inventions that others might have been willing go develop had the Government held title. Experience indicates that such fears are largely unfounded and that, in any case, even if the Government held title, the likelihood of other firms developing most inventions would be small without some exclusive rights. It would seem that the Government would have to do more than merely take title. It would, in turn, in most cases have to grant someone else an exclusive license. But it could be asked what advantage there is to going through the cost and effort of such an exclusive licensing effort as opposed to allowing the inventing contractor, who is intimately familiar with the invention to retain rights with appropriate march-in provisions if no work is done on the invention to commercialize it?

For a variety of practical reasons, it would be a mistake to believe that a title-inthe-Government licensing approach could be as effective in promoting utilization as leaving title-in-contractors. First, as mentioned previously, a title-in-the-Government approach might eliminate the incentive for many grantees and contractors to report inventions. In the case of the university community it is the principal investigator who normally starts the process moving by identifying inventions. Since publication, and not patents, are critical to the careers of university investigators many are not motivated to report inventions. 26 However, this can be overcome by aggressive programs at the university to induce reporting, especially by an active licensing program that offers some possibility of financial reward for the inventor. Such incentives to the inventor are completely lost when the Government automatically takes title. Within the business sector, a similar decrease in reporting might result, although probably for different reasons.

Secondly, the government would be faced with an enormous increase in workload. For example, under a title-in-the-Government approach DOD would be faced with some 1500-2500 inventions a year on which a decision would have to be made concerning the filing of patents. If DOD continued to base that decision solely on potential military applications, it ought to be obvious that patent applications will not be filed on a number of inventions that have commercial potential but not military potential. Therefore, if one is to honestly argue that a title-in-the-Government approach will not have negative impacts on innovation, one must be prepared to say that DOD and other agencies must screen invention disclosures for commercial application, a task which is now done by DOD contractors who have the opportunity to elect rights (i.e., a first option on rights to inventions). However, that would require a substantial increase in DOD staff and resources devoted to such task.

To duplicate the efforts now undertaken by many contractors and a number of universities, the Government agencies would have to be prepared to discuss the inventions with various industrial experts, to run patent searches, and to undertake a substantial amount of sophisticated market and technology analysis that is beyond their normal missions and capabilites.

Thirdly, Government licensing efforts will be hampered by the fact that the Government will not have available to it the expertise and know-how of the inventor and the technical team that conceived the invention. Successful patent licensing often requires transfer of more than a bare right in patent. Agreements to provide technical assistance may be required which the Government could not offer. Moreover, in the case of many inventions coming from the larger firms, the invention may simply be an improvement on existing technology controlled by the inventing firm. Because of the existence of dominant background patents, the invention will be of no use to anyone but the inventing corporation.

Fourthly, it is not always obvious at the time an invention is made that it will ultimately have commercial importance. In many cases, it is the perserverence of the inventor or other technical personnel with the firm who foresee an invention's possibilities that persuades a company to go ahead with development.

For example, Battelle Columbus Laboratories did a study to identify the factors which influenced the movement of ten current technologies from their original conception state into actual use. They concluded: "The technical entrepreneur, whose importance was highlighted in the study of the "factors", is also a "characteristic" important in nine of the ten innovations. This is the strongest conclusion that emerges from the study. In fact, in three innovations, the technical entrepreneur persisted in the face of the inhibiting effect of an unfavorable market analysis. If

26 In this regard it should be kept in mind that about two-thirds of the basic and applied extramural research supported by the Government is performed by universities and other nonprofit institutions.

any suggestion were to be made as to what should be done to promote innovation, it would be to find—if one can, technical entrepreneurs."27

We believe "technical entrepreneurship" will largely be lost under a title-in-theGovernment approach. Accordingly, it is unreasonable to believe that Government licensing would be as effective in promoting the development of contractor inventions as leaving title-in-the-contractor.

Impact of patent policy on public health

The following discussion provides a clear case study of the impact of patent policy since one can compare the results of the Department of Health, Education, and Welfare's (DHEW) pre-1968 title-in-the-Government oriented policy with its experience since that time when a more title-in-the-contractor oriented approach was adopted.

28

A significant portion of Government R&D is devoted to medical research." DHEW, NSF, the Department of Agriculture and to a lesser extent other agencies such as DOD and the Veterans Administration support extramural research in the medical life sciences. Out of such research new compounds are often synthesized which may have pharmaceutical potential. Experience at NIH and studies by the General Accounting Office 29 and Harbridge House 30 clearly support the conclusion that a title-in-the-Government patent policy that did not make an exception for medical research would endanger the public health. However, proponents of a titlein-the-Government approach have never suggested that medical research be excepted from the policy. Indeed, even the President's Statement on Government Patent Policy unfortunately specifically singles out health as an area in which the Government should take title.

The GAO and Harbridge House reports noted above, which were based on extensive interviews with National Institutes of Health grantees and staff, concluded that the pharmaceutical industry would not utilize its risk capital to pursue further development of potential pharmaceutical agents generated with DHEW support without a guarantee of some patent exclusivity. (With the passage of the Medical Devices Act of 1976, which requires premarket clearance of many medical devices, it is becoming increasingly apparent that the same need for patent protection applies to the medical device area.) In some situations, the GAO discovered investigators with hundreds of compounds with potential therapeutic value on their shelves with no source to test their market potential. The GAO criticized DHEW for its failure to use its discretion to enter into Institutional Patent Agreements (which it had not done since 1958) or to make timely determinations of rights after identification of inventions.

Since 1969, when DHEW began using its discretion as suggested by the GAO, until the fall of 1974, DHEW estimates that the intellectual property rights to 329 inventions made in performance of DHEW-funded research were being managed by institutions with Institutional Patent Agreements (IPA) or by successful nonprofit petitioners for the purpose of soliciting further industrial support. During this period, these organizations have negotiated 44 nonexclusive and 78 exclusive licenses under patent applications filed on the 329 inventions. Since 1974, to the end of fiscal year 1976, the number of inventions held by such organizations has increased to 517. DHEW estimates that the risk capital generated under the licenses on these 517 inventions has been approximately $150,000,000.31

An example of the stake which an innovating company has in the development of a new drug may be seen from the following. Prior to the 1962 amendments to the Food and Drug Laws the average cost of developing a new drug was estimated to be $534,000.32 The development cost of a new drug in 1973 was estimated to be $11,500,000, a figure which escalates to $24,400,000 when the cost of research on projects which do not result in marketed drugs in included. With these figures in

27 Battelle Columbus Laboratories, "Science, Technology and Innovation, Summary Report", Feb. 1973, p. 8. 28 Over one-third of the Federal R&D budget for basic research in FY 1977 went for the life sciences which include medical and related research.

29 GAO Report B-164031. Ibid (Note 7)

30 Harbridge House Report. Ibid (Note 9)

31 Science Policy Implications of DNA Recombinant Molecule Research. Hearings before the Subcommittee on Science, Research and Technology of the Committee on Science and Technology, U.S. House of Representatives, 95th Cong., 1st Sess. (No. 24), p. 965. It should also be noted that over 60 percent of the inventions retained by IPA holders or petitioners have not yet been licensed and many will never be licensed or brought to ultimate use. Accordingly, the mere retention of patent rights is clearly no guarantee that commercialization will occur.

32 Scherer, "The Economic Effect of Mandatory Patent Licensing," p. 59, U.S. Energy Research and Development Administration, Public Meeting 11277.

mind it is little wonder that the return on R&D investment in the drug industry has dropped sharply since 1960 (it is_currently calculated to be 3.3 percent).33 The May 26, 1977 testimony of the Patent Counsel of DHEW, given before the Subcommittee on Science, Research and Technology of the House Committee on Science of Technology includes examples of inventions which have been licensed by universities and nonprofit organizations that have reached or are near reaching the market place. As noted in that testimony most of the examples are pharmaceutical products and medical devices. No comparable examples were known at the time the GAO and Harbridge House ran their studies.

This experience strongly supports the general proposition that the less restrictive the patent policy the greater is the transfer of technology.

Why does such a first-option-in-the-contractor policy promote the transfer of technology:

1. It reduces the uncertainties as to the status of invention rights and thereby permits:

(a) the prompt filing of appropriate patent applications by the contractor-grantee; (b) an early effort by experienced technology transfer groups and patent management organizations to locate and engage private enterprise in further development of inventions;

(c) an early decision by the industrial developer that the intellectual property rights in the innovation being offered are sufficient to protect its risk investment. 2. It is a recognition by the agency that the nature of the research being supported through funding under a grant or contract is fundamental or basic and that inventions and the making of them are by-products of and not a specific object of the grant or contract.

3. It is a recognition that any invention evolved will require further development to bring it to the marketplace-development which should involve private enterprise since under our free enterprise system private parties and not the Government should engage in such activity.

4. It provides motivation for a contribution by a commercial organization, in cash or in kind, to Government-funded research projects-the certainty of the grantee (contractor) having the first option to any invention arising from such project providing the basis for this now recognized attitudinal change by industry.

5. It provides a climate which encourages the investigator-inventor's continuing participation in the transfer of his inventive technology to the public-a particularly important consideration where university-generated inventions are involved since such inventions tend to be embryonic in nature.

6. It more fairly recognizes the equities and contributions of all of the parties to the inventive technology.

7. It provides the opportunity for the university-contractor to generate income as consideration for the technological innovation being offered, which income is earmarked to support further research at the university—the public thus benefits a second time.

8. It permits timely consideration to be given to foreign patent protection and thereby enhances the possibility of generating payments from foreign sources for the transfer of the patented technology under license with an attendant favorable impact upon the balance of trade.

Impact of patent policy on economic growth and jobs

It should be obvious that without the introduction of new products into the economy, economic growth and job expansion would come to an eventual halt. While people can disagree whether particular technological innovations are good or bad, we doubt that anyone would seriously argue that a slow-down in technological innovation would not result in slower economic growth. Yet, the fraction of R&D performed in this country that is Government supported has now reached around two-thirds. Hence, it is inescapable that a Government patent policy that discouraged investment in the development of the inventions made during that research would have a negative effect on economic growth.

Although the relationship between innovation and long-term economic growth and job expansion are intuitively and historically obvious, several studies serve to highlight this.

A 1967 Department of Commerce study 3 and a more recent update of that study by John Flender and Richard Morse of the MIT Development Foundation, Inc. 35

33 Schwartzmann, "Innovation in the Pharmaceutical Industry," p. 70.

34 "Technological Innovation: Its Environment and Management", U.S. Panel on Invention and Innovation. (Washington, D.C. GPO, 1967).

35 John O. Flender and Richard S. Morse, "The Role of New Technical Enterprises in the U.S. Economy", M.I.T. Development Foundation, Inc., Oct. 1, 1975.

lend strong support to the proposition that sales growth and job creation occurs more rapidly in innovative companies than in mature (dominant) companies. And even more significant for purposes of this analysis is the fact that job expansion at young (i.e. small) high technology companies was even more spectacular. For example, the authors found that during a five year period six mature companies with combined annual sales of $36 billion in 1974 experienced a net gain of only 25,000 jobs during the five years, whereas five young, high technology companies with combined sales of only $857 million had a net increase in employment of 35,000 jobs (five "innovative" companies with $21 billion sales total had a net increase in employment of 106,000 jobs). These findings indicate that a patent policy that would deemphasize the needs of smaller firms and emphasize concerns with larger and more dominant firms could have a negative impact on job expansion.

The potential harm that could accrue from discounting the need to be concerned with inventions from nondominant firms is further emphasized by a study done by Gelman Research Associates. An international panel of experts selected the 500 major innovations that were introduced into the market during 1953-73 in the U.S., U.K., Japan, W. Germany, France, or Canada. Of the 319 innovations produced by U.S. industries, 24 percent were produced by companies with less than 100 employees. Another 24 percent were introduced by companies with 100 to 999 employees. Inasmuch as it seems apparent from the foregoing discussion that a first option to title to inventions in the contractor is much more likely to bring about innovation, it is indisputable that it is also much more likely to encourage economic growth and job expansion.

Impact of patent policy on foreign competition

American industry is in increasing competition with foreign corporations in hightechnology areas and a title-in-the-Government patent policy must inevitably work to the advantage of foreign firms at the expense of American industry and labor. The taking of title by the Government will effectively prevent the American inventing corporation from obtaining foreign patent protection. Without Government foreign filings no American firm could gain any exclusive rights in foreign markets. Moreover, historically, the Government agencies have had neither the incentive, the staff, the budget, nor sufficient knowledge of market conditions to file for foreign patents in anything more than a small number of cases.36

If the Government takes title to U.S. rights in inventions and dedicates them, these inventions are equally available to foreign based firms that would export commercial embodiments of these inventions into the U.S.

If one combines these facts with the difference in the relationship between business and Government in certain foreign countries as compared to relations in the U.S., certain disturbing implications arise. In some foreign countries industry is highly socialistic and state controlled. In others, major companies may enjoy state subsidies and support. The result of all this is that the same invention that U.S. firms may not develop without the exclusivity afforded by patent rights may be developed by Japanese, German, or other foreign firms that enjoy monopoly advantages in their home markets through means quite apart from patents. In turn these products are exported into the United States and displace American products and American jobs.

In short, given the difference in industry-Government relations in many of the technologically advanced foreign countries as compared to the United States, a titlein-the-Government policy is most likely to favor foreign companies. The mere speculative concern that there might be a few isolated cases where leaving title in a contractor might result in activities by that contractor which are in violation of the antitrust laws should not control the whole of Government patent policy when other remedies are available through those laws. The U.S. economy does not operate in a vacuum and to formulate a policy of title-in-the-Government primarily upon hypothetical and mistaken concerns about the impact that policy will have on competition within the United States ignores the many adverse effects such policy would have.

The case-by-case approach

Where the disposition of patent rights in inventions made in whole or in part with Federal funds is deferred until there has been an identification of the inven

36 Statistics by the Committee on Government Patent Policy show that during the period of Fiscal year 1970-75 the Government filed for foreign patents on an average of 77 contractor inventions, and the preponderance of these were by only two agencies, DOE and NASA. This is approximately one-tenth the number of contractor inventions upon which the Government filed U.S. patent applications.

tion the certainties associated with title-in-the-contractor are lost. As pointed out before such certainty is a strong incentive to the transfer of technology.

Thus, any attempt to transfer the inventive technology would have to await the decision by the Government that title will be left with the contractor. As a consequence, valuable time will be lost in transferring the technology because it is seldom that a bureaucratic decision is made expeditiously. Moreover, the administrative and associated paperwork burden in the deferred determination approach militate heavily against the viability of this approach as a realistic alternative to the title-in-the-contractor approach. For example, a preponderance of DOD contracts now include clauses allowing the contractor to retain patent rights. As was mentioned before, it is unlikely that DOD could expeditiously process each contractor request for patent rights under a deferred determination procedure with present staffing.

Deferred determination advocates would claim that the Government can make a better judgment after the invention is identified, and that exclusivity will not always be needed. Implicit in this claim is the assumption that Government personnel will either be in a position: (i) to determine if the existence of exclusive patent rights is needed as an incentive to further development; or (ii) to find a better qualified firm to commercialize the invention through a Government licensing effort after taking title to the invention.

In regard to the question of whether exclusivity is needed for private investment to be made in an identified invention, it should be recognized that if the Government determines that exclusivity is not needed but is wrong, no further development may take place.

Moreover, for the Government to be right more often than not when making a deferred determination would require extensive technical, marketing, and economic studies of the firms, technology, industries and market involved. The cost to taxpayers of such programs could be more than any savings they would produce for consumers. This appears to be true, since in most deferred determination cases exclusivity has been deemed necessary, and the costly determination process has been engaged in simply to confirm this fact. This has been substantiated in practice by NASA, DHEW and NSF (the three agencies who have historically made the largest number of deferred determinations) by the grant of over 90 percent of the requests for "greater rights" over a period spanning ten years.

Similarly, the ability of Government personnel to decide after an invention is identified that utilization will best be promoted by the Government's taking title and offering the invention for licensing, assumes that commercial developers, other than the inventing contractor, can be found (presumably but not necessarily on a nonexclusive basis). There is really no effective means for Government personnel to ensure that other firms, whether licensed exclusively or nonexclusively, would do a better job of developing the invention than a willing contractor or a licensee of the contractor. As noted previously, other firms often lack some of the "know-how" of the contractor and will not have the inventor or co-inventors working for them. One can be sure that in most cases the inventing organization will have little interest or incentive to transfer its know-how to another firm, possibly a competitor. Moreover, the very process of attempting to find alternative developers will simply serve to delay private investment and cool the interest of the inventing contractor. It will also force the Government into the expense of filing patent applications in order to assure that a patent is available if exclusive licensing is ultimately deemed neces

sary.

It is important also to emphasize that a deferred determination that is truly geared to resolve the questions that trouble opponents of the title-in-the-contractor approach would be so costly, complex, and time consuming as to discourage many contractors from requesting rights in the first instance, especially small busnesses and universities. They may even neglect to report the invention under such circumstances. In all likelihood, without a request for rights to trigger the deferred determination process, most agencies will have no incentive to do anything with the disclosure, and the invention will fall into the public domain to be available to all and, in most cases, practiced by no one, as seems to be the case with a very substantial portion of the 28,000 patents now in the Government's patent portfolio. Indeed, under a deferred determination approach the agencies would probably be devoting so many resources to those cases where rights were requested that they would have insufficient personnel or interest to study inventions and encourage development and marketing where rights were not requested.

Summary and commentary

It is believed apparent from careful examination of the impact of alternative patent policies on the various objectives of Government patent policy that the title