National Academies Press: OpenBook

Foreign Participation in U.S. Research and Development: Asset or Liability? (1996)

Chapter: Foreign Participation in Publicly Funded U.S. R&D

« Previous: Foreign Participation in Privately Funded U.S. R&D
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 90
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 91
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 92
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 93
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 94
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 95
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 96
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 97
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 98
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 99
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 100
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 101
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 102
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 103
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 104
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 105
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 106
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 107
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 108
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 109
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 110
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 111
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 112
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 113
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 114
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 115
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 116
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 117
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 118
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 119
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 120
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 121
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 122
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 123
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 124
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 125
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 126
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 127
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 128
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 129
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 130
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 131
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 132
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 133
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 134
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 135
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 136
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 137
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 138
Suggested Citation:"Foreign Participation in Publicly Funded U.S. R&D." National Academy of Engineering. 1996. Foreign Participation in U.S. Research and Development: Asset or Liability?. Washington, DC: The National Academies Press. doi: 10.17226/4922.
×
Page 139

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Foreign Participation in Publicly Funded U.S. R&D The past 10 to 15 years have also witnessed the growth of foreign involve- ment in publicly supported U.S. R&D activities at U.S. universities and federal laboratories.) This trend is less well documented than is the growth of foreign participation in privately funded U.S. R&D, and it has stimulated intense contro- versy. This is because these institutions, the two main pillars of the nation's publicly funded basic research enterprise and a unique source of national com- petitive advantage, are viewed as particularly vulnerable to foreign exploitation. At a time when many elements of the nation's innovation system are per- ceived to be faltering in the face of increasing global competition, publicly funded U.S. basic research capabilities remain internationally preeminent. This is par- ticularly true of those resident in the nation's academic institutions. In an effort to capitalize on this strength, the federal government is trying to enlist the re- search assets of universities and federal laboratories to bolster the competitive- ness of U.S.-based companies in world markets. However, the internationaliza- tion of U.S. industry and the general openness of the nation's basic research enterprise have heightened concerns about losing government-funded intellectual property and its associated economic value to foreign entities. The strength of the U.S. basic research enterprise is due primarily to its scale, its highly decentralized and pluralistic structure (i.e., multiple institutional play- ers with different research agendas), and its openness to the free flow of ideas and talent from throughout the world. Underlying these sources of strength is the fact that most basic research is conducted in publicly funded, noncommercial institu- tions with reward systems that provide powerful incentives for researchers to share rapidly and widely the results of their research. Ironically, these very 90

PUBLICLY FUNDED R&D 91 sources of strength make the U.S. basic research enterprise relatively easy for foreign nationals to access and exploit (National Academy of Engineering, 1993~. To date, three concerns have driven debate and policy on foreign involve- ment in publicly supported U.S. R&D, whether conducted by foreign firms, gov- ernments, academic institutions, or individual researchers. First, the United States may not receive an adequate quid pro quo for allowing various foreign entities to participate in publicly funded U.S. R&D. Second, some foreign governments appear to deny U.S. researchers reciprocal access to comparable publicly funded research within their borders. Third, some foreign countries do not appear to carry their fair share of the global basic research burden. Faced with these con- cerns, the American public and the Congress have become increasingly willing to restrict foreign access to U.S. government-supported research. This chapter examines the scope and nature of foreign participation in three areas of publicly funded R&D activity: research universities, federal laborato- ries, and federally funded industrial R&D programs and industry-led consortia. It concludes with an assessment of the costs and benefits of this participation. UNIVERSITY-BASED RESEARCH The primary locus of foreign participation in U.S. publicly funded research and development has been the nation's academic research enterprise home to roughly half of the nation's basic research activity. In 1994, American universi- ties and colleges are estimated to have performed $21 billion worth of research and development, or roughly 12 percent of total U.S. R&D performed that year. Also in 1994, academic institutions performed more than 48 percent of all basic research, 13 percent of all applied research, but less than 2 percent of all develop- ment work conducted in the United States. Sixty percent of all academic R&D conducted in 1994 was paid for by the federal government (National Science Foundation, l995b). Foreign entities establish ties with U.S. universities for many reasons. For- eign students, visiting scholars, and company researchers are drawn to U.S. re- search universities to learn about some of the most advanced and creative re- search methods in the world. Foreign governments, companies, and nonprofit institutions establish ties with research universities to gain timely access to new knowledge and technology in particular fields or to engage university-based re- searchers in the solution of particular technical problems. Foreign firms license technology generated and owned by U.S. universities, and their U.S. affiliates look to U.S. research universities for highly trained science and engineering tal- ent.2 Often, foreign firms look upon their investments in U.S. research universi- ties as a way to build goodwill and enhance their reputations in the U.S. research community.3 At the same time, several related factors have encouraged U.S. research uni- versities to solicit more foreign participation in recent years. The nation's indig-

92 FOREIGN PARTICIPATION IN U.S. RESEARCH ED DEVELOPMENT enous supply of advanced degree students, postdoctoral researchers, and faculty in science and engineering has fallen behind demand, forcing universities to turn increasingly to foreign talent to make up the shortfall. Similarly, the failure of federal R&D budgets to keep pace with the rising fiscal demands of research universities has encouraged these institutions to seek other patrons and clients, including foreigners. Finally, rapid growth in the scope, quality, and accessibility of foreign-based science and engineering capabilities has encouraged U.S.-based academic researchers to seek out foreign collaborators in many fields. Foreign participation in U.S. academic R&D has manifested itself in mul- tiple ways in recent years. These include: · The involvement of foreign students, postgraduates, and visiting scholars and researchers from foreign firms in university research activities; . Collaborations between foreign academic institutions/researchers and their U.S.-based counterparts; · Sponsored and open-ended underwriting of university research by foreign governments, foreign corporations and their U.S. subsidiaries, and foreign non- ~. . . profit Institutions; · The cooperative activities of foreign corporate laboratories operating near U.S. research universities: · Foreign support for the construction of buildings, purchase of equipment, and other in-kind contributions; and · The hiring of U.S. faculty as consultants or advisors to foreign corpora- tions and governments. Foreign Students, Researchers, and Faculty One measure of expanding foreign participation in U.S. university research and development has been the growing dependence of these institutions on for- eign-born graduate students, postdoctoral researchers, and faculty. In 1980, non- U.S. citizens accounted for 12.8 percent of graduate enrollments in U.S. science and engineering fields, 21 percent of all science and engineering doctorate recipi- ents, and a third of all postdoctoral researchers in science and engineering fields at U.S. universities. By 1991, non-U.S. citizens accounted for 23 percent of enrollments, 37 percent of doctorates, and more than half of postdocs working in science and engineering fields (Figure 4.1) (National Science Foundation, 1993a). U.S. engineering schools are particularly dependent on foreign talent (Figure 4.2~. As of 1991, 47 percent of all graduate students, 59 percent of all doctorate recipients, and 69 percent of all university-based postdoctoral researchers in en- gineering were non-U.S. citizens. The extent of foreign participation at the gradu- ate level varies among engineering fields, with significantly more participation in petroleum, mining, and agricultural engineering, and significantly less in aero- space and biomedical engineering, and in engineering science (Figure 4.3~.

PUBLICLY FUNDED R&D 100 - 90 - 80 - 70 - O 60- o a) a) 50 - 40 - 30 - 20 - 10 - _ — ~ - O— 1980 1 982 Postdocs Doctorate recipients — Graduate students ,~_ _- 1 1 1 1 1 1 1 1 1 1 1 1984 1986 1988 1990 93 FIGURE 4.1 Foreign postdocs, doctorate recipients, and graduate students in science and engineering fields, 1980 through 1991. SOURCE: National Science Foundation (1993a). 100 - 90 - 80 - 70 - 60 - , o ~ 50- ,,-- a' a) 40 - 30 - 20 - 1 n - - Jr Or Of - Postdocs Doctorate recipients Graduate students 1980 1 982 1984 1986 1988 1990 FIGURE 4.2 Foreign postdocs, doctorate recipients, and graduate students in engineer- ing, 1980 through 1991. SOURCE: National Science Foundation (1993a).

94 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT Engineering (total) Petroleum ~ a Mining Agricultural Chemical Industrial Metallurgical/Material Electrical Mechanical Nuclear Civil Engineering Science Aerospace Biomedical Engineering (NEC*) * NEC = not elsewhere classified ~~ ~ ~ ~ ~ ~S~: ,~ ~~:~ S ~ ; ~ ; ~~ ~ ~ ~ ~ S~:~ T.~?? · S~ T~ ~ S ~ ~ ~ ~ ~~. , ~ - . . ~T~ ~ ~ ~~ ~S~ ~ .~ ~~ at. . . ~~ ~ ~~ S?.:~ ?:~i~: ITS,, ~ ~ ~~ ~~ ~ ~ ~~ _ . _ _ _ ~ ~ ~ ~ ~ _ ~ ~ ~ i~ _. .~ .. ~? ~ ~ ~~: S: S: Si S.~.~S ~ .. i- /- ~\ S~ S~ ~ ______ ~ 4 ~ i i. ~S ~ i. T ~ . ~~ ~ ~ ~ ~- ~ ~ ~ T? ST ~ i~ -: ~~: -hi- i-- --; ~ ~ 1 1 1 1 1 1 1 [A 1 983 C] 1 990 · 1 991 0 10 20 30 40 50 60 70 80 FIGURE 4.3 Non-U.S. citizens enrolled full-time in graduate engineering programs, percent of total enrollees by field of study, 1983, 1990, and 1991. SOURCE: National Science Foundation (1993a). During the past 15 years, U.S. engineering faculties have experienced similar growth in the level of foreign involvement. In 1975, only 10 percent of engineer- ing faculty under 36 years of age were from other countries. In 1985, the latest year for which data are available, the foreign-born share stood at 50 percent (Na- tional Research Council, 1988~.4 The sudden rise in foreign graduate enrollment, postdoctoral research, and faculty employment is a function of three inter-related developments: the rapid growth of university research activities during the past decade and the resulting increase in demand for research personnel; a rapid increase in American industry's

PUBLICLY FUNDED R&D 95 demand for science and engineering graduates, mostly those with B.S. degrees; and an absolute decline in the number of U.S.-born science and engineering stu- dents who pursue advanced engineering degrees or academic careers (National Research Council, 1988~. In contrast to the well-documented situation with foreign students, post- doctoral researchers, and resident faculty, little is known about the number, re- search focus, or duration of stay of foreign researchers visiting U.S. universities. Because most U.S. universities are highly decentralized, they do not keep tabs on foreign visitors, unless these individuals have an appointment at the host univer- sity or require contact with the federal government. In 1991, for example, the Massachusetts Institute of Technology (MIT) reported hosting 1,250 foreign scholars, "including professors, visiting scientists and engineers, post-does, re- search affiliates and others" on campus, of whom 91 were paid fully or partially by foreign industrial firms (Massachusetts Institute of Technology, 19911.5 Collaboration between Foreign and U.S. Academic Researchers Another dimension of deepening foreign involvement in U.S. academic re- search is the increase in research collaboration between U.S. academic research- ers and their counterparts abroad in many science and engineering fields. One measure of this growth is the number of journal citations that include the names of both U.S. and foreign researchers (Figure 4.41. Between 1980 and 1991, the share of scientific and technical articles with international coauthorship more than doubled, from 5.2 percent to 11.0 percent. Papers in the earth and space sciences, mathematics, and physics were more likely than those in other fields to exhibit coauthorship. Although no national data are collected on the extent of research collabora- tion between U.S. universities or university-based research centers and their coun- terparts overseas, anecdotal evidence suggests an increase in such linkages (Godfrey, 1991~. For example, in testimony at a December 1993 National Acad- emy of Engineering workshop on the flat panel display industry, Jay William Doane, director of the Liquid Crystal Institute at Kent State University, under- lined the importance of his institution's research collaboration with the Univer- sity of Stuttgart, Tokyo University of Agriculture and Technology, and the Slov- enian J. Stefan Institute to active matrix display research and development.6 Foreign Funding of U.S. University Research Between 1975 and 1994, the share of university research funded by the fed- eral government fell from 67 to 60 percent and that supported by state and local governments decreased from 10 to 8 percent. Over the same period, industry more than doubled its contributions to academic research, from 3 to 7 percent of the total; the share of academic research supported by nonprofit institutions re-

96 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT 20 - 15 - 10 - 5- O- Mathematics Earth and space sciences '~'x: " \~? ,/~ - _ _ — .' )~- _-- _~- < Life sciences \ ~ I - ,"' ..* , _ .~-- __ Engineering and technbology 4v .. fib - . i. e`' Be, or 4L Physics .~ · ,~ _ I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Chemistry 1976 1981 1986 1991 FIGURE 4.4 Internationally coauthored articles as a percent of all articles, by science and engineering field, 1976 through 1991. SOURCE: National Science Board (1993~. mained virtually unchanged, between 7 and 8 percent, while the contributions by universities themselves increased from 12 to 18 percent (Table 4.1~. Within the context of these shifts, foreign governments, nonprofit institutions, and compa- nies have begun to play a small role in underwriting U.S. academic R&D. Data on foreign funding of U.S. university-based research are scant and dated. Although some public institutions are required by state law to report the receipt of foreign funds, most do not differentiate support obtained from domestic versus foreign sources. Factors that contribute to the haphazard tracking of foreign fi-

PUBLICLY FUNDED R&D TABLE 4-1 Support for U.S. Academic R&D, Percent by Contributing Sector, 1975, 1980, 1985, 1990, and 1994 1975 1980 1985 1990 1994 Federal government 67.1 67.5 62.6 59.2 60.1 State and local governments 9.7 8.2 7.8 8.1 7.6 Industry 3.3 3.9 5.8 6.9 6.9 Academic institutions 12.2 13.8 16.7 18.5 17.9 All other sources 7.6 6.6 7.2 7.3 7.4 NOTE: Columns do not necessarily total to 100 due to rounding. SOURCE: National Science Board (1993) and data from the National Science Foundation Science Resource Series, National Patterns of R&D Resources series, accessed via the World Wide Web. 97 nancial support include the lack of uniform university accounting procedures, the multiplicity of funding sources and channels, and the decentralized nature of ex- changes between donors and a broad spectrum of university offices, departments, and individual researchers. According to the most recently published survey, foreign-sponsored research stood at $74.3 million in fiscal 1986, or roughly 1 percent of total university R&D expenditures (U.S. General Accounting Office, 1988a).7 An informal inquiry by the National Science Foundation (NSF) found that foreign support of university R&D had not changed as of fiscal 1988 (National Science Foundation, 1990c).8 More recent data collected from leading research universities by the study com- mittee suggest that the foreign share of academic research funding has not grown appreciably since the GAO survey was conducted (Table 4.2~. Also in fiscal 1986, more than half of foreign funds for U.S. academic R&D were concentrated in five institutions: Texas A&M University, Harvard Univer- sity, MIT, Oregon State University, and the University of Wisconsin. Foreign support ranged from 1 to 9 percent of total R&D expenditures at these institutions and averaged 4 percent. In fiscal 1988, three top-20 research universities, Texas A&M, the University of Michigan, and Harvard, reported receiving more than 2 percent of their research funding from foreign sources (National Science Founda- tion, 1990c). Another nine top-20 institutions received between 1 and 2 percent of their research funding from foreign sources. Six years later, in fiscal 1994, the foreign share of total sponsored research at most of the nation's top research universities had not changed significantly.9 In fiscal year 1986, Japanese entities sponsored more university-based R&D in the United States ($9.5 million) than did those of any other foreign country, although the United Kingdom and Germany were also major contributors to U.S. academic research (U.S. General Accounting Office, 1988a). That year, foreign funding of U.S. academic R&D was concentrated in a relatively small number of universities (Table 4.3~.

98 ; - C~ A - en U) ._ CQ ._ . CC ; - Cal Cal . V) on o Ed Ct 50 do 50 o .~ o en Cal 4 - ·_4 X ~ Ct oo of ~ To -¢ oo o ~ Ed o _` Cal o o ^ — o En o _ X Cal o to o do ^ Ct o En do ~ ~ Em, X Cal ~ 8 Ce ~ of o En ~ V V ~ V V ~ ~ V V V V ~ ~ ~ V Z o ~ ~ o o ~o o ~ ~ ~ o o ~ o =^ ~ o~ ~ oo ~ c~ ~ ~ ~ ~ ~ oo ~ c~ ~ ~ c~ ~ ~ ~ o ~ ~ o ~ oo ~ oo ~ ~ ~ oo o c~ ~ ~ ~ ~ ~ ~ oo oo ~ ~ o ~ ~ oo ~ o ~ ~ ~ o ~ ~ N ~ ~ cM v~ ~ oo c~ oo o ~ ~ ~ ~ ~ oo o ~ o ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ - A V V ~ V ~ A ~ V ~ ~ — V ~ ~ V A V V ~ oo ~ ~ ~ ~ U~ ~ ~ ~ ~ ~ oo ~ oo ~ ~ ~ U~ CM ~ 00 0 u~ ~D ~ ~ oo ~ ~ ~ ~ 0~ 0 ~ M0 o ~ ~ ~ ~ ~ ~ o o C~ ~ o ~ ~ ~ o ~ ~ oo o ~ ~ ~ ~ o C~ ~ ~ ~ oo o oo o ~ ~ ~ U~ CN CM ~ ~ ~ CA ~ U~ ~ ~ C~ ~ ~ ~ ~ ~ — O O O kD CN ~ O ') ~ <:) O ~— ~ cr ~ oo 0 c~ O 0 0 ~ — x 0 ~ ~ cr 0 oo O ~ ~ oo ~ ~ O~ 0 0 ~ ~ ~ oo ~ ~ 5\ [_ ~t ~ ~ ~ t_ [_ ~ ~ ~ ~ oo ~ ~ ~ ~ ~ ~ ~ ~ 0 ~ ~ ~ ~ 0 oo ~ C~ U~ ~ ~ ~ ~ ~ ~ ~ C~ ~ ~ ~ ~ oo ~ 0 —~ ~ ~ ~ ~ ~ ~ ~ ~ ,4 ~ ~ ~ ~ ~ -_ ~ ~ ~ ;^ o ~SO o E~ cq - , ._ _ ~o I ~ ~ O-CO C~ C) C~ ~ C~ .~.~> ~ 0 0 0 C~ . . . Ct .> .> .> V, Ct C~ . . . o C) ._ ~ . _ ~ . ~ _ ~ · V~ ~ ~ ~ ~ C ~ C >. c~e E >< ~ =- ~ ~ . . . . . ~ ~ ~ or C oc . ~ ~ ~ ~ U ~ ~ ~i :t ~ O ~ · · ·= =) ~ ~ Ct C~ ~ Ct · ~ ~ · ~ ' - ~ Ct ~ ~ ~ ~ E~ ~ ~ ~ ~ 40 ~ ~ ~ ~ ~ ~ ~ O O O O 0 5_, 0 0 0 . Ct ;> ~ ~ ;> ._ ._ ._ ._ ._ l_ ._ ._ ._ ~ ~ ~ ~ ~ Ct ~ ~ ~ .~= ~ 00 ~ O . . . . . . . ~ O — CN ~ ~ ~ ~C _ _ _ _. _ _ ~ 50 Ct ;^ . . C-0, C~ 3 ._ r~ X . P~ Ct o _ s~ C~ o C~ o ;^ Ct C~ . ._ ~ C:, ~ C ._ .= C) 3 o~ ~ C~ o o ~ C~ o~ o ~ ~ ,= oc ~ — ._ ~ 5.., — Ct ~ ~ U: c~ a~ ._ ~ .~ o ~ C .> ~ :~o Ct C~ - - a~ = ~i o Ct C) ce o ct z 'e o o ~ - c~ o st c) - o ~— z . ~ c~ oo oo o oc . - o c) c~ - c) a~ v . ~ p~ o v:

PUBLICLY FUNDED R&D 99 Three fields geology ($16.4 million), agriculture ($11.5 million), and medicine ($8.4 million) accounted for nearly half of all foreign support for U.S. academic R&D in fiscal 1986 (Table 4.4). There were very few similarities in either the research fields funded or the source of foreign support among the top five recipient institutions. An international ocean-drilling research program funded in part by NSF at Texas A&M University received 93 percent of all for- eign support directed to U.S. university-based geology research. Four universi- ties accounted for nearly 80 percent of all foreign-funded research in agriculture, most of which focused on the agricultural research needs of developing countries (U.S. General Accounting Office, 1988a). In fiscal 1986, a little more than a third of all foreign funds for university research came from businesses; the remaining two-thirds came from foreign gov- ernments and nonprofit organizations. Of the top five U.S. recipients of foreign R&D funding, Texas A&M, Harvard, and Oregon State received the majority of such support from governments and nonprofit institutions (99 percent, 95 per- cent, and 78 percent, respectively). In contrast, nearly all foreign support of research at MIT and the University of Wisconsin came from private companies (98 and 89 percent, respectively). Indeed, foreign firms accounted for more than half of all foreign R&D funding at 13 of the top 20 U.S. research universities in fiscal 1986 (Table 4.5~. Foreign Corporate Participation in U.S. Academic R&D The most controversial aspect of foreign involvement in U.S. university R&D centers on the participation of foreign corporations and their U.S. subsidiaries. This is because foreign firms unlike foreign governments, nonprofit institutions, and academic researchers are well equipped to acquire and apply commercially valuable outputs of publicly funded U.S. academic research, thereby appropriat- ing a significant share of their associated economic returns.~° There are five major ways foreign corporations can participate in U.S. uni- versity-based R&D: through company-sponsored research, either in the form of contracts or grants; by working with university patent and technology licensing offices; via university industrial liaison or affiliates' programs; by involvement in university-industry research centers; and through activities in formal or de facto industrial technology parks located near prominent U.S. research universities. Data on the current magnitude, disciplinary focus, and national shares of foreign sponsored research at individual U.S. universities and federal laboratories are fragmentary. Limited survey data (National Research Council, 1994a; Rob- erts, 1995a) and anecdotal evidence indicate that Japanese companies are more diligent than their U.S. or European counterparts with regard to accessing, moni- toring, and drawing upon the research capabilities of these publicly funded insti- tutions.

100 Cal Cal ~ .= Cal : - ;> ._ it: P" · Cal o o Ct o Cal ·_4 Ct ¢ U) o 4 - o .O o ¢ ~ . ° o D V ·,_ of o ·~ c) O O U. ._ ._ Cal ._ ._ C) C) sit o o EN o Cq ~ O O · - ~ to to O · ~ ~ O U. U. ~4 O ·— . - ~ sit C~ . _ ID ~ ._ ~ .> A t - 0, . O ;> ~ $:: V, ~ o D ~ I: V U 1 ~ ~ ~ ~ ~ c~ ~ ~ ;^ ~4 ~o . ·c ) ~ i ~ =-c~ j -= ~ u, c') ·s - ~ ~ · - ~ ~ ~ ~ 3 ~ ~ C~ ~ ,= o C~ X ~ ~ ~ ~ C~ U~ ;^ ~4 o - ~o o — o ~ ~ Ct C~ ~ o C~ ~ o o i i ca ._ ._ . j ~ . C~ o Ct ~ ;> ~ . oc o o C) E~ C~ o ~ Ct o . ~ , ~ ~ C~ ~ o~ ~S U . <—o, o ~ ~ o Ct ~ j V o oo ~o o a _ ~ ~n oo ~ .~ ~ ~ ~ ~ Ct C~ Ct 5 - o — c~ ~ s~ 'e ·= x f~4 ~

101 5) G) : - o Cal o Cq Cal ._ ~ ~ ~ o ~ ~ ~ ~ to ~ ~ ~ .O con ~ ~ ~ ~ ~ 3 ;^ on o C) En o C) Ct ~ Ct ~ ~ Cal ¢ 3 O ~ O C) . ~ X . ~ US ~ ED ~ ~ o - o Ed o ._ ¢ ' O ~ Cal Cq Ct ;^ o - o C) C) E~ C~ o . _ - Ct C~ .= .> .~ ~ ~ o~ ~ ¢ X ~ 00 U: U: — X ·— o _} E~ ~ O O O ~ ~ 00 O ~D ~ u~ . . . . _4 _~ U. ~ C ~ ~ '_ C.) C [I1 0 _ ~ _ ~ ~ O ~ ~4 o 50 ~4 ~: - . ~ Ct ~ C~ ._ O C) ~ .~ 00 ~ C~ O .O ~ o c) u, c4 - o 5= 'e o oo . c) ~ ~ . - - - o :e ·— - - . ~ ~ . ~ ~ 'e — ~ oc ) ~ o ~D ~ · ~ o · o ~ ~ c~ > . ~ mo << - . ~ c c c ~ ~ ~ cd c~ - o . ~ c~ ~ ~ ~ z X ~ a' ~ E~ ~ ~ ~a

102 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT TABLE 4.4 Foreign Support for U.S. Academic R&D, by University Department, Fiscal 1986 Field $ in Millions Geology Agriculture Medicine Civil engineering Biology Chemistry Chemical engineering Materials engineering Computer engineering Other engineering Mechanical engineering Electrical engineering Physics Nuclear engineering Aeronautical engineering Psychology Other 16.4 11.5 8.4 2.7 2.6 1.5 1.5 1.3 1.1 1.1 .8 .7 .5 .4 .3 .1 21.3 SOURCE: U.S. General Accounting Office (1988a). Research Contracts and Grants As noted, foreign-owned companies accounted for approximately one-third of all foreign-sponsored research at U.S. universities, or one-third of one percent of total academic research funding in 1986, the latest year for which aggregate data are available. Company-sponsored research at U.S. universities comprises research contracts and research grants. The distinction between the two instru- ments is subtle and varies among institutions. In general, research contracts ob- ligate university-based researchers to provide their corporate sponsor with more- frequent and more-formal reports on their progress than are required with grants. Contracts also usually specify particular deliverables, whereas grants are gener- ally more open ended. National statistics on the sponsorship of academic re- search do not distinguish between contracts and grants because of the definitional vagaries and reporting inconsistencies among institutions. Research grants may demand more of a quid pro quo from university-based researchers than the term "grant" implies. For example, companies providing research grants to university-based researchers may receive favorable consider- ation in licensing negotiations, even though they do not receive royaltyfree or exclusive rights. For example, at the University of California at Berkeley and MIT, some engineering departments have agreed to accept visiting fellows from major industrial donors. During the 1991-92 academic year, Berkeley's College

PUBLICLY FUNDED R&D TABLE 4.5 Total Foreign Support for R&D and Corporate Share of that Support at the Top 20 U.S. Research Universities, Fiscal 1986a Total Foreign R&D Support ($l,000S) Corporate Share (% of Total) 1. Massachusetts Institute of Technology 5,304 98.2 2. Univ. of Wisconsin-Madison 2,380 89.5 3. Stanford Univ. 562 93.2 4. Cornell Univ. 245 29.4 5. Harvard Univ. 10,781 4.8 6. Univ. of Michigan 450 83.9 7. Texas A&M Univ. 15,200 1.3 8. Johns Hopkins Univ. 2,118 n/a 9. Univ. of California-Los Angeles 782 50.4 10. Univ. of Washington 2,068 38.3 11. Pennsylvania State Univ. 673 69.4 12. Univ. of Pennsylvania 228 100.0 13. Univ. of California-San Diego 728 96.5 14. Univ. of Minnesota 268 61.9 15. Univ. of Arizona 586 100.0 16. Yale Univ. 194 87.6 17. Univ. of Illinois-Urbana 206 0 18. Univ. of Texas-Austin 752 96.9 19. Univ. of Southern California 30 n/a 20. Univ. of California-Berkeley 187 52.5 aUniversities ranked according to volume of R&D expenditures in 1986. SOURCE: U.S. General Accounting Office (1988a). 103 of Engineering hosted 17 fellows from Japanese companies as part of a research program funded with between $2 and $3 million in grants. The salaries and expenses of these fellows were paid by their sponsor companies. There are no recent national aggregate data on the disciplinary focus of for- eign corporate sponsorship of university research. However, in fiscal 1986, nearly 50 percent of foreign support of university research, including funding by foreign governments, companies, and nonprofit organizations, was concentrated in geol- ogy, agriculture, and medicine (see Table 4.3~. Stalson's (1989) survey of eight leading engineering schools found that foreign corporate sponsors demonstrated an interest in research from all of the traditional engineering disciplines, as well as the science-oriented fields of biotechnology and computer science. Although Japanese firms awarded the majority of foreign-sponsored contracts in these dis- ciplines, German and French companies were significantly represented, with a smaller presence of Canadian, British, Dutch, Italian, and other European firms. According to Stalson, there was no discernible pattern of national specialization

104 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT in any field of engineering research except civil engineering and construction, which appeared to be the exclusive province of the Japanese. As of 1993, over 60 percent of all foreign company-sponsored research at the University of California at Berkeley was concentrated in three locations: the department of chemistry, the electronics research laboratory, and the space sci- ences laboratory. The largest single research contract by a foreign firm, $1.8 million to the space sciences laboratory, came from French-owned Fairchild Space Company, itself a contractor to NASA. British companies accounted for almost all the foreign corporate research support in chemistry. The German com- pany, Siemens, provided 85 percent of all foreign funds that went to the electron- ics research laboratory; the remaining 15 percent came from Japanese firms. All foreign corporate support of the university's Engineering Systems Research Cen- ter and department of optometry came from Japanese companies, as did a major- ity of foreign research support for the department of mechanical engineering and mechanical design.~3 For the most part, foreign firms appear to be sensitive to the charge that they are taking more from U.S. research institutions than they are contributing. They have sought ways to avoid even the appearance of impropriety by concentrating on "precompetitive" research and seeking U.S. corporate partners when engaging universities in contract research (Stalson, 1989; U.S. Congress, House, 1993~. Nevertheless, in recent years, a small number of controversial agreements be- tween U.S. universities and foreign firms have had a negative effect on the public's perception of foreign involvement in U.S. government-funded R&D. A case in point was the 1988 agreement between Hitachi Chemical Research (HCR), a subsidiary of the Japanese firm Hitachi Chemical Company, Ltd., and the University of California, Irvine (UCI). Under the terms of this agreement, the Japanese company agreed to build a facility on university-owned property to house rentfree the basic research laboratories of the university's department of biological chemistry and a basic research laboratory for the company's propri- etary programs. Normally, intellectual property developed using University of California resources, including facilities and equipment, is owned by the univer- sity. In this case, however, Hitachi owns all intellectual property developed within its proprietary research laboratory, and in the case of "joint investigations of at least one UCI investor and at least one HCR inventor," UCI and HCR will each own an equal interest in the invention (National Research Council, 1992c). From the outset, public criticism of the UCI-HCR agreement was intense. Many questioned whether the two institutions could keep their research activities separate or whether the U.S. taxpayer was being called upon to subsidize the proprietary research of a Japanese company. Others doubted whether UCI fac- ulty collaborating with HCR could avoid potentially harmful conflicts of interest. Although little has transpired to substantiate these fears since the shared research facility opened its doors in 1990, public skepticism about the arrangement lingers.

PUBLICLY FUNDED R&D 105 It is not unusual for U.S. or foreign-owned firms to establish R&D facilities proximate to leading research universities in order to draw upon their resident technical expertise and research activities. (See discussion of industrial technol- ogy parks, below.) It is also fairly common for companies to supply instrumen- tation, equipment, and other in-kind support to academic research projects in exchange for facilitated or preferential access to intellectual property resulting from the research. In the committee's view, what has made the UCI-HRC agree- ment so controversial is that it is such an obvious attempt by a privately owned Japanese firm to draw upon the intellectual resources of a university that are underwritten by the U.S. government. Also controversial have been cases in which universities or university-affili- ated research institutions have effectively sold to foreign-owned firms the right of first refusal to all intellectual property resulting from a broad stream of research that has been partially, and in some cases largely, subsidized by U.S. public mon- ies. Some of the more highly publicized agreements of this type include the 1989 deal between Harvard Medical School, its teaching hospital Massachusetts Gen- eral Hospital, and the Japanese cosmetics firm Shiseido Co. Ltd., in the area of skin research;~4 the 1990 agreement between the University of California at San Diego and the Swiss pharmaceutical company, Ciba-Geigy;is the 1992 agree- ment between Scripps Research Institute and the Swiss pharmaceutical company Sandoz Pharma;~6 and the 1993 agreement between the University of California at San Francisco and the Japanese firm Daiichi Pharmaceuticals, involving re- search on atherosclerosis.~7 The Scripps-Sandoz Pharma agreement has been the most controversial of the four. This is so in part because it was the largest research agreement ever concluded between a U.S. research institution and an industrial partner, and also because its terms were so expansive. Under the original agreement, which was to begin in 1997 and continue for 10 years, Scripps would receive $300 million and in exchange would grant Sandoz the right of first refusal to license any intellec- tual property resulting from institute's research. The agreement included the option to extend the agreement for an additional 6 years. Controversy arose be- cause Scripps receives 60 percent of its funding from the National Institutes of Health (NIH). The agreement was subsequently modified at NIH's insistence (National Institutes of Health, 1994a,b). In the opinion of the committee, it is in the best interests of the United States for publicly supported research universities and other publicly funded institutions to offer to private-sector investors the right of first refusal for intellectual prop- erty resulting from specific research supported by these investors. Such provi- sions offer an important incentive to private companies to invest in high-risk research activities that leverage public R&D monies. They also can yield sub- stantial benefits to society at large, in the form of better trained science and engi- neering graduates, advances in knowledge, and improvements in instrumenta-

106 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT lion. If the intellectual property includes technology that may be commercial- ized, another effect may be higher performance and lower-cost products and ser- vices. At the same time, the public interest is poorly served when universities grant to a private firm rights of first refusal to all intellectual property that might follow from an extremely broad stream of publicly subsidized research in effect pro- viding the firm with a windfall and potentially preventing or delaying useful ap- plication of the resulting intellectual property by other companies. Each of the cases that received extensive public scrutiny has involved different circumstances and different trade-offs, whose implications for the public interest are difficult to assess. Nevertheless, universities and other performers of publicly funded R&D need to recognize and address the public perception that they are giving away the store. Patent and Technology Licensing Programs In 1980, Congress passed amendments to the Patent and Trademark Act, known as the Bayh-Dole Act (P.L. 96-480~. This legislation made it possible for universities, other nonprofit organizations, and small businesses to retain rights to most of their federally funded inventions. Universities are granted consider- able autonomy in commercializing these technologies under the act. In return for this autonomy, the amendments require that universities meet two provisions. First, they must give preference to U.S.-based businesses, particularly small com- panies, in licensing federally funded technologies. Second, they may grant ex- clusive rights or sell these technologies only to companies willing and able to manufacture substantially in the United States products embodying the invention or produced through application of the invention (U.S. General Accounting Of- fice, 19924.~9 In response to this change in law, U.S. research universities have greatly expanded their patenting and technology licensing programs during the past 14 years. What few data exist suggest that foreign companies played an appreciable role in this increased activity. A 1991 GAO survey of patent licensing by 35 leading U.S. research universities found that roughly 15 percent (29 of 197) of all exclusive licenses issued by universities for technologies developed with NSF or NIH funds were sold to foreign companies (18) or to their U.S. subsidiaries (1 1) (U.S. General Accounting Office, 1992~.2° As of May 1994, foreign companies held 21 (or 8 percent) of MIT's 225 active exclusive licenses. European firms held 13 of these and Japanese and Canadian firms claimed 4 each. Foreign com- panies also claimed 23 (roughly 16 percent) of MIT's 145 currently active nonex- clusive licenses. By way of comparison, foreign-owned companies held just 3 of 27 active exclusive licenses from the University of California at San Diego, 5 of 65 exclusive licenses from the University of Michigan, and only 1 of 34 active exclusive licenses issued by the Georgia Institute of Technology.2i Neither the

PUBLICLY FUNDED R&D 107 GAO data nor those provided by individual research universities provide infor- mation on the nature or relative importance of exclusive licenses granted to for- eign and domestic firms. When universities turn to foreign firms to license their technologies, they seem to do so for reasons specific to each transaction. In his testimony at a November 9, 1993, NAE workshop on access by foreign companies to U.S. tech- nology, John Wiley, then dean of graduate studies and vice president for research at the University of Wisconsin at Madison, illustrated this point with two ex- amples. In the first case, the university licensed a technology for making a sili- con-backbone polymer to a Japanese firm after having spent several years at- tempting unsuccessfully to interest U.S. chemical, fiber, and other companies in the process. In the second case, the university licensed to a Japanese pharmaceu- tical company what turned out to be a highly lucrative patent for a vitamin D derivative found to be effective in treating osteoporosis in women eating a very low-fat diet. Given the generally high-fat diet of most Americans, and hence, the relatively small projected U.S. market for the derivative, no U.S. pharmaceutical companies were interested in the technology. The Japanese diet, in contrast, made for a large potential domestic market for the derivative, which made the patent attractive to the Japanese firm (Wiley, 19931. Wiley's comments, as well as those made by other university research ad- ministrators, suggest that university technology licensing offices generally turn to foreign licensees as a last resort when no prospective U.S. licensee can be found.22 Industrial Liaison Programs Industrial liaison programs (ILPs) charge membership fees to companies in return for providing them with general access to the results of university research, to researchers, and to laboratories in specified fields. As part of its 1992 survey of 35 leading U.S. research universities, GAO gathered information on the growth of industrial liaison programs. Thirty of these institutions had at least one indus- trial liaison program. Carnegie Mellon University alone accounted for 59 of 278 such programs that were identified. Of the 30 universities with ILPs, 24 had a total of 499 foreign companies enrolled in at least 1 industrial liaison program.23 Nine of the 24 host institutions reported holding more than $10,000 in stock through endowments in at least one of the participating foreign companies. Eighteen of the universities surveyed provide liaison program members, whether domestic or foreign, with access to the results of federally funded research before those results are made generally available, while the other 12 institutions do not. Three institutions MIT, Stanford, and the University of California at Berkeley, which accounted for 290, or 58 percent, of all foreign liaison program participants—do not provide ad- vance access to research results (U.S. General Accounting Office, 1992~.24

108 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT Only a few of the universities surveyed reported that they distinguish be- tween U.S. and foreign companies when reviewing and approving requests to join their liaison programs, allowing members to participate in program activi- ties, or assessing membership fees. Both the University of California at Berkeley and MIT require significantly greater participation fees from foreign companies than they do from U.S.-owned firms.25 The universities of Michigan, Washing- ton, and Wisconsin, as well as Columbia University, place certain restrictions on foreign participation in at least one of their industrial liaison programs (U.S. Gen- eral Accounting Office, 19921. University-Industry Research Centers26 University-industry research centers (UIRCs) represent a third channel through which foreign companies participate in U.S. university-based science and engineering research. During the 1970s and 1980s, increased support from state and federal governments fueled a rapid proliferation in the number of UIRCs. These seed or matching funds encouraged research universities to institutionalize more formal R&D relationships with private companies. The most aggressive federal sponsor of UIRCs during the 1980s was the National Science Foundation (NSF), which helped establish a raft of university-based centers including Engi- neering Research Centers, Science and Technology Centers, State/University/In- dustry Cooperative Research Centers, Industry/University/Cooperative Research Centers, Materials Research Centers, and Supercomputer Centers.27 Data gathered by Cohen et al. (1994) from over 1,000 UIRCs based at more than 200 U.S. university campuses indicate that on average, foreign firms ac- counted for roughly 12 percent of all companies involved in nearly 470 centers reporting at least some foreign participation (Table 4.6~. Among the 17 technol- ogy areas identified, the level of foreign involvement exceeded 12 percent in five areas: biotechnology, pharmaceuticals, biomedicine, chemicals, and agriculture and food. For the most part, the policy statements and actions of state governments appear to support foreign corporate involvement in local research universities.28 In many if not most instances, however, the federal government restricts or places conditions on foreign participation in the centers it supports. For instance, the NSF does not allow foreign governments to participate in the Engineering Re- search Centers it funds. However, foreign-owned firms "are not excluded, as long as the Center can demonstrate that a quid pro quo relationship exists, where information exchange is substantially equal in both directions." In fiscal 1994, NSF classified 9.5 percent of total industrial participation in these centers as for- eign. Foreign researchers may work in the centers and center researchers may spend time in labs abroad. Some centers have opted for more restrictive policies on membership for foreign firms, however.29 Stalson (1989) found that the strongest industry opposition to involvement in

PUBLICLY FUNDED R&D TABLE 4.6 Foreign Company Involvement in U.S. University- Industry Research Centers as Percent of Total Industry Participation, by Technology Area, 1992 Technology Area Percent Foreign Involvement Biotechnology Pharmaceuticals Biomedicine Chemicals Agriculture and food Scientific instruments Advanced materials Aerospace . energy Transportation Environmental technology and waste management Industrial automation and robotics Computer software Semiconductor electronics Telecommunications Computer hardware Other technology areas 7.4 6.9 6.8 5.7 3.7 .6 0.2 9.s 9.5 9.4 9.3 8.8 8.s 7.7 6.2 6.1 9.2 SOURCE: Wesley Cohen, Carnegie Mellon University, unpublished data. 109 the university-industry centers came from firms in sensitive, competitive areas or in fields where the United States was believed to have a strong technological lead. For example, some U.S. companies involved in semiconductor and micro- electronics research, magnetics technology, and materials science voiced concern about foreign companies also participating in these centers. Industrial Technology Parks Industrial technology parks established near prominent U.S. research univer- sities have provided another avenue for foreign access to U.S. R&D. Some states have established formal parks, such as North Carolina's Research Triangle Park (a major center for biotechnology research), to promote state or regional eco- nomic development. Other de facto technology parks have been developed near major American research universities by firms seeking to exploit the "dense ex- ternalities" or "spillovers" that come from daily interaction with university re- searchers or researchers working for other companies. Dalton and Serapio (1995) have documented how foreign corporate R&D facilities are clustered around major U.S. research universities, such as MIT, Princeton, Stanford, and the Uni- versity of California at Berkeley (Chapter 3, Figure 3.61.

110 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT Data on the scope and nature of foreign company research activities in the Princeton and Research Triangle Park cases indicate that the R&D investments appear to be both quantitatively and qualitatively significant, at least for some of the major foreign firms. For example, although the U.S. R&D laboratories of most Japanese electronics companies are focused on highly applied research and have relatively small research staffs and budgets, the NEC Research Institute in Princeton is a notable exception. As of fiscal 1994, it employed 135 people, including 78 permanent and 26 visiting technical personnel. The institute had an operating budget of approximately $20 million, and its staff was engaged in basic computer and physical sciences research. Siemens Research Corporation, also in Princeton, had a staff of 145 with an annual budget also in the range of $20 million.30 Research efforts at Siemens have been focused on software engineer- ing, imagery, learning systems, and optical data processing. At Research Tri- angle Park, Glaxo, a British chemical and pharmaceutical company, Rhone- Poulenc and Reichold Chemicals, French and Japanese chemical companies, and Ciba-Geigy, a Swiss pharmaceutical firm, collectively employ nearly 1,700 re- searchers (Dalton and Serapio, 1995; Serapio and Dalton, 1994J. Other Forms of Foreign Corporate Participation in U.S. Academic R&D Beyond the more institutionalized interaction of foreign companies and uni- versity-based researchers, a variety of relationships has developed in which uni- versity faculty act as independent professionals—often consultants, advisors, or board members. In general, American universities do not collect data on such interactions. However, a 1990 MIT survey of its faculty's consulting contacts found that foreign companies accounted for less than one-quarter of all such en- counters (Massachusetts Institute of Technology, 1991~. In several celebrated instances, foreign-owned companies have hired away U.S. university-based researchers. For example, when NEC set up its research laboratory in Princeton, it hired prominent researchers from the University of California at Berkeley, MIT, AT&T Bell Laboratories, and other top U.S. re- search institutions to lead its basic research effort in various areas of the com- puter and physical sciences (Business Week, 19924. Foreign companies have also made unrestricted gifts, such as endowed sci- ence and engineering chairs, and have given funds for physical infrastructure to U.S. research universities. For instance, as of 1991, 30 of the 215 endowed chairs at MIT were funded by foreign-owned corporations. Here again, however, no national data are available. The Special Case of Japanese Involvement in U.S. Research Universities While limited, existing data suggest that many Japanese-owned companies pursue more aggressively closer interactions with the U.S. academic research

PUBLICLY FUNDED R&D 111 community than do their European or U.S. counterparts. The special interest of Japanese firms in the U.S. academic research enterprise is confirmed in a recent survey of over 240 leading European, Japanese, and U.S. it&D-intensive compa- nies that measured the intensity of these firms' use of universities (mostly U.S. research universities) for four defined purposes: collaborative research, gleaning innovative ideas, determining technology trends, and training company personnel (Figure 4.51. Roberts (199Sa3 observes that Japanese companies draw upon the assets of research universities for all four purposes more intensively than do ei- ther European- or U.S.-owned companies.3i One major exception may be U.S. academic research in biotechnology, which, according to Dibner et al. (1992), draws significantly more intensive interaction from the large European pharrna- ceuticals and chemicals companies than it does from the smaller Japanese drug a. WAS. 80 - 60- .= I an a, 40— o Q an lo: o 20 - Collaborative research Obtaining innovative ideas Determining technology trends Training company personnel O- l North American firms European firms Japanese firms FIGURE 4.5 Use of university resources by North American, European, and Japanese companies, by type of activity, 1992. NOTE: Of 244 companies sampled, useable re- sponses were received from 95 firms (39 percent), of which 46 were from the United States (42 percent response), 27 from Europe (34 percent), and 22 from Japan (40 percent). Most of the university resources cited by survey respondents were located in the United States. SOURCE: Roberts ( 1 995a)

2 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT FOREIGN PARTICIPATION IN U.S. FEDERAL LABORATORY R&D Like the nation's academic research enterprise, the nation's federal labora- tory system has experienced growing foreign participation during the past de- cade. The more than 700 federal laboratories had a combined budget for fiscal 1993 of $23.4 billion. This total includes intramural agency laboratories as well as federally funded research and development centers (FFRDCs). FFRDCs and many intramural laboratories are government-owned, contractor-operated facili- ties managed by universities (e.g., Los Alamos, Lincoln Laboratory), university consortia (e.g., Brookhaven, Fermilab), industrial contractors operating on a not- for-profit basis (e.g., Oak Ridge National Laboratory and Sandia National Labo- ratory), or independent nonprofit organizations (e.g., MITRE Corporation, Draper Laboratory, RAND). Other intramural agency laboratories are government- owned and government-operated, such as those at the National Institutes of Health, National Institute of Standards and Technology, Naval Research Labora- tory, Naval Surface Weapons Center, and certain facilities at the National Aero- nautics and Space Administration. Federal laboratories are diverse in size, character, and mission. Most are single-office facilities employing a small number of researchers; a small percent- age are large organizations that employ thousands of scientists and engineers. Collectively, these laboratories employ roughly 120,000 scientists and engineers nationwide. In 1993, these institutions performed approximately 18 percent of all U.S. basic research, 16 percent of all applied research, and 13 percent of all tech- nology development work (National Science Board, 1993~. A History of Foreign Involvement With the exception of laboratories performing classified research, U.S. fed- eral laboratories have traditionally been open to foreign researchers. During the past 4 decades, the U.S. government has entered into agreements with foreign governments that have allowed thousands of non-U.S. researchers to work at fed- eral laboratories. During the 1960s and 1970s, the responsibilities of the federal laboratory system grew to include the construction and operation of major user facilities, such as particle and photon accelerators, environmental research parks, and materials laboratories. These new facilities opened the laboratory system increasingly to U.S. and foreign researchers from industry and academe. Since 1980, growing concern about the nation's international competitive- ness has led the federal government to take a number of steps designed to in- crease the extent to which federal laboratories support the activities of U.S. in- dustry. These have included the creation of offices of technology transfer in the federal laboratories, changes in patenting and royalty mechanisms that allow com- panies to acquire exclusive licenses and permit federal laboratory researchers to receive a percentage of royalties from commercialized technology, and reimburse-

PUBLICLY FUNDED R&D 113 ment schemes that allow private companies to conduct proprietary research at federal user facilities, such as the synchrotron light source at Brookhaven Na- tional Laboratory. Growing numbers of domestic and foreign visiting research- ers have been drawn to the federal laboratories, as these facilities have moved into the business of technology transfer and technical support to U.S. industry. By the mid 1980s, visiting researchers accounted for a significant share of the work conducted at federal laboratories. In its study of foreign participation in 50 federal research laboratories, GAO (1988b) reported that in addition to perma- nent laboratory employees, 4,657 U.S. and 3,597 foreign visiting researchers con- ducted R&D at these laboratories in fiscal 1986. Fifty percent of the visiting U.S. and 57 percent of the visiting foreign researchers were affiliated with universities and other nonprofit organizations. There were more Japanese, 758 (13.4 percent of the total pool of visiting foreign scientists), conducting research at the 50 fed- eral laboratories than any other nationality. There were 448 researchers from the United Kingdom and 438 from the People's Republic of China. From 1988 to 1991, the Department of Energy's (DOE) 10 multiprogram laboratories, operating with an annual budget of $6.2 billion and a total staff of roughly 52,300, hosted 11,000 visiting foreign scientists and engineers.32 Ap- proximately one-third of the visitors came from Japan. Two DOE laboratories, Lawrence Livermore National Laboratory and Los Alamos National Laboratory (LANL), hosted half of all foreign visiting researchers during the 3-year period (National Research Council, 1994a). From 1988 to 1990, LANL hosted 4,000 foreign visitors, 700 of them Japa- nese. The primary focus of foreign visiting researchers' activity at LANL has been basic research. Japanese visitors to LANL have been characterized by the laboratory's administrators as "more strategic" that is, more focused in their approach and better prepared to identify and pursue research and technologies of potential commercial importance than other visiting foreign scientists. In addi- tion to their interest in basic research, the Japanese scientists have also followed closely U.S. advances in the modeling of computer software codes and in laser sciences, particularly photolithography (National Research Council, 1994a). In 1993, NIH, with a total resident technical staff of approximately 16,000, hosted approximately 1,700 foreign visiting researchers as part of its visiting fel- lows, visiting associates, and visiting scientists programs. As participants in these programs, foreign researchers generally spend 2 years or longer at NIH engaged almost exclusively in basic research. Stipends and salaries are provided by the agency. In order to participate, visiting researchers must agree to certain intellec- tual property rights provisions. Together, Japanese and Chinese scientists ac- counted for over one-third of participants in the program (Table 4.7~. Although NIH lacks complete data on the institutional affiliation of foreign visitors, the agency estimates that only 4 percent of visiting researchers from Japan were from private industry.33 In its 1988 survey, GAO found that research administrators and managers at

114 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT TABLE 4.7 Participants in NIH Visiting Scientist Programs, Top 10 Countries, Fiscal 1993 Country Numbera Japan China Italy Korea India Russia United Kingdom France Germany Israel 325 287 131 105 94 92 86 77 77 75 aU.S. permanent residents not included in count. SOURCE: National Institutes of Health, unpublished data. federal laboratories supported the open exchange of information with foreign re- searchers in basic scientific fields. However, in fields with commercial potential, research managers and administrators at NIST (then the National Bureau of Stan- dards), Langley Research Center, Oak Ridge National Laboratory, and Sandia stated that they gave preferential access to U.S. researchers and organizations and carefully reviewed requests for access by foreign researchers and organizations (U.S. General Accounting Office, 1988b). Foreign Corporate Participation in Federal Laboratories In recent years, several federal laboratories have also entered into formal relationships with foreign-owned companies as part of broader efforts to collabo- rate more with U.S.-based industry in areas of precompetitive, commercially rel- evant R&D. Since the late 1980s, for example, several federal labs have signed cooperative research and development agreements (CRADAs) with foreign com- panies.34 Between March 1988 and January 1995, NIST negotiated over 500 CRADAs, 34 of them with foreign companies. Over the same period, NIH en- tered into 237 CRADAs, 26 of which were with foreign companies, 6 of these Japanese. Argonne National Laboratory and Oak Ridge National Laboratory have negotiated 92 and 108 CRADAs, respectively, and each has 2 CRADAs with foreign companies.35 Federal laboratories that wish to grant exclusive licenses for their technol- ogy to or engage in CRADAs with private organizations "must give preference to business units located in the United States that agree that products embodying inventions made under the [CRADA] or produced through the use of such inven-

PUBLICLY FUNDED R&D 115 tion will be manufactured substantially in the United States." In addition to this economic performance requirement, federal laboratory directors must consider whether the home governments of would-be foreign participants "permit and en- courage United States agencies, organizations, or other persons to enter into co- operative research and development agreements and licensing arrangements on a comparable basis; . . . have policies to protect the United States intellectual prop- erty rights; and, for classified or sensitive research, whether the foreign govern- ment has adopted adequate measures to prevent the transfer of strategic technolo- gies to destinations prohibited under U.S. national security export controls."36 Implementation of the "substantial U.S. manufacturing" requirement has been approached differently by different agencies. NIST, for example, allows its eight laboratory directors considerable discretion in the negotiation of CRADAs, including authority to decide whether the participation of a U.S. multinational company or foreign-controlled firm complies with the letter and spirit of the law. If a laboratory director considers a CRADA proposal to be particularly sensitive, he or she can request direction from the NIST Director.37 In contrast, DOE, whose 10 multiprogram laboratories had entered into over 1,200 CRADAs as of early 1995, allows less discretion to its contractor-operated labs in the negotiation of CRADAs in general and in the implementation of per- formance requirements in particular. Exercising a higher degree of centralized control, the DOE initially developed strict guidelines for compliance with the "substantial U.S. manufacturing requirement" (referred to as the U.S. Competi- tiveness Article within the modular CRADA), extending its scope to include all intellectual property generated under a CRADA, including subject inventions, patents, copyrights, trademarks, protected CRADA data, and mask works. Fur- thermore, these early DOE guidelines stipulated that such intellectual property should be practiced only in the United States. Both large and small U.S. companies expressed concern about the highly restrictive nature of DOE's initial set of guidelines, which they believed would weaken their ability to compete globally. Required to manufacture some prod- ucts abroad in order to meet foreign content laws, provide just-in-time delivery services, and maintain competitive freight charges, and unwilling to accept whole- sale liens on CRADA-related intellectual property, the big three U.S. automakers and the nation's leading computer manufacturers insisted that the competitive- ness article be waived or modified in the DOE CRADA agreements.38 Several major U.S. multinational companies have indicated that delays in the CRADA negotiation process related to the U.S. competitiveness article discouraged them from concluding CRADAs with DOE. In 1993, DOE took steps to streamline the CRADA negotiation process in response to criticisms that its procedures were too bureaucratic and time-consum- ing. In the process, the agency also added some flexibility to the U.S. competi- tiveness article by defining a process and general criteria by which a would-be CRADA participant could satisfy the performance requirement without acceding

116 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT to the highly restrictive "sample language" or the all-inclusive definition of intel- lectual property.39 However, in order to take advantage of these alternative eligi- bility guidelines, DOE still must approve the company's eligibility before the firm and laboratory can proceed with their joint work statement (U.S. Department of Energy, 1993~. In short, this requirement continues to impose significant de- lays on the CRADA negotiation process for firms that will not accept the "sample language." Subsequent to the introduction of performance and reciprocity requirements for foreign-controlled firms into CRADA regulations, the federal government has extended similar requirements to cover a range of financial-assistance agreements, such as contract research, joint ventures, and research grants, between federal laboratories and private companies.40 Most noteworthy in this regard is the En- ergy Policy Act of 1992, which includes more detailed economic performance requirements for participating firms as well as more extensive reciprocity re- quirements for foreign-controlled firms.4i With regard to the latter, DOE is obli- gated to determine whether the prospective foreign participant's home govern- ment: allows U.S.-owned companies opportunities, comparable to those afforded to any other company, to participate in any government-sponsored joint ventures similar to those authorized under the act; affords U.S.-owned companies local investment opportunities comparable to those afforded foreign firms in the United States; and affords adequate and effective protection for the intellectual property of the U.S.-owned companies. Several large U.S.-owned companies have criticized DOE for lax and incon- sistent enforcement of the eligibility requirements set forth in the Energy Policy Act of 1992. General Electric cited both the performance requirements and the reciprocity provisions of section 2306 of the act when it objected to DOE plans to include the European-based company Asea Brown Boveri in a program launched in 1993 to develop advanced gas turbine concepts under a cost-sharing agreement with the agency.42 Similarly, AlliedSignal Incorporated, also citing section 2306, has objected to Oak Ridge National Laboratory's decision to include a Japanese company, Kyocera, in an advanced ceramics research project.43 FOREIGN PARTICIPATION IN RECENT FEDERAL INDUSTRIAL TECHNOLOGY INITIATIVES Federally funded industrial R&D programs and consortia, such as the De- partment of Commerce's Advanced Technology Program (ATP), the Technology Reinvestment Project (TRP), administered by the Advanced Research Projects Agency (ARPA), the Semiconductor Manufacturing Technology Research Cor- poration (SEMATECH), and the U.S. Display Consortium (USDC), also have received considerable attention with regard to foreign corporate participation. These initiatives are among the most high-profile elements of recent efforts by the federal government to expand its support of industrially relevant civilian tech-

PUBLICLY FUNDED R&D 117 nology development and application (Advanced Research Projects Agency, 1993; Committee on Science, Engineering and Public Policy, 1992; National Academy of Engineering, 1993~. The ATP was established under the 1988 Trade Act to fund R&D in busi- nesses, especially small and medium-sized companies. The goal of ATP is to help firms develop generic, precompetitive technology that will stimulate high- risk, high-potential products, processes, and technologies. Eligibility require- ments for foreign firms wishing to participate in the program are set forth in the 1991 Technology Administration Authorization Act and are identical to those of the 1992 Energy Policy Act.44 The Commerce Department decides whether a foreign-owned company is eligible to participate in the program on a case-by- case basis. Since the program's inception in 1990, 15 of 413 participating orga- nizations in ATP have been foreign owned.45 Thus far, ATP administrators have not found it difficult to implement the eligibility requirements for foreign-owned firms (U.S. General Accounting Of- fice, 1994a). However, it is generally acknowledged that the uncertainties sur- rounding the certification process, particularly those related to reciprocity, tend to discourage foreign-owned firms from applying for ATP funds. It is notewor- thy that while foreign-owned firms represented only 3 percent of organizations competing for ATP funds in 1994, half of these companies received ATP awards compared with only 7 percent of all other applicant organizations.46 The TRP was launched in March 1993 with a budget of $472 million "to stimulate the transition to a growing, integrated, national industrial capability" through the support of technology development and deployment activities of U.S companies. The TRP, managed by ARPA in cooperation with five other federal agencies, elicited more than 2,800 proposals during its first phase. As of October 1994, the program had funded on a cost-sharing basis 212 projects involving 1,631 organizations, only five of which were foreign firms. Two of those, Mitsui Engineering and Shipbuilding (Japan) and Kvaerner Masa Marine (Canada), are involved in the TRP project designed to help Bath Iron Works Corporation of Maine diversify and modernize its shipbuilding operations to compete more ef- fectively in the global commercial shipbuilding market.47 Foreign firms seeking to participate in a TRP consortium must meet eligibility requirements that are basically the same as those for the ATP.48 However, regardless of the nationali- ties of the countries participating, only U.S.-owned firms may submit proposals. Unlike the ATP and the TRP, most of the publicly subsidized industry-led R&D consortia established in the United States during the past decade have not permitted foreign membership, although some nevertheless involve foreign firms. One such example is SEMATECH, a consortium of 14 U.S. semiconductor manu- facturers founded in 1987 to provide U.S. manufacturers with the capability to achieve world leadership in semiconductor manufacturing technology by 1993. Since 1988, DOD has provided half of SEMATECH's $200 million annual oper- ating budget.49 Membership in the consortium is restricted to U.S.-owned com-

118 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT panics. However, SEMATECH has entered into technical alliances with its Eu- ropean counterpart, the Joint European Semiconductor Submicron Initiative (JESSI). Furthermore, most of the U.S. companies participating in the consor- tium are extensively involved in various types of commercial and technical alli- ances with foreign firms.50 The U.S. Display Consortium (USDC), another industry-led, public-private partnership, was launched in mid-1993 with $20 million in public funds "to orga- nize the U.S. manufacturing expertise to develop the U.S. infrastructure required to support a world-class manufacturing capability for high definition displays in the United States." To achieve this, the USDC brings together 13 flat-panel dis- play manufacturers and developers, 52 flat-panel display equipment and materi- als suppliers, the flat-panel display user community, and the U.S. government. Membership in the consortium is currently limited to U.S.-owned firms.5i THE IMPLICATIONS OF FOREIGN PARTICIPATION IN PUBLICLY FUNDED U.S. R&D There are three major concerns about foreign participation in publicly funded U.S R&D. The first is that foreign students, researchers, and companies take away more technology, know-how and, most importantly, economic value Jobs and profits) from their involvement in U.S. publicly funded R&D than they give back to the United States. The second is that other nations' publicly funded R&D, conducted outside the United States, is not as accessible to U.S. citizens, particularly U.S.-owned firms, as is U.S. publicly funded R&D to foreign enti- ties. The third is that foreign governments, or more specifically, foreign taxpay- ers, may not be carrying their fair share of the global basic research burden. The following discussion addresses in turn each of these concerns and their associated implications. Is the Quid Pro Quo Adequate? There is no quantitative way to determine whether or not the United States receives benefits commensurate with those derived by foreign students, research- ers, and corporations from their access to and participation in U.S. publicly funded research. It is possible, however, to review the costs and benefits of different types of foreign R&D involvement, both to foreign nationals, those who carry out publicly funded U.S. research, and U.S. citizens. Such an analysis suggests that some forms of foreign participation yield a more adequate quid pro quo than others. Foreign Graduate Students, Postdoctoral Researchers, and Long-Term Visiting Researchers Clearly, foreign students, postdoctoral researchers, and other long-term vis- iting researchers derive many benefits from their involvement in publicly funded

PUBLICLY FUNDED R&D 119 U.S. research at universities and federal laboratories. They receive what many consider to be the world's finest advanced training in science and engineering, much of it subsidized directly or indirectly by U.S. taxpayers.52 In the process, they are given access to cutting-edge research and knowledge in many fields. In some cases, such as university-industry research centers and federal laboratories involved in CRADAs, they also have access to the proprietary research of U.S.- owned firms. Many of these foreign researchers or newly minted Ph.D.'s return eventually to their home country and take with them the skills and specialized knowledge acquired in the United States skills and knowledge likely to be em- ployed by the foreign competitors of many U.S.-owned companies. Although difficult to quantify, the nation's heavy reliance on foreign-born graduate students, postdoctoral researchers, and faculty may entail certain liabili- ties. Some argue, for instance, that the large presence of foreign faculty and teaching assistants in U.S. engineering programs may contribute to cultural ten- sion and language barriers that discourage U.S.-born students, particularly women and ethnic minorities,53 from pursuing engineering degrees. Others contend that foreign students are taking slots in graduate schools and jobs in the U.S. workforce that otherwise would go to Americans. Still others believe that the availability of abundant foreign talent may be allowing U.S. universities and the nation's educa- tional establishment generally to avoid addressing fundamental problems in the U.S. educational system.s4 While some of these arguments are more compelling than others, the com- mittee remains convinced that foreign students and long-term visiting researchers are an indispensable asset to U.S. university and federal laboratory research. As key members of U.S. science and engineering departments, foreign graduate stu- dents and postdoctoral researchers represent the underpinning of many advanced research projects. Without them, laboratories would have difficulty raising re- search support from government, industry, and internal sources. These research programs, in turn, induce high-quality faculty to stay. Numbers alone confirm the important role of foreign postdoctoral research- ers and long-term visiting researchers in the work of federal laboratories. In the DOE multiprogram labs, for example, there is on average one foreign visiting scientist or engineer for every five members of the resident technical staff. Simi- larly, the NIH hosts about one long-term foreign visiting researcher for every eight members of its resident staff. Foreign postdoctoral researchers and long- term visiting researchers have made numerous intellectual contributions that have strengthened U.S. university-based research. These investigators come from both companies and nonprofit institutions. Some examples include Tokyo Electric Power researchers' work in MIT's power engineering laboratory; Japanese con- struction company researchers' contributions to civil engineering research at MIT, Purdue, and Stanford (Stalson, 1989~; the contributions of French, Canadian, German, and Japanese researchers from academe, industry, and government— to research at the Center for Ultrafast Optical Science at the University of Michi-

120 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT gan, Ann Arbor (U.S. Congress, House, 1993~; and the critical role of Japanese scientists and engineers in the research program of the Kent State University's Liquid Crystal Institute.55 Furthermore, laboratory administrators attest to the significant intellectual contributions made by many of their foreign visiting researchers. In response to a 1988 GAO survey, a majority of the federal laboratory directors contacted con- cluded that overall "the federal laboratories and the United States benefitted more than foreign researchers and their countries through the collaboration on research and development" (U.S. General Accounting Office, 1988b). More recently, at a National Research Council workshop on Japanese participation in U.S. federal laboratories, an administrator at Sandia National Laboratory observed that "(f~or- eign visitors, Japanese, Germans, and others . . . combine interest in what is being done at Sandia with competence in the field and provide a stimulus which encour- ages Sandia researchers to innovate" (National Research Council, 1994a). At the same workshop, a representative from Pacific Northwest Laboratory noted that "compared to American companies, the Japanese send better-trained, better- educated people to the lab, and they are more proactive in seeking access to the lab." Once they graduate, foreign-born students may continue to benefit their host institutions. If foreign alumni stay on as faculty, they are often able to use their ties to their home countries to attract the next generation of foreign talent and to help lure research projects and other funds from firms and government agencies. If they leave the university for positions in U.S. or foreign industry, they are likely in the future to turn to their former graduate institutions to hire graduates, contract research, or seek technical assistance. Finally, like U.S.-born alumni, foreign graduates provide funds for endowed chairs, scholarships, and new or renovated facilities, and they encourage their employers to do likewise.56 U.S. industry also derives many benefits from the presence of foreign sci- ence and engineering students and faculty. Currently, more than half of foreign engineering graduate students enter and remain part of the U.S. engineering workforce for at least 2 years after graduation (Finn et al., 1995; National Science Foundation, 1993a). As of 1989, more than 10 percent of all doctoral-level engi- neers employed in the United States were foreign nationals (Figure 4.6~. U.S. companies may also derive indirect benefits from repatriated foreign alumni. It is claimed, for example, that the entire energy establishment of South Korea is run by graduates of Rensselaer Polytechnic Institute and MIT and that all senior civil engineering experts in the People's Republic of China are Cornell graduates. Pre- sumably, these individuals' educational experience in the United States would make them more inclined to call on U.S. firms when considering sources for foreign equipment and technology (Stalson, 1989~. The benefits reaped by U.S. industry and the U.S. economy as a whole from the inflow of foreign talent, however, are greater than numbers alone suggest.

PUBLICLY FUNDED R&D Total, scientists and engineers Total, scientists Physical scientists Earth, atmospheric, and ocean scientists Mathematical scientists Computer specialists Agricultural scientists Biological scientists Medical scientists Psychologists Social scientists Total, Engineers Aeronautical /astronautical Chemical Electrical/electron ics Materials Mechanical Nuclear Systems design Other 121 =G~==~=...... 2;"2:, ".i ".~";::."22,;.~>':.';. ' ;."' if'."".' '''.' .'' '' " "' "':""''."'.";'~'' ' : :'.: ' :.:.:.:;.~: -,:.:;:;: :::~:~.:i: :~.:i.:~: ~ :.:.:..::.:.:...: :.:.: :~.:...~.:.~:.:.:.:::.:.:::_ I. _ ::.,,.,.,_, l 1 2 3 4 5 6 7 8 9 10 11 12 13 14 FIGURE 4.6 Non-U.S. citizens as a percent of employed Ph.D. scientists and engineers in the United States, total and by field, 1989. SOURCE: National Science Foundation (1993a). Foreign-born graduates of U.S. universities constitute a large pool of elite techni- cal talent with special knowledge of their nation's or region's cultures, languages, political economies, and markets knowledge that is particularly valuable to U.S. firms as economic competition becomes increasingly global. In conclusion, the committee is convinced that the participation of foreign graduate students, postdoctoral researchers, and other long-term visiting research- ers in U.S. academic and federal laboratory research has yielded significant net benefits to the U.S. economy and national innovation system.

122 Short-Term Visiting Researchers FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT In contrast, the committee believes that short-term foreign visiting research- ers those whose stay in the United States lasts between 3 months and 1 year are likely to take away from U.S. research universities and federal laboratories more than they contribute to them. There are few data on the number of short- term visitors to U.S. universities and federal laboratories, and the committee knows of no systematic effort to prove or disprove the more "extractive" charac- ter of foreign short-term visiting researchers. However, committee members who have hosted foreign visiting researchers at their institutions note that those indi- viduals climb a very steep learning curve during the first 10 to 12 months in their host laboratories. They are therefore less likely to make significant contributions to the work of the host lab if their tenure is less than a year. Presumably, steep learning curves also diminish the ability of short-term visitors to extract intellec- tual assets from their host institutions. However, anecdotal evidence suggests that some short-term foreign visitors have been particularly aggressive in collect- ing and transmitting back to their home institutions information about their U.S. host organization's research activities (National Research Council, 1 994a). More- over, the fact that each foreign visiting researcher has access to the work and talents of multiple resident U.S. researchers bespeaks an inherent imbalance in the flow of ideas and information between visitors and their host institutions in any individual transaction. Foreign Corporate Participation in Publicly Funded U.S. R&D Foreign-owned firms have benefitted from participating in publicly supported U.S. R&D in several ways. They have gained access to many of the world's leading researchers in various fields of science and engineering and have been able to assess and acquire intellectual property and human capital outputs of U.S. public R&D activities. To a certain extent, involvement in publicly funded re- search has also provided those companies another window on the R&D activities of U.S.-owned firms engaged in collaborative research at university research cen- ters or federal laboratories. Although it is impossible to place an economic value on these benefits, the committee believes they are greater than the current scope of foreign corporate participation measured in terms of research sponsored, re- search personnel exchanged, and technology licenses acquired suggests. Through their participation in the research activities of U.S. universities and federal laboratories, foreign corporations undoubtedly extract more intellectual property from the United States than they would in the absence of such ties. It is equally certain, in the view of the committee, that in some cases, foreign-owned firms and their stakeholders have benefitted at the expense of American firms and stakeholders. At the same time, the committee firmly believes that the flow of knowledge

PUBLICLY FUNDED R&D 123 and technology between foreign firms and U.S. universities and federal laborato- ries is not unidirectional and that foreign firms, through their participation in U.S. research, return economic value to U.S. citizens in many different ways. Numer- ous case examples confirm that foreign-owned firms have contributed material support as well as intellectual resources to U.S. research universities and federal laboratories. This has enhanced the productivity and quality of these institutions' research efforts, thereby strengthening the nation's overall research infrastruc- ture.57 Furthermore, whether foreign companies receive more from U.S. universi- ties and federal laboratories than they contribute to these institutions says very little about the impact of foreign involvement in this research on the welfare of Americans. There is some evidence suggesting that the foreign-owned compa- nies with the most extensive ties to U.S. publicly funded research institutions also tend to have significant U.S.-based manufacturing and R&D operations.58 These operations employ U.S. citizens, buy from U.S.-based suppliers and equipment vendors, import as well as generate technology and know-how applied within the U.S. economy, and pay U.S. taxes. Even if foreign companies commercialize abroad intellectual property li- censed from U.S. publicly funded research institutions, it should not be assumed automatically that their gain represents a loss for the United States. In some instances, foreign firms have acquired or licensed technology generated with U.S. public funds and gone on to develop and commercialize the technology over- seas.59 Yet, as noted, foreign-based licensees have also enabled publicly funded research institutions to earn revenues on technology that otherwise would not have been commercialized. Moreover, by commercializing this otherwise un- exploited technology, foreign firms have in some cases helped supply U.S. citi- zens with better or cheaper goods or services.60 As far as the transfer of codified technology is concerned, those who oversee the technology licensing offices of U.S. universities and federal laboratories sug- gest that foreign-owned companies are for the most part "customers of last resort" for U.S. government-subsidized intellectual property (U.S. General Accounting Office, 1988b; Wiley, 1993~. That is, these companies license technology that no U.S.-owned company is willing to invest in. In addition to their preference for dealing with U.S.-based firms, U.S. re- search universities and federal laboratories, in the committee's judgment, have generally made good-faith efforts to comply with federally mandated economic performance requirements. There have been several well-publicized instances in which publicly supported U.S. R&D institutions have negotiated R&D coopera- tion or licensing agreements with foreign firms that do not appear to be in the best interests of the United States. The committee believes that these cases are the exception rather than the rule.6i Increased fiscal austerity at the federal level, defense downsizing, and grow- ing pressure on publicly funded research institutions to contribute to national and

24 FOREIGN PARTICIPATION lN U.S. RESEARCH AND DEVELOPMENT regional economic development have made research universities and federal labo- ratories aware of the need to demonstrate their service to the nation's economic interests. Moreover, these publicly funded institutions and their foreign corpo- rate customers or patrons appear equally aware of the public-relations hazards associated with involving foreign firms in publicly supported research. As a result, they exercise considerable care to avoid even the appearance of impropri- ety.62 Indeed, the negative fall-out from the few high-visibility cases has, in the committee's opinion, discouraged foreign corporate involvement in U.S. govern- ment-subsidized R&D. At the same time, the manner in which some federal agencies have enforced the performance requirements embodied in recent technology transfer legislation suggests they may be overreaching in their efforts to keep the economic benefits of public R&D spending within the United States. Some agencies have devel- oped economic performance criteria and procedures to enforce them that in the committee's view, are fundamentally at odds with the competitive R&D and tech- nology management practices of multinational companies, strongly discourage foreign corporate involvement, and are at times in conflict with the core missions of the agencies themselves. In order to engage U.S.-owned multinational compa- nies in CRADAs and other forms of collaborative R&D, some agencies have, on a case-by-case basis, made a more liberal interpretation of what constitutes an adequate economic quid pro quo. Nevertheless, extensive restrictions invoked in the name of U.S. economic interests—on private-sector use of publicly subsi- dized intellectual property continue to discourage leading U.S.-owned high-tech- nology companies from collaborating with federal laboratories. With respect to foreign-owned firms, the interpretation and enforcement of economic performance requirements by various federal agencies appear to be particularly vulnerable to political and legal challenges from without. Hence, these requirements serve to discourage the involvement of foreign-owned firms that might otherwise seek to participate in publicly funded U.S. R&D. Given the growth of technical competence overseas and the prominent role of U.S.- and foreign-owned multinational companies in areas of R&D and technology of di- rect relevance to the core missions of U.S. federal agencies, the committee be- lieves it is not unreasonable to question whether economic performance require- ments as currently administered truly serve the nation's interests. Reciprocal Access to Publicly Funded Foreign R&D Reciprocal access describes, in a comparative way, the access U.S. and for- eign citizens have to one another's R&D enterprises. In part because the costs and benefits to the United States in this area are difficult to quantify, reciprocal, or "equitable," access is viewed by many U.S. policymakers as an important indi- rect measure of whether foreign participation in the U.S. R&D enterprise is an asset or a liability. Not only does a lack of reciprocal access offend the American

PUBLICLY FUNDED R&D 125 public's sense of fairness, it may also reduce the ability of U.S.-owned firms to compete internationally. Inspired by this reasoning, U.S. lawmakers have made reciprocal access a prerequisite for foreign corporate participation in many areas of publicly funded U.S. R&D.63 Several recent surveys suggest that U.S. government, academic, and indus- trial researchers have had few problems accessing research capabilities and ac- tivities within academic and government-operated laboratories abroad. For ex- ample, an informal survey (National Science Foundation, 1990b) of U.S. companies with foreign R&D operations in 1990 found that all sponsored univer- sity-based research in their host countries. None of the firms noted any problems establishing relationships with universities in Europe, Canada, South America, or Australia. Of 26 officials interviewed, three said their efforts to tap Japanese university research met with some government resistance.64 At the same time, others have found that U.S. and European companies invest considerably less effort drawing on the research capabilities of U.S. research universities than do their Japanese competitors (Roberts, 1995a). A 1988 GAO survey found that research managers and administrators at eight federal laboratories did not have "difficulty getting access to foreign labora- tories and that, except for some isolated instances, foreign researchers have readily exchanged information with federal laboratory researchers" (U.S. General Ac- counting Office, 1988b). In 1989, researchers at NIST reported that they had good access to Japanese national laboratories, including those of the Ministry for International Trade and Industry (MITI) and Nippon Telephone and Telegraph (NTT) (National Research Council, 1989a). What little information exists concerning access by U.S. industry-based re- searchers to foreign government laboratories suggests that at least for intramural research activities, accessibility has not been a problem. As noted by a National Research Council report on U.S. access to Japanese government laboratories, the principle concern may be inadequate exploitation of foreign laboratories by U.S. researchers, not lack of access (1989a). To address this concern, the National Science Foundation and other federal agencies have initiated programs designed to encourage U.S.-based researchers from academe, industry, and government to spend time working in foreign research institutions.65 Access by U.S.-owned companies to government-funded industrial R&D consortia in Europe appears to be fairly comparable to access by foreign-owned firms to consortia in the United States. Participation in European Community research and development (EC R&D) programs, for example, is open to all legal entities established and regularly carrying out research within member countries. Moreover, all contractors taking part in community-funded R&D are obligated to exploit the results of that work "in conformity with the interests of the commu- nity."66 As of 1993, 25 U.S.-owned companies had participated in 53 projects (roughly 5 percent of all projects) of the community's largest research program, the European Strategic Programme for Research and Development in Informa-

126 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT tion Technology. At least 1 U.S. company is participating in each of the 14 other EC R&D programs.67 The policies of EC member states regarding foreign par- ticipation in national R&D programs appear to be consistent, at least on paper, with those of the EC. These policies require that corporate participants be estab- lished in the sponsoring country and exploit resulting intellectual property in a manner consistent with the sponsoring nation's interests.68 In Japan, access by foreign firms to publicly funded industrial R&D pro- grams appears more tightly regulated than it is in the United States or in the European Community. Much more than its American or European counterparts, the Japanese government uses privately owned industrial consortia to plan, ex- ecute, and supervise most of the nation's investment in industrially relevant R&D. For the most part, U.S. and European Community R&D programs award extra- mural grants and contracts through relatively transparent and competitive selec- tion processes. That is, firms, universities, and other nonprofit research organiza- tions compete for funding by submitting proposals that are then evaluated for their relative merit by panels of experts. In contrast, the development and imple- mentation of publicly funded industrial R&D programs in Japan is characterized by a more closed and arguably more strategic process of project and participant selection. Leading Japanese companies in a particular technology area negotiate the establishment and funding of particular research programs with MITI or other government agencies (Hane, 1993; Heaton, 19881. In Japan, foreign firms are generally not directly involved in the planning of a particular research program, but some, including U.S. companies, have been invited specifically to participate in certain initiatives because of the particular expertise they bring to the table. For example, MITI invited Motorola to partici- pate in the research consortium on micromachine technology, Texas Instruments was asked to join the MITI program on atomic manipulation, and IBM joined an agency-supported research program on fuzzy logic.69 However, as suggested by repeated unsuccessful attempts by the U.S. company AlliedSignal Inc. to join a MITI-sponsored research program on advanced structural ceramics, the alloca- tion of public research contracts and grants in Japan is not decided through a transparent competitive process solely on the relative technical and economic merits of the research proposals submitted.70 There is little doubt that international asymmetries of access to publicly funded research persist and impose costs (however difficult to measure) on U.S. citizens. The committee is not convinced, however, that recent U.S. government attempts to redress these asymmetries through the introduction of reciprocity re- quirements in federal R&D legislation are a constructive response. Indeed, reci- procity requirements, in concept as well as implementation, appear to be fraught with many more liabilities and hazards for the United States than are economic performance requirements. First, conditioning the nondiscriminatory treatment of foreign-owned firms on their home governments' compliance with U.S. laws contradicts a longstanding

PUBLICLY FUNDED R&D 127 commitment of the United States to the unconditional "national treatment" of multinational enterprises, a principle of international economic intercourse that the United States with good reason has long urged other nations to abide by.7i Indeed, there is some evidence the inclusion of reciprocity requirements in osten- sibly "domestic" legislation has encouraged at least a few major U.S. trading partners to consider introducing similar rules.72 Second, the process of assessing whether a particular firm's home country is in compliance with each of the various reciprocity requirements is cumbersome, time consuming, and very difficult to administer. Enforcement is made all the more difficult by the fact that federal agencies are called upon to render judg- ments in areas of policy where it is often difficult to determine what constitutes "comparable investment opportunities" or "adequate protection of U.S. intellec- tual property." The vagueness of these requirements has made it possible for some U.S.-owned companies to challenge both the eligibility of particular for- eign firms to participate in the U.S. R&D enterprise as well as the enforcement processes of some federal agencies. The committee believes that these chal- lenges have had a chilling effect on the interest of foreign-owned firms (as well as of federal R&D administrators) in participating federally funded R&D initiatives. Given the differences in the structure and scale of publicly funded R&D activities among countries, the value to the United States of having reciprocal access will vary. In Japan, for instance, a much smaller share of leading-edge basic or generic research is reportedly conducted within publicly funded institu- tions than is true in the United States. Such asymmetries shift the debate for some from reciprocal access to the broader issue of international burdensharing in basic research. International Burden Sharing in Basic Research In terms of its scale, scope, accessibility, and productivity, U.S. basic re- search is globally preeminent. It is clear that foreign companies and countries draw more heavily on the results of publicly funded U.S. basic research than the United States or U.S. companies do on the research output of publicly funded research abroad. Nonetheless, international comparisons of R&D expenditures confirm that most of America's important trading partners invest as much if not more of their gross national product in basic research than does the United States (Table 4.8~. Japan is an important exception to this general trend, spending a smaller share of its gross domestic product and nearly half as much per capita on basic research as the United States (Irvine et al., 1990~. Given Japanese companies' success in accessing U.S. basic research, the relatively low level of Japanese basic research spending has been particularly disturbing to many observers, who see it as evidence of Japanese free riding on the basic research output of others. This negative perception is compounded by the widely held view that much of

128 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT TABLE 4.8 Govemment Expenditures on Academic and Academically Related Research as a Percentage of GDP and Per Capita' 1987 Expenditure as a percentage of GDP (1987) Expenditures per capita (1987 $a) United Kingdom 0.211 26.1 Germany 0.261 34.7 General France 0.135 17.2 university Netherlands 0.329 40.3 funds United States 0.070 12.7 Japan 0.156 20.6 United Kingdom 0.083 10.3 Germany 0.090 11.9 Separately France 0.140 17.9 Academic budgeted Netherlands 0.106 13.0 research research United States 0.223 40.6 Japan 0.055 7.3 United Kingdom 0.295 36.4 Germany 0.351 46.7 Subtotal France 0.275 35.0 Netherlands 0.435 53.3 United States 0.293 53.3 Japan 0.212 27.9 United Kingdom 0.103 12.7 Academically Germany 0.145 19.3 related France 0.178 22.7 research Netherlands 0.098 12.0 United States 0.043 7.9 Japan 0.021 2.7 United Kingdom 0.398 49.1 Germany 0.496 66.0 Total France 0.453 57.7 Netherlands 0.532 65.3 United States 0.336 61.1 Japan 0.232 30.6 aSpending in national currencies converted to U.S. dollars using OECD 'purchasing power pari- ties' for 1987 calculated in early 1989. SOURCE: Irvine et al. (1990).

PUBLICLY FUNDED R&D 129 what the Japanese classify as basic research is, in fact, more application oriented and proprietary in character than similar research conducted by other advanced industrial economies. Moreover, many believe that the most important basic re- search in Japan takes place within closed corporate laboratories rather than in the more open and accessible publicly supported research institutions (Hicks, 1994; Hicks and Hirooka, 1992~. SUMMARY The participation of foreign individuals and institutions in publicly funded R&D activity in the United States appears to be extensive and growing. Govern- ment surveys of the number of foreign graduate students, postdoctoral research- ers, and other visiting researchers at U.S. universities and federal laboratories document a significant increase in the level of foreign involvement since the mid- 1970s. At the same time, meaningful data on the scope, growth, and nature of foreign institutions' involvement in publicly funded U.S. R&D are fragmentary, dated, and scarce. Available information suggests that foreign institutions account for a very small share (less than 2 percent) of total sponsored research at U.S. universities and federal laboratories, which is concentrated in a small number of U.S. institu- tions. As of the mid-1980s, Japanese institutions sponsored more U.S. univer- sity-based R&D than firms of any other nationality. During this period, most foreign-sponsored research at U.S. universities was funded by not-for-profit in- stitutions and was focused mostly in agriculture, medicine, and geology. Only fragmentary data exist with which to assess the current magnitude, disciplinary focus, and national shares of foreign-sponsored research at individual U.S. uni- versities and federal laboratories. However, limited survey data and anecdotal evidence indicate that Japanese companies are more diligent than their U.S. or European counterparts with regard to accessing, monitoring, and drawing upon the research capabilities of these publicly funded institutions. Foreign participa- tion has been minimal in recently established, federally supported industrial R&D initiatives, such as the Department of Commerce's Advanced Technology Pro- gram. As with U.S. privately funded R&D, the involvement of foreign nationals in R&D sponsored by the federal government carries both risks and opportunities for the United States. The committee believes that the extensive presence of foreign graduate students, postdoctoral researchers, and other long-term foreign visiting researchers at U.S. universities and federal laboratories has, on balance, yielded significant benefits to the U.S. economy and its innovation system. In contrast, the committee believes that short-term visiting researchers have contrib- uted much less to the work of U.S. research universities and federal laboratories, both overall and relative to what the researchers themselves take away. Although the committee could identify various costs and benefits of foreign

130 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT institutional involvement in U.S. publicly supported research institutions, it was unable to determine the economic impact of this involvement. Through their participation in the research activities of U.S. universities and federal laborato- ries, foreign corporate participants undoubtedly extract more knowledge and in- tellectual property from the United States than they would in the absence of such ties. Likewise, the committee does not doubt that access to U.S. publicly funded R&D activities and institutions has brought extensive benefits to foreign-owned firms and their stakeholders abroad, in some cases at the expense of American firms and their American stakeholders. Nevertheless, there is also considerable evidence to confirm that foreign firms have contributed significant material support, technology, and intellectual re- sources to research universities and federal laboratories. Indeed, many if not most foreign-owned companies that have extensive ties to U.S. publicly funded research institutions appear to be good corporate citizens. For instance, they establish U.S.-based manufacturing and R&D operations that employ Americans, pay U.S. taxes, buy from U.S.-based suppliers and vendors, and import as well as develop technology and know-how that is then applied in this country. U.S. research universities and federal laboratories, in the committee's judg- ment, have generally made good-faith efforts to comply with federally mandated economic performance requirements for foreign company participation in U.S. publicly funded research. Foreign-owned companies are for the most part "cus- tomers of last resort" for U.S. government-subsidized intellectual property. However, the highly restrictive and somewhat inconsistent enforcement of these requirements by some federal agencies has raised questions about how well these requirements serve agency missions specifically and national interests more gen- erally in the increasingly global economy. Aside from this general bias against multinational companies (both U.S. and foreign-owned), the performance require- ments and their underlying political-economic logic serve as a disincentive to potential foreign-owned participants in publicly funded U.S. R&D activity. With regard to mandated reciprocity requirements, it appears that federal agencies have yet to develop effective and credible procedures for establishing compliance or noncompliance. Indeed, in conception as well as implementation, reciprocity requirements appear to be fraught with significant short- and long- term liabilities and hazards for the United States. U.S. government, academic, and industrial researchers seem to have few problems accessing publicly funded research capabilities and activities in aca- demic and government-operated laboratories abroad. Furthermore, access by U.S.-owned companies to government-funded industrial research consortia in Europe appears to be comparable to that extended foreign-owned firms in the United States. Japan, however, has restricted foreign participation in its publicly funded industrial R&D consortia to a greater extent than either the United States or the European Community through its closed, strategic process of project and participant selection.

PUBLICLY FUNDED R&D 131 Given differences in the organization, scale, and sophistication of publicly funded R&D activities among countries, the value to Americans of greater access to publicly funded foreign R&D is not clear. Nevertheless, lack of reciprocal access may have damaging results. In some cases, it may disadvantage U.S.- owned firms in international competition, with negative consequences for their U.S.-based stakeholders. But its greatest cost to the nation may be the extent to which it offends the American public's sense of fairness, thereby undermining public support for efforts to negotiate remedies to these asymmetries in interna- tional forums. International comparisons of R&D expenditures confirm that most of Amer- ica's advanced industrialized trading partners invest as much if not more of their gross national product in basic research than does the United States. Japan, de- spite recent efforts to expand its basic research capabilities, spends a smaller share of its gross domestic product and only about half as much per capita in this area compared with the United States. The committee believes strongly that the federal government should continue to encourage Japan to assume a role in the global basic research community that is commensurate with its industrial, techno- logical, economic, and diplomatic standing in the world. NOTES 1. In 1994, public funds accounted for 36.1 percent of all R&D dollars invested in the United States 57.6 percent of all basic research funding, 36.6 percent of all applied research funding, and 29.7 percent of the nation's total investment in technology development (National Science Founda- tion, l995b). Data on the scope, composition, nature, and growth of foreign participation in U.S. publicly funded research is more scarce and fragmentary than those on foreign involvement in U.S. privately funded R&D. Accordingly, it is more difficult to assess its importance and consequences for the U.S. innovation system and national economy than foreign participation in U.S. privately funded R&D activity. 2. U.S. affiliates of foreign-owned firms employed 104,500 U.S. residents in R&D activity in 1992 (U.S. Department of Commerce, 1995). 3. For a discussion of the reasons foreign firms become involved in U.S. academic research, see National Science Board (1993), Serapio (1994), Stalson (1989), Massachusetts Institute of Technol- ogy (1991), and U.S. Congress, House (1993). It is instructive to compare the findings of the above-cited studies with those of a National Science Foundation (NSF) report (199Ob), which among other things reported on reasons why U.S. firms invest in academic research abroad. Based on informal inquiries to 21 U.S. multinational com- panies, accounting for more than half of U.S. overseas industrial R&D in 1988, NSF found that the principal reason these firms sponsored research at foreign universities was "to benefit from the work of individual scientists and engineers who are leading, world-renowned experts in their respective fields." The desire to foster goodwill and promote the company's reputation within the scientific community as well as the desire to take advantage of tax laws also were cited as important incentives. 4. These data probably overestimate the share of foreign-born engineering faculty, according to several experts, due to problems with the survey methods used and a lack of information on emigra- tion of foreign-born Ph.D's. 5. The University of Texas at Austin reported 181 international scholars in science and engineer-

32 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT ing fields for fiscal 1993, roughly 10 percent funded by foreign companies; correspondence from Dale Klein, University of Texas at Austin, to Proctor Reid, NAE, March 17, 1995. 6. Remarks by J. William Doane at the National Academy of Engineering workshop on the flat panel display industry, Committee on Technological Innovation in Small Business, December 7, 1993; correspondence from Doane to Proctor Reid, NAE, March 31, 1995. 7. The U.S. General Accounting Office defined foreign sources as foreign governments and individuals, nonprofit organizations headquartered in a foreign country, businesses headquartered in a foreign country, U.S. subsidiaries of foreign corporations, and joint-venture businesses in which the foreign partner has a controlling interest (U.S. General Accounting Office, 1988a). 8. In 1990, the National Science Foundation Division of Science Resources Studies conducted an informal telephone inquiry into fiscal 1988 levels of foreign funding of U.S. academic research. Division staff contacted essentially the same population of universities that the General Accounting Office had surveyed in 1988. 9. The data collected by the NAE via an informal telephone inquiry looked at foreign-sponsored research as a share of total sponsored research (not total R&D expenditures) at the institutions can- vassed. Not all of the institutions approached were able to distinguish readily between research spon- sored by foreign organizations and that sponsored by U.S. organizations. Moreover, tracking and ac- counting procedures with regard to sponsored research appear to vary significantly among institutions. Representatives from several institutions contacted by the NAE noted that the volume of for- eign awards received can fluctuate, in some cases significantly, from one year to the next. Neverthe- less, multiyear data provided by several of the top 20 research universities show that the foreign share of total sponsored research averaged over 3 to 5 fiscal years (FY) was very close to its present share for the single-FY estimate provided in Table 4.2 for FY 1993 or FY 1994. For example, the foreign share of total sponsored research at the University of Minnesota (FY 1992-94), Stanford and Penn State (FY 1989-93), and the University of Illinois-Urbana (FY 199~94) averaged 0.5 percent or less. At the University of California at Berkeley, the foreign share averaged 1.7 percent for the 5-year period, FY 1989-93. 10. This is not to say that the only, or even the most important, reason foreign-owned firms par- ticipate in U.S. university-based research is to acquire commercially valuable intellectual property. Available data suggest that much of the financial and other material "research" support provided to U.S. universities by foreign-owned companies takes the form of outright gifts and grants for which no specific deliverables other than "goodwill" are promised in return. Nevertheless, the fact that a grow- ing number of foreign-owned companies are availing themselves of the relatively free access to the intellectual assets and research activities resident at U.S. universities has led many American observ- ers to question whether the nation loses more than it gains from foreign corporate participation in the U.S. academic research enterprise. 1 1. For example, during the 1991-1992 academic year, Berkeley's College of Engineering hosted 17 visiting industrial fellows from Japanese companies that had provided between $2 and $3 million in research support, mostly in the form of grants or gifts. The fellows' salaries and expenses were paid by their companies. Those sent are typically engineers with M.S. degrees and 5 to 10 years experience in the company's R&D laboratory. (Personal communication between David Hodges, University of California at Berkeley, and William Spencer, SEMATECH, January 20, 1992.) 12. In 1989, Helena Stalson prepared a draft report for the National Academy of Engineering, "Foreign Participation in Engineering Research at U.S. Universities," which was based on interviews conducted at the following eight universities: Carnegie Mellon, Columbia, Cornell, Massachusetts Institute of Technology, Princeton, Rensselaer Polytechnic Institute, University of Illinois (Urbana), and University of Wisconsin (Madison). 13. Correspondence from Marion Lentz, UC Berkeley, to Proctor Reid, NAE, June 14, 1994. 14. The agreement signed by Massachusetts General Hospital and Shiseido Co. Ltd. in August 1989 provides for Shiseido to spend up to $9 million a year over 10 years to support a 100-person skin research center at the hospital's Charlestown Navy Yard facilities in return for rights to commercial-

PUBLICLY FUNDED R&D 133 ize any discoveries the center makes. (See "Dollars for Science" in the September 10, 1989, Boston Globe.) In 1989, Shiseido established two U.S.-based research centers, the Cutaneous Biology Re- search Center and Shiseido America Technocenter (Dalton and Serapio, 1995). 15. As of 1993, Ciba-Geigy had nine U.S. research facilities working in areas of biotechnology, pharmaceuticals, advanced components, chromatography, resins and plastics, and plant development (Dalton and Serapio, 1995). 16. See Gibbons (1992), Healy (1993), and National Institutes of Health (1994a,b). As of 1993, Sandoz Pharmaceutical employed roughly 180 R&D professionals at its research and manufacturing facility in East Hanover, NJ. The Swiss parent company also had seven other U.S.-based research facilities (Dalton and Serapio, 1995). 17. In 1993, the University of California at San Francisco (UCSF) and Daiichi Pharmaceuticals entered into an agreement to establish a center to study atherosclerosis. In exchange for providing $20 million in research funding over 5 years, Daiichi received first rights to negotiate for exclusive licenses to any drugs developed at the center as well as the right to delay publication of research results until it decided whether or not to apply for a patent. It is noteworthy that UCSF requires foreign sponsors of university research to pay all direct and indirect costs of research for which they are awarded patent rights (Helm, 1994). 18. In her testimony before the House Subcommittee on Regulation, Business Opportunities and Energy of the Committee on Small Business on March 11, 1993, former NIH Director Bernadine Healy noted how the Scripps-Sandoz agreement illustrated potential contradictions in U.S. policy concerning the Bayh-Dole Act and agency implementing regulations. "These global first refusal rights may seem to conflict with some of the policy behind Bayh-Dole, such as the preference for collaboration with small business and the promotion of free competition and enterprise. However, the grant of these global rights may not conflict with the U.S. Department of Commerce regulations implementing Bayh-Dole [37 CFR § 401.1(a)] [which] provide for: '. . . the rights of research orga- nizations to accept supplemental funding from other sources for the purpose of expediting or more comprehensively accomplishing the research objectives of the government sponsored project.' Simi- larly, [37 CFR § 401.7] states that the small-business preference: . . is not intended, for example, to prevent nonprofit organizations from providing larger firms with a right of first refusal or other options in inventions that relate to research being supported under long-tenn or other arrangements with larger companies. Thus, the Scripps-Sandoz agreement illustrates a potential contradiction between what some view as the policy underlying the Bayh-Dole Act and the regulations implementing it. "The Bayh-Dole Act articulates multiple policies and objectives which may prove difficult to reconcile. For example, an action by a nonprofit organization, while clearly advancing one objective of the Act, such as promoting collaborations between commercial and nonprofit organizations, might not advance and may, in fact, appear to undermine another objective of the Act, like promoting free competition and enterprise. Needless to say, application of the objectives of the Bayh-Dole Act requires a balancing of statutory objectives in the overall best interests of the American public. That is the challenge for NIH and other agencies that support research. It would be foolhardy and irrespon- sible for federal agencies to champion open competition at the expense of denying the institutions it supports an opportunity to seek additional research funding. To restrain collaborations between fed- eral grantees and industry simply because the competitive advantage of a given collaboration may not be readily apparent, while simultaneously undermining the best opportunities to rapidly commercial- ize inventions, would also be unwise. In short, we must achieve a balance, remembering that the general intent of the legislation is to promote product development not commercial monitoring." (Healy, 1993). 19. For an informative discussion of the origins and consequences of Bayh-Dole, see Wisconsin BioIssues (1994). 20. The 18 foreign companies granted exclusive licenses included 4 French firms, 3 British com- panies, 2 Swiss companies, and 1 Canadian, 1 Israeli, 1 Finnish, and 1 Japanese company. Five licenses were awarded to one or more unidentified foreign-owned pharmaceutical company(ies). At

134 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT least five of the foreign firms of known nationality have affiliates of parent companies with manufac- turing or R&D facilities in the United States (U.S. General Accounting Office, unpublished data). 21. Correspondence from Joel Moses, MIT, to Proctor Reid, NAE, June 14, 1994; correspondence from Jean Fort, UCSD, to Proctor Reid, NAE, April 3, 1995; correspondence from Homer Neal, University of Michigan, to Proctor Reid, NAE, April 4, 1995; correspondence from Robert Nerem, Georgia Institute of Technology, to Proctor Reid, NAE, February 7, 1995. 22. See, for example, the testimony of Susan D. Wray, director, Office of Patent, Copyright and Technology Licensing, University of Florida at Gainesville in U.S. Congress, House (1993). 23. Since a foreign firm could participate in more than one university's industrial liaison program, the number of foreign firms participating is probably less than the 499 reported (U.S. General Ac- counting Office, 1992, p. 17). 24. It should be noted that neither the National Science Foundation nor the National Institutes of Health guidelines for universities and other recipients of federal research support address the issue of whether industrial liaison program members should be given advance access to research results. 25. Foreign participants in the University of California at Berkeley's industrial liaison programs pay fees twice as high as those submitted by U.S.-owned firms. MIT charges foreign-owned firms roughly 30 percent more than it does U.S.-owned firms. 26. This section draws heavily on University-lndustry Research Centers in the United States by Cohen, Florida, and Goe (1994). 27. Seed money for these centers has been provided to a select group of universities by the spon- soring federal agency with the expectation that the host institutions will raise matching funds from industry, state and local governments; and internally. 28. See, for example, discussion of the efforts by state governments to attract foreign direct in- vestment and associated economic activity to their jurisdictions in Feller (1994) and Kayne (1992). 29. Data are from the NSF ERC Database. Proctor Reid, NAE, phone conversation with Lynn Preston, Engineering Education and Centers Division, NSF, December 1, 1995. Stalson's (1989) found that foreign participation in ERCs and other university-industry re- search centers at the eight universities she surveyed was relatively modest with only limited participa- tion from European firms in basic chemical and power engineering research. Moreover, it appeared that a number of European industrial commitments to these centers were "inherited" when European firms acquired participating U.S. firms. 30. Remarks by William Gear, President, NEC Research Institute, and Knut Merten, President and CEO, Siemens Corporate Research, at the NAE roundtable on foreign participation in U.S. re- search and development, November 9, 1993. 31. These findings, in turn, are consistent with the tendency of Japanese companies to acquire externally developed technology by way of licenses. See Chapter 3, pp. 56, 66-67. 32. The 10 multiprogram labs of DOE are: Argonne National Lab, Brookhaven National Lab, Idaho National Engineering Lab, Lawrence Berkeley Lab, Lawrence Livermore National Lab, Los Alamos National Lab, Oak Ridge National Lab, National Renewable Energy Lab, Pacific Northwest Lab, and Sandia National Laboratories (National Research Council, 1994a, p. 90). 33. National Research Council (1994a) and unpublished data provided to the study committee by the National Institutes of Health. 34. The 1986 Federal Technology Transfer Act authorized CRADAs between government-oper- ated laboratories and industry. Under a CRADA, a private organization provides personnel, equip- ment, and/or financing for a specified R&D activity that complements the mission of its federal laboratory partner. CRADAs include provisions for the allocation of rights to intellectual property resulting from the cooperative research. In 1987, Executive Order 12591 directed agencies to del- egate authority for entering into CRADAs to their respective laboratories, and contained guidelines for the granting of intellectual property rights under these agreements. In 1989, the National Com- petitiveness Technology Transfer Act extended authority for negotiating CRADAs to contractor-op- crated government laboratories. See Committee on Science, Engineering, and Public Policy (1992).

PUBLICLY FUNDED R&D 135 35. Correspondence from Wanda Bullock, NIH, to Proctor Reid, NAE, April 4, 1995; correspon- dence from Robert Petit, ORNL, to Proctor Reid, NAE, April 10,1995; correspondence from Stephen Lake, ANL, to Proctor Reid, NAE, April 10, 1995. 36. See the Stevenson-Wydler Technology Innovation Act of 1980 (P.L. 96-480) as amended by the Technology Transfer Act of 1986 (P.L. 99-502) and the Advanced Technology Program Act of 1989 (P.L. 101-189), and Executive Order 12591. For further discussion of the reciprocity provisions of these and other recents acts of law, see Schwartz and Caplan (1993). 37. Proctor Reid, NAE, phone discussions, March 8, 9, and 31, 1995, with David Edgerly and Bruce Mattson, NIST, Office of Technology Services. NIST is currently reviewing its implementa- tion of the economic performance requirements of the Bayh-Dole Act and CRADA legislation. 38. The automobile industry's "umbrella" CRADA with DOE stipulates that the R&D will be done in the United States and that manufacturing for the first 2 to 3 years will be done predominantly in the United States and will continue in the United States where practical thereafter. 39 In 1993, DOE modified its modular CRADA agreement to offer would-be private sector par- ticipants the option of either accepting the "sample U.S. competitiveness language in toto" or agree- ing "to provide specific economic benefit to the U.S. economy under one or more criteria [eight in total] of the U.S. competitiveness work sheet." The sample U.S. competitiveness language states: In exchange for the benefits received under this CRADA, the Parties therefore agree to the following: A. Products embodying Intellectual Property developed under this CRADA shall be substan- tially manufactured in the United States; B. Processes, services, and improvements thereof which are covered by intellectual Property developed under this CRADA shall be incorporated into the Pc~icipant's manufacturing facilities in the United States either prior to or simultaneously with implementation outside the United States. Such processes, services, and improvements, when implemented outside the U.S., shall not result in reduction of the use of the same processes, services, or improvements in the United States; land C. In the event that it is not feasible to meet the requirements of A. and B., a plan for providing net benefit to the U.S. economy is attached in Document B [the U.S. competitiveness work sheet]. If a would-be participant chooses option C, it "must furnish a description of specific economic or other benefits to the U.S. economy which are related to the commercial use by Participant(s) of the technology being funded under the CRADA and which are commensurate with the Government's contribution to the proposed work." The benefits criteria set out in the U.S. competitiveness work sheet include: 1. Direct or indirect investment in U.S.-based plant and equipment. 2. Creation of new and/or higher-quality U.S.-based jobs. 3. Enhancement of the domestic skills base. 4. Further domestic development of the technology. 5. Significant reinvestment of profits in the domestic economy. 6. Positive impact on the U.S. balance of payments in terms of product and service exports as well as foreign licensing royalties and receipts. 7. Appropriate recognition of U.S. taxpayer support for the technology (e.g., a quid pro quo commensurate with the economic benefit that would be domestically derived by the U.S. taxpayer from U.S.-based manufacture). 8. Cross-licensing, sublicensing, and reassignment provisions in licenses that seek to max mize the benefits to the U.S. taxpayer." (U.S. Department of Energy, 1993, pp. 56-57, 86.) Whereas a DOE laboratory can move directly to receive approval of the CRADA joint work statement from its regional operations office if the would-be participants accepts the sample U.S. competitiveness language, the laboratory must receive approval first from its operations office or DOE headquarters of the participant's alternative statement of specific economic benefits to the U.S. economy before it can proceed with the joint work statement. (U.S. Department of Energy, 1993). In the DOE modular CRADA, intellectual property is defined as "patents, trademarks, copy- rights, mask works, protected CRADA information [generated information which is marked as being

136 FOREIGN PARTICIPATION IN U.S. RESEARCH AND DEVELOPMENT protected CRADA information by a party to the CRADA] and other forms of comparable property rights protected by Federal Law and other foreign counterparts." With the permission of their re- gional operations office, DOE laboratories may work with a more narrow definition of intellectual property that includes only patents, trademarks, copyrights, and mask works. 40. See, for example, the Energy Policy Act of 1992 (42 U.S.C. §13525), and the Technology Administration Authorization Act (15 U.S.C. §278n) in the American Technology Preeminence Act of 1991 P.L. 102-245. 41. For a company to be eligible to receive financial assistance under the Act, the DOE secretary must find "that the company's participation in the Program would be in the economic interest of the United States, as evidenced by investments in the United States in research, development, and manu- facturing (including, for example, the manufacture of major component or subassemblies in the United States); significant contributions to employment in the United States; and agreement with respect to any technology arising from assistance provided under this section to promote the manufacture within the United States of products resulting from that technology (taking into account the goals of promot- ing the competitiveness of United States industry), and to procure parts and materials from competi- tive suppliers." 42 U.S.C. §13525 Energy Policy Act of 1992, Sec. 2306 Limits on Participation by Companies. 42. "General Electric has circulated a paper implying that the Swedish-Swiss ABB should not be allowed to participate [in the advanced gas turbine programmed because its home governments do not protect intellectual property rights . . . The paper also suggests that if ABB were involved in the programme, it would use the fruits of the research not to create employment in the U.S. but to strengthen its technological capacity in Europe." (Financial Times, 1994) 43. AlliedSignal Corp. protested Oak Ridge National Laboratory's selection of a Japanese and a French firm in a conservation procurement for advanced ceramic manufacturing technology. How- ever, DOE refused to obtain information to make a finding under Section 2306. "Lack of DOE action and inconsistent application of the law has been the subject of Congressional correspondence from Senate and House authorization and appropriations committees." (Correspondence from Maxine Savitz, AlliedSignal Corp., to Proctor Reid, NAE, June 2, 1994; Inside Energy, 1994.) In response to criticism that it lacked a clear and consistent implementation strategy for these eligibility requirements, DOE submitted a statement of policy guidance for public comment in Febru- ary 1995. (Federal Register, 1995) 44. The Technology Administration Authorization Act of 1991 includes "buy American" provi- sions (15 U.S.C. §278n) that prohibit the use of funds authorized under the legislation to procure components manufactured in a foreign country whose government unfairly maintains a procurement policy discriminating against U.S. products or services. 45. Telephone discussion between Connie Chang, Advanced Technology Program, NIST, and Proctor Reid, NAE, February 24, 1995. 46. Ibid. 47. Two U.S.-based ocean carriers, American Automar Inc. and Great American Lines Inc., are also participants in the Bath Iron Works TRP. The Japanese and Canadian participants will transfer commercial shipbuilding technology to Bath. (Defense News, 1993; Journal of Commerce, 1993) 48. "ARPA Technology Reinvestment Project, section 2.2.2. 'Guidelines for Assembling a Team of Eligible Participants.' In general, an 'eligible firm' as defined by legislation is a company or other business entity that conducts a significant level of its research, development, engineering, and manu- facturing activities in the United States. A firm not meeting this test may still be an 'eligible firm' if its majority ownership or control is by United States citizens. In addition, a foreign-owned firm may be an 'eligible firm' if its parent company is incorporated in a country whose government encourages the participation of U.S.-owned firms in research and development consortia to which that govern- ment provides funding, if that government also affords adequate and effective protection for the intellectual property rights of companies incorporated in the United States. Determination of eligibil- ity of firms in this last category [foreign-owned firms] will be made by the Secretary of Commerce as

PUBLICLY FUNDED R&D 137 mandated by 10 U.S.C. 2491(9). No prior certification of eligibility will be issued or accepted, and the burden of establishing eligibility will ultimately rest on the proposer." (Advanced Research Projects Agency, 1993) 49. William Spencer, executive director of SEMATECH, announced on October 5, 1994, that SEMATECH would no longer seek a DOD subsidy. (Corcoran, 1994) 50. See Committee on Science, Engineering, and Public Policy (1992); Issues in Science and Technology (1994); testimony by S. Richard Deininger, Director of SEMATECH's National Re- source Program, before the NAE study committee, November 9, 1993; and Grindley et al. (1994). 51. The issue of foreign participation in the U.S. Display Consortium, like in SEMATECH, is primarily a political one. Both consortia represent a public-private response to an onslaught of for- eign competition. As such, it is politically very difficult to open them up to foreign participation, even though there may be some technological advantages to doing so. Note that the Microelectronics and Computer Technology Corporation, a privately funded consortium, is trying to expand its funding and technology base by recruiting foreign-owned firms. Proctor Reid, NAE, phone discussion with Peter Mills, CEO, U.S. Display Consortium, February 1994. 52. For data on sources of funding for foreign doctoral candidates in science and engineering, see National Science Foundation, 1993a, Table A-21, pp. 94-96. 53. These are the very subgroups targeted by the National Research Council (1988) study as the greatest potential source of U.S.-born engineers in the coming decades. 54. See "Foreign Nationals Change the Face of U.S. Science," Science (261), 1993, pp. 1769- 1775. 55. See note 6, above. 56. To be sure, contributions by foreign alumni to U.S. universities are a small fraction of those by U.S.-born alumni. Considering the countries of origin of the majority of foreign-born engineering students and faculty the People's Republic of China, Korea, and India this should come as no surprise. Nevertheless, there are numerous examples of foreign alumni generosity. For example, in gratitude for kindness shown by a professor in the mid-1950s, Gordon Wu of Hong Kong has en- dowed a $1.5 million Sollenberger chair at Princeton University, having earlier financed construction of a Gordon Wu Hall at the university. The Mogami Geotechnical Laboratory at Cornell University has been endowed by a former student and Japanese national who now teaches at Kyoto University (Stalson, 1989). 57. See the comments of Dr. Gerard Mourou, director of the National Science Foundation's Cen- ter for Ultrafast Optical Science at the University of Michigan, on the contributions of two large foreign sponsors, Fujitsu and Thomson, to the center's basic and long-term research program (U.S. Congress, House, 1993, pp.l02-103); testimony of John Wiley, University of Wisconsin, before the NAE study committee, November 9, 1993 (Wiley, 1993); and the comments of federal laboratory representatives in U.S. General Accounting Office (1988b) and National Research Council (1994a). 58. Of the 17 foreign-owned firms involved in industrial affiliate programs at the Georgia Institute of Technology, 16 are known to have U.S.-based manufacturing and R&D operations (Correspon- dence from Robert Nerem, Georgia Institute of Technology, to Proctor Reid, NAE, January 1995). Of 102 foreign-owned members of Stanford University's industrial liaison programs, 34 have U.S.-based manufacturing and/or R&D facilities (Correspondence from Marianne Meredith, Stanford, to Proctor Reid, NAE, April 7, 1995). 59. In at least two instances, foreign companies have replicated technologies developed by U.S. federal laboratories, thereby denying U.S.-owned firms opportunities to exploit first-mover advan- tages. The two examples cited most often in this context include the KEVA code (developed by Los Alamos National Laboratory), used to model the workings of the internal combustion engine, and the laminar-flow clean room concept (developed by Sandia National Laboratory), used in microelectron- ics manufacturing. (National Research Council, 1994a) 60. See, for example, p. 107 above, Wiley (1993), and U.S. Congress, House (1993). 61. Some observers have questioned whether most university technology licensing offices are

138 FOREIGN PARTICIPATION lN U.S. RESEARCH AND DEVELOPMENT adequately funded and staffed to negotiate effectively with corporate partners U.S. or foreign in the best interest of the United States. In testimony before the House Subcommittee on Science on October 28, 1993, Robin Frank Risser, CEO of Picometrix Inc., made the following observation regarding his company's participation in the University of Michigan's Center for Ultra-Fast Optical Science: Our biggest area of concern surrounds background technology which might have been available for exclusive licensing by Fujitsu had USL and Picometr~x not paid particular attention to this area. I believe this to be a problem area. University researchers often operate in an environment where there are significant bamers to constructively reducing inventions to practice by filing patent appli- cations. Patent budgets at the University of Michigan are small compared to the amount of research. As a result, it is usually difficult to patent many background discoveries prior to entering into a research contract with a foreign company. Since foreign companies usually obtain options for exclu- sive licenses on patents "conceived or reduced to practice" during the term of the agreement, this means that they could reach back to obtain exclusive rights to background technology that they did not really fund. In my opinion, this occurrence could be substantially reduced if patent budgets were substantially increased, faculty were better educated, and other battlers to patenting were reduced. Perhaps requiring that a percentage of federally funded research and development be set aside for foreign and domestic patents is appropnate. (U.S. Congress, House, 1993, pp. 60-61) 62. See, for example, Stalson (1989), Massachusetts Institute of Technology (1991), U.S. House of Representatives (1993). 63. See discussion on pp. 114- 117 and notes 36 and 40, above. 64. Although U.S.-industry contacts with foreign universities are widespread, actual U.S. corpo- rate funding of foreign university research is small compared to industrial R&D spending. Compa- nies supported foreign university research at about the same rate that they support R&D in U.S. universities (i.e., roughly 1 percent of total company-funded R&D, or, by NSF's (199Ob) estimates, $50 million to $100 million per year). The U.S. companies surveyed indicated that the principal rationale for establishing linkages with foreign universities was to benefit from the work of world-renowned scientists and engineers. Other reasons given included fostering good will and promoting the company's reputation in the host country's scientific community. 65. See, for example, the program announcement for NSF's (199Sc) Division of International Programs as well as the U.S.-Japan Manufacturing Fellowship Program administered by the U.S. Department of Commerce, which was launched during the final year of the Bush Administration. U.S. Department of Commerce (1993b). 66. Correspondence from Helen Donoghue, Second Secretary, Science and Technology, Delega- tion of the European Communities, Washington, D.C., to Proctor Reid, NAE, May 18, 1994. 67. Ibid. 68. Very little information is available on the specific requirements individual European nations impose on would-be foreign participants in their national R&D programs. For information regarding the policies of several OECD countries, see Organization for Economic Cooperation and Develop- ment (1992b). In late 1993 and early 1994, the U.S. Department of State conducted a survey of U.S. access to publicly funded research abroad as part of an ad hoc interagency working group chaired by the Council of Economic Advisors. The findings of the survey confirm that the formal policies of individual EC member states with respect to international access to national R&D programs are, for the most part, consistent with those of the European Community proper. However, some survey re- spondents indicated that, in some cases, national practices vis-a-vis foreign-owned firms were more restrictive than their formal policies would suggest. Proctor Reid, NAE, discussions with Anthony Rock and Gary Couey, U.S. Department of State, March 14, 1995. 69. The High-Speed Civil Transport "Piper" project was designed from the outset to include for- eign collaborators (National Research Council, 1994b). 70. See note 42, above. 71. The rationale for extending national treatment to multinational firms is essentially the same as that underlying the U.S. commitment to free trade and free investment: They offer the most effective

PUBLICLY FUNDED R&D 139 route to economic growth and rising living standards for all nations. Moreover, the United States remains the world's largest source of foreign direct investment. U.S.-owned multinationals account for a majority of U.S. exports and depend increasingly on foreign markets. The U.S. commitment to the principle of national treatment is manifested in bilateral and multilateral investment treaties it has negotiated with other countries, and the federal government's official negotiating position within the Organization for Economic Cooperation and Development (OECD), General Agreement on Tariffs and Trade (GATT), and most recently, the Asian-Pacific Economic Cooperation Forum. The United States and its trading partners have negotiated exceptions to national treatment in areas concerning national security and public order. Investment treaties and OECD code have also listed specific excluded sectors; these vary somewhat from one agreement/ treaty to another. Despite these inconsistencies and exceptions, the U.S. commitment to national treatment has been strong and has clearly contributed to a movement in this direction at the interna- tional level. Federal R&D programs that condition national treatment on foreign-government (non-for- eign-owned firm) compliance with/or application of U.S. laws and standards undermine U.S. efforts to push national treatment in international forums. As recent developments in the EC suggest, U.S. moves toward reciprocity in this and other areas of so-called domestic policy may encourage other nations to introduce similar requirements. The consequences of this for the United States in the long term, if not more immediately, could be costly (Graham and Krugman, 1995). 72. While there has been legislation put forward in the European Parliament that would introduce similar reciprocity "conditional national treatment" requirements in EC programs, it has not yet be- come law. See European Parliament (1994).

Next: Findings and Recommendations »
Foreign Participation in U.S. Research and Development: Asset or Liability? Get This Book
×
Buy Paperback | $41.00 Buy Ebook | $32.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

During the past decade, foreign participation in U.S. research and development—through acquisition of R&D-intensive businesses, links with universities, and other arrangements—has expanded rapidly.

This emergence of foreign influence has drawn a mixed response—some regard the trend as a positive corollary to the expanding involvement of U.S.-owned companies in national markets abroad. Others consider it a net liability for Americans that often benefits foreign companies and their home economies at U.S. expense.

There exists a large gap in expert and public understanding of the drivers, nature, and consequences of foreign participation in the nation's technology enterprise. This volume seeks to close this gap and reviews:

  • The nature of R&D activities and how they contribute to economic development.
  • The causes, scope, and nature of foreign involvement in U.S.-based R&D activity and the associated costs, risks, benefits, and opportunities of this trend.
  • The merits and liabilities of policies to regulate foreign R&D participation.
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!