Particularly in light of recent advances in information and communication technologies, international boundaries have become considerably less important in structuring the conduct of R&D and the use of research collaborations. Indicators of R&D globalization illustrate these R&D landscape changes for each of the R&D-performing sectors. Growth in international academic research collaboration is exhibited by the substantial increase in international co-authorship trends. (See chapter 6.) Extensive global growth in public-sector and industrial R&D activities is detailed below.
The rapid rise in international cooperation has spawned activities that now account more than 10 percent of government R&D expenditures in some countries. A significant share of these international efforts results from collaboration in scientific research involving extremely large "megascience" projects. Such developments reflect scientific and budgetary realities: Excellent science is not the domain of any single country, and many scientific problems involve major instrumentation and facility costs that appear much more affordable when cost-sharing arrangements are in place. Additionally, some scientific problems are so complex and geographically expansive that they simply require an international effort. As a result of these concerns and issues, an increasing number of S&T-related international agreements have been forged between the U.S. government and its foreign counterparts during the past decade.
International governmental collaboration in S&T and R&D activities appears to be a growing phenomenon. There are few sources of systematic information on government-to-government cooperative activities, however. A report by the U.S. General Accounting Office (GAO 1999) provides a snapshot of seven Federal agencies' international S&T agreements that were active during FY 1997. The GAO accounting is only for official, formal agreements and therefore provides a lower-bound estimate of the number of governmental global S&T collaborations. Most international cooperation is continuous and ongoing and takes place outside the framework of official, formal agreements. Nonetheless, the GAO study found that these seven agencies-DOE, NASA, NIH, NIST, the National Oceanographic and Atmospheric Administration (NOAA), NSF, and the Department of State-participated in 575 such agreements with 57 countries, 8 international organizations, and 10 groups of organizations or countries. Fifty-four of these agreements were broad-based bilateral arrangements between the U.S. government and governments of foreign countries-commonly referred to as "umbrella" or "framework" agreements. The remaining 521 agreements were bilateral agreements between research agencies and their counterparts in foreign governments and international organizations (381) or multilateral agreements (140) to conduct international cooperative research, provide technical support, or share data or equipment.
Generally, such agreements-which are indicative of government interest to cooperate internationally in R&D-have no associated budget authority. Nor is there a system in place to link international S&T agreements with actual spending on cooperative R&D. According to a study by the Rand Corporation, the U.S. government spent $3.3 billion on R&D projects involving international cooperation in FY 1995 (which may or may not have been associated with international S&T agreements) and an additional $1.5 billion on non-R&D activities associated with international S&T agreements (Wagner 1997).
Among the seven agencies that GAO reviewed, DOE participated in the largest number of official international S&T agreements (257, or 45 percent of the 575 total). (See text table 2-17.) This total included almost 100 multilateral agreements with the International Energy Agency (IEA), which represents the United States and 23 other countries with common scientific interests and priorities. NASA was second among the seven agencies in terms of participation in total international S&T agreements (127, including 15 multilateral agreements with the European Space Agency).
In addition to the 140 multilateral agreements, these seven agencies participated in bilateral S&T agreements with countries from almost every region of the world. In terms of the sheer numbers, U.S. agencies were most active in their participation with Japan (78): DOE and NASA reported the largest number of their bilateral S&T agreements with that country. After Japan, U.S. S&T agreements were most commonly reported with Russia (38), China (30), and Canada (25). DOE reported more agreements with Russia and China than did any other agency; NASA accounted for the largest number of agreements with Canada. The prevalence of DOE and NASA in these and other international S&T agreements reflects the megascience attributes associated with their missions. Of the other five agencies in the GAO report, only NIST reported more than five bilateral agreements with any single country (Japan and South Korea) in FY 1997. NIST also listed five agreements with Russia and three with Canada.
Overall, more than 90 percent of the international S&T agreements active in FY 1997 resulted in research projects or other research-related activities. In cases in which this activity did not occur, funding problems that developed after the agreements were signed or changes in research priorities generally were the reasons for their discontinuation.
International S&T collaboration can and does increasingly take place under less formal agreements, however. Consequently, these measures of formal agreements do not necessarily represent the level or intensity of R&D relationships or international collaboration between scientific communities in various countries.
International R&D collaboration is on the rise in the private sector as well-as is indicated by the rising number of formal cooperative agreements or alliances between firms, the growth of overseas R&D activities performed under contract and through subsidiaries, and an increase in the number of R&D laboratories located abroad (OECD 1998a). The expansion of international industrial R&D activity appears to be a response to the same competitive factors that foster domestic collaborations. Firms reach beyond their home borders as a way of addressing rising R&D costs and risks in product development, shortened product life cycles, increasing multidisciplinary complexity of technologies, and intense competition in domestic and global markets.
Industrial firms increasingly have used global research partnerships to strengthen their core competencies and expand into technology fields they consider critical for maintaining market share. In these partnerships, organizations can expand opportunities and share risks in emerging technologies and emerging markets. During the first half of the 1970s, strategic alliances were almost nonexistent, but they expanded rapidly late in the decade. For example, the number of newly made partnerships in the three core technologies-information technologies, biotechnology, and new materials-rose from about 10 alliances created in 1970 (Hagedoorn 1996) to about 90 in 1980. R&D-related international strategic technology alliances increased sharply throughout the industrialized world in the early 1980s and accelerated as the decade continued, reaching 580 such partnerships in 1989. In the early 1990s, the annual formation of newly established alliances at first tapered off from that reported in the 1980s and then rapidly increased to a peak of more than 800 new alliances formed in 1995. Since then, there has been a steady decrease in the number formed, to 564 in 1998-a total that nonetheless exceeds the number formed during any year prior to 1989. For the entire 1980--98 period, U.S., European and Japanese firms collectively entered into almost 9,000 strategic technology alliances. Most of these alliances were formed in the 1990s; most involved U.S. firms; and most were signed to foster R&D partnerships in just a few high-tech areas, notably information technologies and biotechnology. (See figure 2-36, text table 2-18, and appendix table 2-67.)
As the number of alliances has increased, the forms of cooperative activity have changed as well. The most prevalent modes of global industrial R&D cooperation in the 1970s were joint ventures and research corporations. In these arrangements, at least two companies share equity investments to form a separate and distinct company; profits and losses are shared according to the equity investment. In the second half of the 1980s and into the 1990s, joint nonequity R&D agreements became the most common form of partnership. Under such agreements, two or more companies organize joint R&D activities to reduce costs and minimize risk while they pursue similar innovations; participants share technologies but have no joint equity linkages (Hagedoorn 1990, 1996).
Between 1990 and 1998, more than 5,100 strategic technology alliances were formed, of which 2,700 were intraregional (that is, made between firms located within the broad regions of Europe, Japan, or the United States) and 2,400 were interregional (between firms located in separate regions). Of course, many of the more than 500 intra-European alliances are also multinational because they generally involve firms from more than one European country (in contrast with the numerous intra-American and much less numerous intra-Japanese firm partnerships in which all partners have the same national ownership). For the 1990--98 period, U.S. companies participated in 80 percent of known technology alliances, about half of which were between two or more U.S. firms and about half of which included a non-U.S. company. European companies participated in 42 percent and Japanese companies in 15 percent of the 5,100 alliances formed in the 1990s. (See text table 2-18).
Consistent with overseas R&D funding trends (detailed below), just a handful of European firms account for most of that region's alliances. Of the 4,700 European alliances reported during the entire 1980--98 period (a figure that includes double-counting of partnerships with two or more European firms), the most active participants were British firms (1,036 alliances), German firms (994), French firms (715) and Dutch firms (680). More than 100 alliances were also formed by companies with Italian (338), Swiss (267), Swedish (278), and Belgian (119) ownership. Additionally, a substantial number of the international technology partnerships involved firms located outside of these major regions. During the entire 1980--98 period, Canadian firms entered into 198 strategic technology alliances (mostly with U.S. companies), South Korean firms joined 119, Russian (and other former Soviet Union) firms joined 90, Chinese firms joined 86, Australian firms joined 63, Israeli firms joined 51, and Taiwanese firms joined 48.
Most intraregional and interregional alliances have been between firms sharing research and technology development in information technologies (IT) and biotechnology. These two technologies alone account for two-thirds of all alliances formed since 1990. The only other technologies for which firms consistently have entered into a substantial number of partnerships relate to advanced materials and non--biotechnology-based chemicals. (See appendix table 2-67.) Forty-four percent of the technology alliances formed worldwide since 1990 dealt with information technologies such as computer software and hardware, telecommunications, industrial automation, and microelectronics. Of the roughly 2,300 IT alliances formed during this period, most have been between U.S. companies (50 percent) or between European and U.S. firms (19 percent). Among the 1,100 strategic biotechnology alliances, the regional distribution has been more diverse, although U.S.-U.S. and U.S.-European interregional partnerships are more prevalent than any other (each type accounting for more than one-third of the biotechnology total). Consistent with R&D funding trends and indicative of known core strengths, U.S.-Japanese collaborations are more common in IT activities than in biotechnology.