 1996, NASA’s portion of all patents assigned to the federal government averaged 10%, or three times the proportion of U.S. patents assigned to the federal government. This percentage was not constant throughout the 20-year period, however. It dropped below 10% during 1979-1982, again in 1986, and most recently, in 1995-1996. NASA’s best patenting years, as a percentage of total federal patenting, were 1983-85 and 1989-93. Given the time lag of several years between invention and patent assignment, these two "spurts" most likely resulted from R&D activity associated with the Space Shuttle and International Space Station programs.

Another data source of indeterminate but potential value -- one that is not readily available for the total federal government -- is the number of waivers NASA has issued to private sector firms of rights to NASA inventions. (Through 1985 NASA’s Inventions and Contributions Board denied only 10% of all patent waivers requested; between 1985 and 1995, the board denied none.) An application for a patent waiver normally reveals the nature of the technology that the contractor suspects may have value -- whether for its own development, or to protect its business position in a particular market segment. While NASA received 2,620 patents to its own inventions during 1976-1996, the Agency also waived patents rights to 1,716 inventions to which it might otherwise have retained title, representing an increase in patenting activity of 65%, over and above the 2,620 patents assigned to NASA. (By comparison, only 13% of NASA’s patents retained during the period have been licensed.) Without comparable patent waiver figures for other federal agencies, we cannot compare NASA’s total patenting activity to that of the entire federal government.

Since patents have no inherent or uniform value, owning twice as many patents doesn’t make us twice as capital-rich, today or tomorrow. Aggregated numbers of patents give us, at best, a rough indication of an organization’s intellectual property (but not necessarily its inventiveness). However, average numbers of patents assigned, as a variable of average R&D investments over a significant period of time, can give us a proximate idea of the nominal return on those R&D investments. (This is so much in the same way that we might observe that the average family in a certain country has 2.37 members, while not ever expecting to meet a .37 family member.) Similarly, we can compare average nominal dollars invested in R&D and average nominal patent yields. In the case of the Federal government, the annual R&D expenditure (as both source and performer), between 1976 and 1996, averaged $62.1 billion. This figure can be used to gauge the direct public "cost" of Federal R&D. During this same period, the overall Federal government averaged 1,311 patents per year, at an average nominal public cost of $50 million each, while the direct public "cost" of each NASA patent averaged $40 million.

Now, if you paid for the R&D behind the patent that I own, then the nominalvalue to me of that patent on the day I acquire it (not counting legal and associated fees), is at least what you paid that I did not have to pay. I might have been able to get the R&D done cheaper, or it might have cost me more. But at least at the point that I acquired the patent, before I invested in bringing a product to market (the outcome of which I can only speculate), what you paid for the R&D behind the patents represents their nominal value to me in my own R&D costs avoided. Between 1976 and 1996, NASA issued 1716 patent waivers to its contractors and grantees. As the average nominal cost of a patent resulting from NASA funded R&D, is $40 million, so the aggregate nominal value to the recipients of 1716 waivers of NASA patent rights was $68.8 billion, a sum which equals 64% of the total NASA R&D budget for the two decades. Keeping in mind that the primary purpose for which the Congress has appropriated R&D dollars to NASA has been to carry out various aerospace science and engineering programs (most of which were executed on contracts by the private sector) $68.8 billion over 20 years is a nice return on the aerospace industry’s investment in influencing NASA patent policy in the 1960’s. The extent to which, and how, the $68.8 billion resulted in economic or social benefits to the public at large depends on one’s appraisal of the social value of the aerospace industry’s products. Certainly, it is not negligible. Effective national defense and a strong and large export sector (i.e., aircraft) are significant national assets.

Beyond this -- that is, to ascertain the consequences of patent policy for the content of a nation’s innovations -- we need to look further, at the classifications to which NASA patents were assigned by their USPTO examiners. A "patent class" is a category, defined by technological and functional principles, to which patent examiners assign patent applications. Classifications ease the task of seeking out "prior art" for an invention. Using the USPTO database, we determined the patent class distribution of the 2,620 patents assigned to NASA from 1976 to 1996. By simple rank order, the overwhelming number of NASA’s patented innovations during this twenty-year period have occurred in "measuring or testing" (or research instrumentation), and aeronautics. After that, the numbers appear to go "all over the map." To make a little more sense of this map, we grouped these classes into technology clusters. What we see clearly from these clusters is the extent to which NASA’s innovations have resulted not only in producers’ goods, but goods for a distinctively post-World War II sector, the R&D sector.

The significance of this finding is twofold: (1) The national R&D enterprise has become a discrete economic sector, serving itself in variable degrees, and, (2) technological improvements in instrumentation and scientific discovery are concurring phenomena. Thus the largest area of technology affected by NASA innovation is not peculiar to aeronautics or space technology, but belongs to the whole universe of scientific and engineering research and development. Its benefits to the general public about which Senators Long, Kefauver and Morse were so concerned, were secondary at most, and depended (as they still do) on the allocations of the market place, which is not known for its social conscience.

Next in frequency of appearance among NASA patent classifications are inventions in the general area of energy generation or propagation -- both electrical and chemical. After that, aeronautics; and after aeronautics, materials (in which we have included biochemical processes), then stock materials or miscellaneous articles. Bringing up the rear of patent classes in which NASA has received 5 or more patents throughout the 20-year period are communications and information processing technologies (excluding software). The relative infrequency of patents in communications and information technologies is most interesting since these technologies are the ones popularly thought to be the chief engines of modern economic, political, and social change. However, advances in materials, biochemical processes and products, and energy technologies may turn out to have been more central to our industrial capacity at the end of the 20th century. The most important distinction to be made, however, is between consumer goods -- which are highly visible to the public -- and producers’ goods, which can have a substantial long-term impact on the nation’s economy but are generally invisible to the general public.

The most objective available documentation of private sector interest in NASA inventions are NASA’s patent licenses. NASA’s intellectual property files document 338 licenses issued during 1976 to 1996, representing 13% of all patents assigned to the agency. Is that a lot? Until we have a better feel for what a "good" licensing rate for an R&D organization with vested patent rights is, we cannot know whether 338 is "good" or "bad", or indicates a "strong" or "weak" interest in NASA technologies. Setting that arguable question aside, does commercial interest in NASA patented innovations, as indicated by licensing, follow the same distribution among technologies as NASA patents?

Comparing the frequency distribution of NASA’s licenses among patent classes with the frequency distribution of NASA patents among those same classes, we find aeronautics far down the list with less than five licenses. This is not a great surprise, since the aircraft industry probably owns most patents in aviation technologies and should have little need to license NASA patents in that same field. But, note also that communications and information processing systems have generated relatively little licensing interest, just as they generated relatively few NASA patents. The NASA inventions that have stimulated the most licensing interest -- and thus commercial interest -- are those involving electric motors, synthetic materials and adhesives, bio- and physical chemistry processes, instrumentation, and -- the surprise on the list -- surgery. (But perhaps surgery should not be such a surprise, if we consider the necessary interest of NASA engineers on miniaturization of instruments and mechanisms.) So, we might ask: Could it be that Kuznets’s observation about pre-World War II technological innovation -- that, within industries, it reaches points of diminishing returns -- now applies to innovation in communications and information processing industries? As new industries mature, do such components as capitalization, market strategy, and entrepreneurship become more important than technological innovation?

While completed case studies of randomly selected NASA inventions are still a few turns down the road, the necessary interim analysis of NASA patent, patent waiver, and NASA licensing data has already yielded some useful information for those who would make policy and manage organizations to promote innovation. Our analysis of NASA patenting and licensing, 1976-1996, has produced:

(1) a baseline for NASA patenting of 10% of patents assigned to the U.S. government, a baseline drawn from 20 years of NASA R&D;

(2) a baseline for NASA patent waivers, or 65% of NASA patents, which represent a recent 20 year average (If we want to do "more" or "less," we can answer the questions, "more than what?" and "less than what?" And do what purpose?);

(3) a baseline of $40 million (unadjusted for inflation) as the nominal average cost or dollar value of a NASA patent, which was the Agency’s average annual R&D budget during 1976-96 divided by the number of patents, on average, assigned to NASA annually during the same period; and

(4) a baseline of $68.6 billion as the nominal value of 20 years’ of patent waivers issued by the Agency to the aerospace and associated industries and academia, representing $68.6 billion in R&D costs avoided by the waivers’ recipients -- NASA contractors and grantees.

We can also make some preliminary observations about the technology areas in which NASA patenting and licensing has occurred. These observations indicate where the "pay-off" from NASA R&D investments can be found, beyond the completion of the aerospace programs for which NASA funds were originally appropriated. That pay-off is not primarily in the consumer marketplace of popularly recognizable products and services, but in the capital goods marketplace, where one acquires instrumentation for research, development, and manufacturing; biochemical and other chemical processes; new ways of harnessing energy; and in medicine, where we seek sophisticated new substances, mechanisms, and instruments.

Moreover, we now have a comprehensive and complete 20-year agency patent data-base and a virtually complete list of NASA licenses to use these same patents, which together constitute a data set from which to randomly (rather than anecdotally) select the case studies necessary to form generalizations about how patent policy influences innovation, and how that innovation percolates through the marketplace. This, as much as the numbers themselves, may be the real prize for tackling the difficult challenge of determining the merit of NASA (and now Federal) patent policy.