The Changing Place of Japan in the Global Scientific and Technological Enterprise1
DAVID C. MOWERY AND DAVID J. TEECE
INTRODUCTION
During the past 20 years, Japan's role within the global network of public and private institutions that influence the creation and adoption of new technologies has changed. Along with dramatic structural change in the Japanese economy, this period has witnessed a transformation in Japanese technological capabilities. From its position in the 1960s as an economy that relied extensively on the transfer and modification of externally developed technologies, Japan has emerged as an economy with many firms that define the technological frontier in their industries.
This transformation in Japanese technological capabilities has created complex challenges for U.S. managers and public officials. The rapidly expanding network of "strategic alliances" among U.S. and Japanese firms has fueled concern that the pursuit by U.S. managers of corporate interests may not coincide with the advancement of U.S. national economic interests. The contrasting structure of the U.S. and Japanese "national innovation systems" means that U.S. firms may have more difficulty accessing Japanese scientific and technological research than Japanese firms have accessing
U.S. research activities and results.2 Added to these and other challenges for public policy is that of assessing the "lessons," if any, of postwar Japanese government industrial and technology policies for a U.S. economy that has produced minimal growth in overall productivity or median household incomes during the past two decades.
Our survey of these issues begins with a brief overview of Japan's emergence as a producer of "state-of-the-art" technology during the past 20 years. We then describe U.S. firms' evolving links to Japanese technology development through alliances, R&D activities conducted in Japan, and other means. The next section discusses some of the complexities for corporate managers and public policy created by these links and by other efforts by policymakers and scholars to divine the "lessons" of Japanese policy.
JAPAN'S CHANGING TECHNOLOGICAL POSITION, 1960-1990
A number of indicators suggest that Japanese firms have progressed during the postwar period from borrowing, modifying, and successfully commercializing foreign technologies to operating at the technological frontier. No individual indicator is definitive, but together they are suggestive. Okimoto and Saxonhouse noted that the ratio of Japanese exports of technology to imports increased from 0.12 in 1971 to 0.30 in 1983.3 This indicator reflects historical trends in licensing and therefore includes royalty payments for technologies licensed years ago. The ratio of technology exports to imports in new contracts alone, however, has risen from 1 in 1972 to 1.76 in 1984—in other words, Japanese firms are now net exporters of intellectual property as measured in new contracts.4 The average annual rate of
growth in Japanese payments for imports of technology declined from 31 percent in the 1950s to 6 percent in the late 1970s, according to Uekusa.5
Alone among industrial nations, Japan registered an increase, rather than a decrease, in the number of patents received per scientist and engineer during 1967–1984, as well as an increase in the number of patents received in foreign countries. Pavitt and Patel point out that on a per capita (adjusted by home-country population) basis, Japanese patenting in the United States increased by more than 650 percent during the period from 1963–1968 to 1980–1985, far more than any other industrial economy.6 The U.S. National Science Foundation (NSF) reported in 1988 that Japanese firms accounted for the largest single share of foreign-origin U.S. patents.7 A comparison of U.S. patenting by U.S. and Japanese firms in 1975 and 1986 reveals dramatic improvements in Japanese patenting performance, relative to that of U.S. firms, in selected technologies (Table 1). Moreover, according to the NSF report, Japanese-origin U.S. patents were cited more than proportionately in other patent applications, an indicator of the high quality of the Japanese patents.8 Japan outstripped the combined totals of Germany, France, and England in patents granted in the United States in 1987 (Table 2).
Other indicators of technological performance also suggest considerable Japanese strength. Technology adoption is a critical factor in national competitive performance, and in some technologies Japanese adoption performance appears to exceed that of the U.S. economy. The rate of adoption and intensity of utilization of advanced manufacturing technologies (including robotics, computer-integrated-manufacturing workcells, and flexible manufacturing systems) in Japanese manufacturing both considerably exceed
TABLE 1 U.S. and Japanese Shares of Total Patents Granted in the United States for Selected Technologies: 1975 and 1986 (percentage)
Selected Technologies |
United States |
Japan |
||
1975 |
1986 |
1975 |
1986 |
|
Lasers |
63 |
50 |
14 |
35 |
Telecommunications |
66 |
52 |
14 |
26 |
Steel and iron |
48 |
37 |
18 |
29 |
Internal combustion engines |
54 |
28 |
17 |
44 |
Semiconductor devices and manufacture |
68 |
57 |
13 |
29 |
Jet engines |
66 |
60 |
4 |
9 |
General-purpose programmable digital computer systems |
77 |
69 |
5 |
19 |
Robots |
63 |
50 |
20 |
29 |
Machine tools—metalworking |
65 |
51 |
8 |
17 |
All technologies |
65 |
54 |
9 |
19 |
SOURCE: Office of Technology Assessment and Forecast, U.S. Patent and Trademark Office, unpublished data. National Science Foundation, The Science and Technology Resources of Japan: A Comparison with the United States, NSF report 88-318 (Washington, D.C.: 1988). |
their levels in U.S. manufacturing.9 Other research has documented the ability of Japanese automotive firms to bring new models to market more rapidly than U.S. auto firms, and Mansfield's research indicates that for technologies based on sources of knowledge outside of the firm, Japanese firms exhibit significantly shorter development and commercialization cycles.10 Mansfield's research also indicates that Japanese firms devote roughly
TABLE 2 Applications (Registrations) in Selected Countries, by Nationality, 1989
twice as large a share of their R&D budget to process research as U.S. firms. Interestingly, Mansfield's econometric analyses suggest that the returns to R&D investments in Japanese industry substantially exceed those in U.S. industry.11
This remarkable record of technological achievement has rested largely on R&D funding from private industry, rather than from government. The
share of national R&D that is privately financed is higher in Japan than in the United States or other industrial countries. The share of national R&D expenditures financed by the Japanese government was only 20 percent in 1987, and this included grants-in-aid to national and private universities. The comparable figure for the United States in 1987 was 49 percent. The large military R&D budget in the United States accounts for much of this difference, but the Japanese share of R&D financed by the private sector is larger than that for the United States even if we consider only nondefense R&D.12 When military R&D expenditures are removed from the figures for both countries, the share of gross national product (GNP) devoted to R&D in Japan has exceeded the American figure since the early 1960s.
The U.S.-Japan contrasts in the direct role of government R&D funding are linked to other important differences in the structure of the U.S. and Japanese "national research systems" (a term meant to include both public and private research organizations). Along with a relatively open market for imports and foreign investment (albeit one that has faced significant demands for protection in recent years), the United States maintains a relatively open research system. More than 50 percent of the basic research performed within the U.S. economy is carried out within universities (including federally funded R&D centers, or FFRDCs). The structural characteristics of the U.S. R&D system, with its high mobility of engineers, scientists and entrepreneurs among firms, heavy reliance on university research for basic science and training, and on small firms for technology commercialization, mean that access by foreign firms to U.S. scientific and technological advances is relatively easy.13
The relative importance within the Japanese R&D system of "open" and "closed" research institutions, respectively universities and private firms, contrasts with that of the United States. Comparative statistics suggesting that Japanese universities account for a larger share of that nation's total R&D investment overstate the performance of R&D within Japanese universities; moreover, the contribution of Japanese universities to basic knowledge historically has been modest in many areas. The Japanese economy's system of industrial finance and governance also makes it difficult for U.S. and other foreign firms to gain access to industrial technologies or innovations through acquisitions of firms or intellectual property.
The structure of both the Japanese and U.S. R&D systems is chang-
ing,14 but in the near future, the U.S. system is likely to remain unparalleled as a performer of basic research, even as the Japanese R&D system strengthens its already impressive capabilities in the creation, modification, and adoption of technology. Among other things, the deeply rooted nature of differences in the structure of the Japanese and U.S. R&D systems means that changes in public policy will work slowly and incompletely to remove structural impediments to access.
U.S. FIRMS' GROWING LINKS WITH THE JAPANESE R&D SYSTEM
The rise of Japan as a technological powerhouse has been paralleled by growth in the importance of the Japanese market for consumer and industrial goods. Foreign firms have pursued a number of approaches to improving their access to Japanese technology and markets, including the establishment of R&D facilities in Japan and the development of complex "strategic alliances" with Japanese firms. Remarkably, in view of the urgency and importance of this issue for U.S. public policy and private managers, federal agencies provide only the most rudimentary data on U.S. and foreign R&D investment in Japan or on the growth of U.S.-Japanese interfirm alliances. In this section, we review the development of both of these approaches to access, relying on fragmentary data from the Japanese Ministry of International Trade and Industry (MITI) and other sources.15 One of the most important implications of the growth of Japanese technological capabilities is the need for better data on the nature of private and public relationships between U.S. and Japanese entities in technology development, transfer, and exchange. All too often, U.S. policy toward Japanese industrial R&D is formulated in a factual vacuum.
Foreign Patenting and R&D Activity in Japan
Patent activity by foreign firms in postwar Japan indicates that foreign firms have had a long-standing interest in utilizing their technology in Japan. Though the data we have found are extremely sketchy and relate only to patent applications, they indicate that foreign firms accounted for 23.2 percent of the total of all patent applications in Japan in 1970, although by 1978 this had fallen to 14.8 percent.16 Chemicals was the most active of these sectors in foreign patenting within Japan during the 1970–1978 period. The United States accounted for 42.7 percent of the total number of foreign patent applications (30,089) in 1989. Germany was the second most significant applicant in 1989, with 19.4 percent of the total (see Figure 1). Table 2 indicates, however, that foreign firms have been less active in patenting in Japan than they have in the United States, Germany, France, or England.
Of perhaps greater interest are the R&D activities of foreign firms in Japan.17Table 3 contains data from a MITI survey of growth in foreign R&D laboratories in Japan during 1975–1990 in manufacturing, which account for the vast majority of foreign R&D activity.18 The number of foreign-owned Japanese R&D facilities in manufacturing grew rapidly during this period, from 51 in 1975 to 123 in 1990, by 1990, this study suggests, total foreign R&D expenditures within Japan amounted to more than 200 billion yen, more than $1.5 billion. Interestingly, despite widespread perceptions of Japanese technological strengths in electronics and manufacturing process technologies, Table 3 suggests that the chemical and pharmaceutical industries account for the largest number of foreign-owned R&D facilities in both 1975 and 1990 and the largest share of foreign firms' investment in 1990. A substantial portion of foreign pharmaceutical firms' R&D investment almost certainly is linked to their efforts to obtain regulatory approval for the introduction of new drugs into the Japanese market. Although foreign clinical trial data are increasingly acceptable to Japanese regulatory authorities, the use of domestic medical personnel and researchers for such trials remains a more effective strategy for gaining regulatory approval. Much of foreign pharmaceutical firms' R&D investment, therefore, appears to be linked to growth in the Japanese consumer market.
A separate tabulation compiled as part of the MITI study contains data on the number and expenditure of Japanese R&D facilities owned by U.S. and non-U.S. firms in 1990 (including nonmanufacturing industries). According to this analysis, U.S. firms operated 83 R&D laboratories at a cost of nearly 83.5 billion yen in 1990. European firms accounted for 58 laboratories but invested a larger amount, more than 122 billion yen. The apparent differences in the average size of U.S. and European R&D laboratories in Japan are not explained in the study, but may reflect differences in the industrial composition of U.S.-and European-owned R&D laboratories within Japan. The European R&D presence within Japan may contain a larger share of chemical and pharmaceutical firms, which typically operate large laboratories, than the U.S. industrial R&D investment, which contains a higher share of software and electronics firms. In the absence of additional data, however, any such conclusion is purely speculative.
These data on foreign firms' R&D investment in Japan differ somewhat from the portrait of foreign firms' innovative activities in Japan presented
TABLE 3 Foreign R&D Facilities in Japan, 1975 and 1990; Foreign Firms' R&D Expenditures in Japan, 1990
|
No. of Labs 1975 |
No. of Labs 1990 |
R&D Expenditure 1990 (billion yen) |
R&D Expenditure 1990 (million dollars) |
Chemicals |
26 |
59 |
25.9 |
199.23 |
Pharmaceuticals |
12 |
25 |
127.3 |
979.23 |
Electric Machinery |
3 |
11 |
15.5 |
119.23 |
Nonelectric Machinery |
1 |
5 |
8.6 |
66.15 |
Instruments |
1 |
5 |
0.6 |
4.61 |
Transportation equipment |
1 |
4 |
3.1 |
23.85 |
Petroleum |
3 |
3 |
14.9 |
114.62 |
Food |
1 |
2 |
1.1 |
8.46 |
Paper/pulp |
2 |
2 |
0.3 |
2.31 |
Ceramics |
0 |
2 |
1.0 |
7.69 |
Nonferrous metals |
1 |
2 |
2.2 |
16.92 |
Rubber products |
0 |
1 |
0.01 |
0.08 |
Total manufacturing |
51 |
123 |
202.9 |
1,560.76 |
Retail services |
4 |
15 |
2.9 |
22.31 |
Other services |
0 |
5 |
4.0 |
30.77 |
Petroleum extraction |
4 |
4 |
15.2 |
116.92 |
Total number of foreign R&D labs in Japan |
59 |
137 |
211.4 |
1,626.15 |
SOURCE: Ministry of International Trade and Industry, Results of the 24th Survey of the Trend of Foreign Affiliates in Japan (Tokyo, 1990). |
by Cantwell.19 Cantwell's data are based on U.S. patents received by large U.S. and foreign firms' Japanese research facilities, and suggest that the contribution of foreign-owned R&D in Japan during 1978–1986 was greatest in motor vehicles, scientific and measurement instruments, and electrical equipment (including semiconductors). The modest contribution of foreign-owned pharmaceuticals research in Japan to these firms' U.S. patents is consistent with our earlier discussion of this R&D investment in Japan, much of which is concerned with clinical testing and approval of new drugs for the Japanese domestic market.
The great contribution of foreign-owned R&D to motor vehicles patenting in Japan, however, does not appear to be matched by high levels of foreign R&D investment or large numbers of foreign-owned facilities in
this industry within Japan. There are at least two possible explanations for this disparity. Cantwell's patent data are classified by the field of application of the patent, while the data presented above are classified by the "primary industry" of the investing firm. Cantwell's data thus treat a patent received on a ceramic engine part by a foreign-owned chemicals R&D laboratory in Japan as a "motor vehicles" patent, while the data in Table 3 classify this laboratory and its associated costs as a chemicals industry R&D facility. Alternatively, this disparity may reflect a failure by foreign motor vehicles firms to reward the significant technological contributions of their Japanese R&D facilities with higher levels of investment.
A recent contractor report to the National Science Foundation provides additional information (some of which appears to be inconsistent with the MITI data) on U.S. firms' R&D presence within Japan.20 The NSF study found 71 U.S. firms with R&D operations in Japan, and received responses from 36 of these to a detailed questionnaire. The authors of the study note that their count of U.S.-owned R&D laboratories yielded an unexpectedly low total, a smaller number than was tabulated in the MITI study discussed above.21 As in the MITI survey, the NSF study found that more than half of the firms responding to the survey were from the pharmaceuticals, chemicals, or petroleum industries, rather than the electronics sector.22 The NSF study also obtained information on staffing patterns and motives for the establishment by U.S. firms of Japanese R&D facilities.
In general, despite the evidence cited above concerning the importance of Japan as a source of new industrial technology, the NSF study suggests that most U.S. firms continue to use Japanese R&D laboratories as instruments for improving or maintaining access to Japanese markets. This conclusion, which
must be qualified by an acknowledgment of the low response rate to the NSF team's questionnaire, is based on the respondents' characterization of their motives for establishing a Japanese R&D laboratory:
Objectives [in establishing a Japanese R&D lab] considered "very important" or "important" were: developing products for the Japanese market; to improve the quality and consumer acceptance of their products (by utilizing Japanese manufacturing technology); and entry into the Japanese R&D scene. Objectives considered ''least important" were to establish a research base in the Far East; to increase the effectiveness of absorbing technology generated in Japan; and to qualify for Japanese government grants and loans for industrial R&D.23
U.S. firms also staffed their Japanese R&D facilities largely with Japanese scientists and engineers, rather than rotating personnel from other research facilities through their Japanese research facilities. This policy may make it more difficult to transfer Japanese-developed technologies to their U.S. or global operations.24
In other words, U.S. firms appear to be utilizing their Japanese R&D facilities to modify products for Japanese consumers and thereby improve their access to the growing Japanese market,25 instead of using their Japanese R&D operations as part of a global technology development strategy. If the NSF study accurately characterizes U.S. firms' motives and R&D operations in Japan, it suggests that many U.S. managers have yet to modify their strategies to take into account the role of Japan as an important source of new technologies. This apparent lack of awareness may also be responsible for the relatively modest presence of U.S. electronics firms in Japan.
Japanese R&D Activity in the United States
Do Japanese firms adopt a different approach to their U.S. R&D? A survey of 100 Japanese subsidiaries in the United States, 10 of them acquired, showed that 45 percent conducted in-house R&D.26 The managers surveyed frequently claimed that "the primary R&D objective of these subsidiary plants with in-house R&D was to expand present business and support present business," apparently through new product development. These results tell us little, however, about how these objectives are pursued. Respondents to the survey indicated that technology transferred from their Japanese parent company was overwhelmingly the most important source of technology. Little basic research is conducted in the United States by Japanese subsidiaries, although there are exceptions (e.g., NEC). Instead many Japanese firms use their U.S. research facilities to adapt technology to the U.S. market, in much the same fashion as U.S. firms historically have used their Japanese R&D facilities. In addition, of course, Japanese firms utilize their U.S. R&D facilities to monitor and assimilate U.S. technological and scientific advances.27
Sectoral differences are considerable. Japanese firms have established at least 70 electronics R&D facilities in the United States, according to data collected by Genther and Dalton and reproduced here as Table 4.28 While many of these labs may be modifying Japanese technology for the U.S. market, there is little doubt that in the area of software development, Japanese firms are shoring up their historic weaknesses by hiring U.S. talent.
"Strategic Alliances" Between Japanese and Foreign Firms
Another important channel for foreign access to Japanese markets and technologies is through long-term agreements among firms that cover joint activities in R&D, product development, manufacturing, or marketing. Such "alliances" between U.S. and Japanese, U.S. and European, and Japanese and European firms have grown significantly in number during the past 20
TABLE 4 Major R&D Facilities of Japanese Electronics Companies in the United States
Company |
Location of Facility |
R&D Activities in Electronics |
Computers and Peripherals |
||
Asahi Optical |
Pentax Technology Broomfield, CO |
Scanners, laser printers (1985) |
Epson |
Epson Technology Center Santa Clara, CA |
Personal computers (1988) |
Fujitsu |
Intellistor Inc. Longmont, CO |
Disk storage devices (1987) |
Hitachi |
Waltham, MA |
Workstation and Unix development (1989) |
Konica |
Konica High-Technology Laboratories Sunnyvale, CA |
Data Storage(1984) |
Matsushita |
Panasonic Technology Palo Alto, CA |
Computer document processing systems |
Mitsubishi |
Horizon Research Waltham, MA |
Data processing (1985) |
Mitsubishi |
Mitsubishi Electric Research Laboratory Cambridge, MA |
Superparallel computers (1991) |
Nakamichi |
Mountain View, CA |
Disk drives (1987) |
NEC |
NEC Technologies Systems Laboratory Boxboro, MA |
Workstations, laptops |
Oki Electric |
Advanced Technology Center |
Laser printers, fax (1990) |
Rohm |
Rohm Research Corp. Boulder, CO |
Printer heads (1990) |
Sony |
Intelligent Systems Research Laboratory San Jose, CA |
Workstations (1988) |
Sony |
Data Storage Laboratory Boulder, CO |
Erasable optical disks (1989) |
Toshiba |
Toshiba America Information Systems San Jose, CA |
Laptops, personal computers |
Computer Software |
||
Ascii |
Media Technology Research Institute San Francisco, CA |
Software and media (1990) |
Canon |
Costa Mesa, CA |
Software, data processing (1990) |
Fujitsu |
Fujitsu Systems San Diego, CA |
Software for POS, handheld computers, routing systems |
Fujitsu |
Information Systems San Jose, CA |
Engineering related software |
Fujitsu |
Open Systems Solutions Emeryville, CA |
Unix software (1991) |
Hitachi |
Hitachi Microsystems San Jose, CA |
Software engineering, design and engineering support |
Company |
Location of Facility |
R&D Activities in Electronics |
Hitachi |
Hitachi Software Engineering San Francisco, CA |
Software (1991) |
Hitachi |
Hitachi Digital Graphics Sunnyvale, CA |
CAD graphics, digitizers |
Kobe Steel |
Kobe Steel Research Laboratory Palo Alto, CA |
Magnetic memories, artificial intelligence (1990) |
Matsushita |
Information Technology Laboratory Princeton, NJ |
Computer graphics, document processing, software (1991) |
Matsushita |
Industrial Equipment Research Laboratory, Wood Dale, IL |
Software for POS (1987) |
NEC |
NEC Research Institute Princeton, NJ |
Al, parallel computing, machine learning (1990) |
Ricoh |
Ricoh Software Research Santa Clara, CA |
Software (1988) |
Seiko Instruments |
San Jose, CA |
Computer graphics |
Sony |
Sony Software Corp. New York. NY |
Music and multimedia software, prepackaged software |
Sony |
Sony Microsystems San Jose, CA |
Unix R&D (1990) |
Sumitomo |
Sumitomo Electric Santa Clara, CA |
Software development (1991) |
Zuken |
Santa Clara, CA |
Computer-aided design (1991) |
Zuken |
San Jose, CA |
Computer-aided design (1993) |
Semiconductors |
||
Canon |
Canon Research Center Palo Alto, CA |
Semiconductors (1990) |
Fujitsu |
Fujitsu Microelectronics San Jose, CA |
Custom gate array design, SPARC |
Fujitsu |
Fujitsu Microelectronics San Jose, CA |
Memories, logic and analog, ASIC |
Fujitsu |
Fujitsu Microelectronics Santa Clara, CA |
Microwave (MIC) and lightwave integrated circuits (MMIC) |
Hitachi |
Brisbane, CA |
Semiconductors (1989) |
Hitachi |
Farmington Hills, MI |
Semiconductors for autos (1989) |
Kawasaki Steel |
Silicon Valley, CA |
Semiconductors (1993) |
Kobe Steel |
Research Triangle Park, NC |
GaAs and superconductive ceramics (1989) |
Kobe Steel |
San Jose, CA |
Semiconductors |
Matsushita |
San Jose, CA |
Semiconductors and software (1991) |
Mitsubishi |
Durham, NC |
Semiconductors (1984) |
NEC |
NEC Electronics Natick, MA |
Semiconductors (ASICs) (1987) |
Company |
Location of Facility |
R&D Activities in Electronics |
NEC |
Mountain View, CA |
VLSI (1986) |
Oki Electric |
Oki Semiconductor Sunnyvale, CA |
Semiconductors (1989) |
Ricoh |
Ricoh California Research Center Menlo Park, CA |
ASICs, CMOS (1989) |
Rohm |
Rohm Research Corp. San Jose, CA |
Semiconductors (1990) |
Sharp |
Camas, WA |
Semiconductors (1988) |
Toshiba |
Toshiba Electronic Components Sunnyvale, CA |
Semiconductors (1984) |
Telecommunication Equipment |
||
Applied Telesis |
Seattle, WA |
Data communications equipment (1989) |
Fujitsu |
Fujitsu Network Systems Raleigh, NC |
Central office switching equipment (1987) |
Fujitsu |
Business Communication Systems Anaheim, CA |
PBX equipment |
Fujitsu |
Telecommunications Research Center Richardson, TX |
Telecommunications equipment |
Fijitsu |
Fujitsu Imaging Systems Danbury, CT |
Fax machines |
Hitachi |
Hitachi Telecom Norcroff, GA |
PBX, faxes (1987) |
Matsushita |
Applied Research Laboratory Burlington, NJ |
Video broadcasting (1981) |
Matsushita |
Communications Systems Laboratory Secaucus, NJ |
Digital cable TV systems |
Matsushita |
Research Triangle Park, NC |
Satellite communications (1991) |
NEC |
Advanced Switching Laboratory Irving, TX |
Central office switches (1989) |
Oki Electric |
Suwanee, GA |
Telecommunications (1989) |
Ricoh |
San Jose, CA |
Facsimile equipment (1979) |
Sony |
Sony Telecommunications Technology Center Paramus, NJ |
Telecommunications |
TDK |
Components Engineering Torrance, CA |
Microwave-related components |
Company |
Location of Facility |
R&D Activities in Electronics |
Toshiba |
Toshiba America Information Systems Irvine, CA |
PBX, cellular systems, fax (1985) |
Optoelectronics |
||
Fujitsu |
Microwave and Optoelectronics Division Santa Clara, CA |
Microwave and lightwave semiconductors |
Hoya Corp. |
San Jose, CA |
Optoelectronics (1989) |
Hoya Corp. |
Hoya Electronics San Jose, CA |
Optoelectronics (1986) |
Hoya Corp. |
Hoya Optics Fremont, CA |
Optical and laser glass (1973) |
Hoya Corp. |
Continuum Santa Clara, CA |
Pulse laser beams |
NTT |
Photonic Integration Research Columbus, OH |
Optoelectronics (1987) |
Olympus |
Torrance, CA |
Optical and electronic products |
Sumitomo |
Research Triangle Park, NC |
Fiber optics |
Television |
||
Hitachi |
Hitachi America Princeton, NJ |
HDTV (1991) |
Matsushita |
Advanced TV-Video Laboratory Burlington, NJ |
HDTV (1990) |
Sony |
Advanced Technology Center (AVIC) San Jose, CA |
HDTV (1989) |
Sony |
Sony Engineering and Manufacturing San Diego, CA |
TV components |
Toshiba |
Toshiba America Consumer Electronics Wayne, NJ |
HDTV receivers (1990) |
Semiconductor Materials and Equipment |
||
Hoya |
Hoya Micro Mask Sunnyvale, CA |
Photo masks |
Hoya |
Probe Technology Santa Clara, CA |
Probe cards |
Kyocera |
San Diego, CA |
Ceramics |
Nikon |
Nikon Precision Belmont, CA |
Wafer steppers applications lab (1990) |
ULVAC |
Fremont, CA |
Semiconductor equipment applications lab (1990) |
Company |
Location of Facility |
R&D Activities in Electronics |
Other Industries |
||
Canon |
Newport News, VA |
Copiers (1990) |
Fuji Xerox |
Palo Alto, CA |
Communications networks for workstations (1992) |
Matsushita |
Speech Technology Laboratory Santa Barbara, CA |
Speech recognition, information processing (1981) |
Matsushita |
Avionics Development Corp. Irvine, CA |
In-flight audio, video systems for passengers (1990) |
Nippon Columbia |
Atlanta, GA |
Multimedia R&D (1992) |
Nippon Denso |
Southfield, MI |
Electronics (1987) |
Sharp |
Hycom, Inc. Irvine, CA |
Flat panel displays (1989) |
Sony |
Sony Transcom Irvine, CA |
In-flight audio, video systems |
Toshiba |
Toshiba America Medical (MRI) San Francisco, CA |
CAT scanners, medical electronic equipment (1989) |
SOURCE: Donald H. Dalton, U.S. Department of Commerce, July 1992. |
years. Although international joint ventures have long been a mainstay of international business operations, the ''alliances" of the past two decades focus more intensively on technology-intensive activities and industries, and frequently are concerned with product development or manufacture for a global, rather than a local, market. Most such alliances involving private firms are motivated by one or more of the following three factors: access to foreign markets; access to foreign technologies; and access to low-cost capital.
In industries like telecommunications equipment or commercial aircraft, the long-standing importance of governments as either purchasers or sources of influence over purchasers has made international collaborative ventures an important means of improving market access. In the semiconductor industry, bilateral trade disputes and the resulting "managed trade" agreements calling for improved market access also appear to have contributed to an increase in collaborative activity.29 Political factors and market access restrictions, however, are not the only factors behind the recent growth in
U.S.-Japan joint ventures. The sheer complexities of transferring and accessing external technologies through licenses, along with the growing technological prowess of Japanese firms, also have played an important role in the growth of U.S.-Japanese collaboration in industrial technology development.30
Tables 5-7 are drawn from a 1987 report by the Japanese Ministry of International Trade and Industry on Japanese participation in international research joint ventures. The report appears to have employed a fairly narrow, legalistic definition of a joint venture, in view of the differences between its tabulation and those drawn from other sources. Nevertheless, the tables yield important information on the growing technological linkages between foreign and Japanese firms. Table 5 displays trends in joint venture formation during 1982–1987, and together with Table 6, yields several interesting insights. Table 5 shows the acceleration in the number of newly formed international research joint ventures, which increased from 7 in 1982 to 36 in 1987.31Table 5 also suggests a high concentration of international joint ventures in the chemicals industry, which accounts for almost one-fifth of the total number of ventures formed during this period. Chemicals ranks second only to electronics in the number of joint ventures formed during 1982–1987. The importance of chemicals (which in this table includes pharmaceuticals) as a focus of international collaboration between U.S. and Japanese firms appears broadly consistent with the prominent role of pharmaceuticals and chemicals in the Japanese R&D investments of foreign firms discussed above.
Table 6 contains information on the nationality of the foreign participants in the international joint ventures covered in the MITI study and categorizes the joint ventures by technology field. U.S. firms dominate both the "conventional" and the "advanced" technology fields, accounting for 49 and 85 percent of the international joint ventures in the two categories respectively. U.S. dominance in computers and communications, biotechnology, integrated circuits, and factory and office automation (all of which are included in the "advanced" technology category) is even more pronounced. Table 7 disaggregates the international joint ventures by research activity. These data are consistent with the findings of other studies that international joint ventures among private firms rarely focus on basic
30 |
See David J. Teece, "Transactions Cost Economics and the Multinational Enterprise: An Assessment," Journal of Economic Behavior and Organization , 7, 1986, pp. 21–45. |
31 |
Like many tabulations of trends in international joint venture activity, the MITI data in Tables 5-7 contain no information on terminated joint ventures. Since these undertakings are renowned for their high "mortality" rate, the MITI data may overstate somewhat the rate of growth in sustained collaborative activity. Any overstatement, however, almost certainly is more than offset by the effects of the MITI study's narrow definition of joint ventures. |
TABLE 5 Formation of New International Research Joint Ventures Involving Japanese Firms, by Year and Industry, 1982–1987
Venture |
1982 |
1983 |
1984 |
1985 |
1986 |
1987 |
Cumulative Total |
Total |
7 |
7 |
23 |
37 |
25 |
36 |
135 |
Manufacturing |
6 |
7 |
19 |
29 |
18 |
30 |
109 |
Food |
0 |
0 |
1 |
2 |
1 |
1 |
5 |
Textiles |
0 |
0 |
0 |
2 |
0 |
3 |
5 |
Chemicals |
1 |
1 |
7 |
7 |
4 |
4 |
24 |
Steel |
1 |
1 |
4 |
0 |
0 |
2 |
8 |
General machinery |
1 |
0 |
1 |
2 |
2 |
5 |
11 |
Electric machinery |
3 |
4 |
4 |
9 |
6 |
6 |
32 |
Heavy electric machinery |
3 |
1 |
0 |
4 |
3 |
0 |
11 |
Household appliances |
0 |
1 |
3 |
2 |
2 |
0 |
8 |
Communication/computer |
0 |
1 |
1 |
2 |
1 |
3 |
8 |
Other electric machinery |
0 |
1 |
0 |
1 |
0 |
3 |
5 |
Transport machinery |
0 |
0 |
1 |
5 |
3 |
1 |
10 |
Instruments |
0 |
1 |
1 |
1 |
0 |
3 |
6 |
Other manufacturing |
0 |
0 |
0 |
1 |
2 |
5 |
8 |
Nonmanufacturing |
1 |
0 |
4 |
8 |
7 |
6 |
26 |
Construction |
0 |
0 |
1 |
1 |
0 |
1 |
3 |
Communications |
0 |
0 |
1 |
2 |
1 |
0 |
4 |
Finance |
0 |
0 |
2 |
3 |
1 |
2 |
8 |
Utilities |
0 |
0 |
0 |
0 |
1 |
0 |
1 |
Other services |
1 |
0 |
0 |
2 |
4 |
2 |
9 |
SOURCE: Ministry of International Trade and Industry, Status Report on International Joint Research and Development of Japanese Private Enterprises (Tokyo: 1987). |
or fundamental research.32 Instead, consistent with the blend of technology access and market access motives that underpin many such undertakings, they are focused on product development and/or modification for global markets.33
A recent study by the U.S. Department of Commerce examined a much broader array of linkages—what might be thought of as strategic alliances—between U.S. and Japanese firms, focusing on a "snapshot" of U.S.-Japan corporate linkages in six high technology industries during 1989–1990 (Table 8). Joint ventures account for less than 40 percent of the number of interfirm collaborative relationships in all of these industries, and in most instances are less frequent than are marketing collaborations
32 |
David C. Mowery, op. cit. |
33 |
"Market-specified" R&D in Table 7 refers to incremental product modifications for new markets. |
TABLE 6 Technology Fields of International Research Joint Ventures Involving Japanese Firms, 1982–87, by Field of Technology and Nationality of Foreign Firm
TABLE 7 International Research Joint Ventures Involving Japanese Firms, 1982–87, by Type of Research and Technology Field
Venture/Type of Research |
Conventional Technology |
Advanced Technology |
Communications/ Computers |
Integrated Circuits |
Factory and Office Automation |
Medical |
Biotechnology |
New Materials |
Nuclear Power |
Basic research |
2 |
1 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
Applied research |
7 |
18 |
3 |
0 |
0 |
0 |
11 |
3 |
1 |
Product oriented |
23 |
38 |
9 |
10 |
3 |
3 |
7 |
4 |
2 |
Market specified |
28 |
23 |
6 |
7 |
6 |
1 |
2 |
1 |
0 |
Other |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
Unspecified |
4 |
3 |
0 |
0 |
2 |
0 |
1 |
0 |
0 |
Total |
65 |
83 |
18 |
17 |
11 |
4 |
22 |
8 |
3 |
SOURCE: Ministry of International Trade and Industry. Status Report on International Joint Research and Development of Japanese Private Enterprises (Tokyo: 1987). |
and agreements.34 The importance of the market access motive for many current interfirm alliances may be inferred from the substantial portion of collaborations in all of these industries that are focused on marketing or marketing and development of new products. Consistent with the MITI data on the research content of international joint ventures, very few of these U.S.-Japan joint ventures in Table 8 are concerned with research, as opposed to production and/or the development of products (an exception to this statement is the biotechnology industry).
Although the full impact of U.S.-Japanese collaboration on the competitiveness of U.S. firms will not be apparent for some time, the visible consequences of these collaborations thus far do not support the critical view of these ventures presented by Reich and Mankin.35 Technology transfer within these ventures is more modest in scope and less uniformly ''outbound" than some assessments assume. Just as U.S. industries vary in their trade balances in goods, the net inflow or outflow of technology through U.S.-Japan collaborations varies across industries. Requiring balance in technology transfer on an industry-by-industry basis makes no more sense than a requirement for such balance in goods trade. In a number of industries, including steel, automobiles, and portions of microelectronics, international collaboration can improve the international competitiveness of the U.S. participants. 36 In other industries, such as robotics, the competitiveness of U.S. systems engineering and software firms and the ability of large U.S. firms to offer a full line of factory automation hardware and software depend on access to foreign hardware through joint ventures and licensing. As we note below, however, the ability of U.S. firms to reap benefits from international collaboration ventures depends on the care with which these undertakings are organized and managed. In particular, U.S. firms participating in international joint ventures may need to strengthen their abilities to learn and absorb new technologies from their partners.
Many U.S.-Japan collaborative ventures involve the purchase by large Japanese firms of significant equity positions in small start-up firms. Do these foreign investments result in the export of critical technological assets that will strengthen Japanese competitors? Very little is known about the economic or technological importance of foreign acquisitions of U.S. high technology start-up firms. Although there are numerous uncertainties on
34 |
Unfortunately, the lack of data on the size of the ventures in Table 8 means that the only basis for comparison of the "importance" of different types of collaborative activity is their number. Adjusting these data for the size of individual collaborative undertakings might yield different conclusions regarding the relative importance of various types of ventures. |
35 |
See R. Reich and E. Mankin, "Joint Ventures with Japan Give Away Our Future," Harvard Business Review, March–April, 1986. |
36 |
For a recent example, see the discussion of the alliance between Ford Motor Company and Mazda, "The Partners." Business Week, February 10, 1992, pp. 102–107. |
TABLE 8 U.S.-Japan "Corporate Linkages," 1989–1990, in Selected High Technology Industries
U.S.-Japan Links 1989–1990 |
Aerospace |
Computers |
Software |
Semiconductors |
Semiconductor equipment |
Biotechnology |
Total |
Marketing |
19 |
36 |
38 |
33 |
11 |
22 |
159 |
Marketing/development |
0 |
2 |
3 |
1 |
1 |
0 |
7 |
Joint venture: |
15 |
18 |
27 |
34 |
10 |
28 |
132 |
R&D/product |
7 |
10 |
12 |
18 |
8 |
10 |
|
Production |
6 |
5 |
13 |
14 |
2 |
4 |
|
Research |
2 |
3 |
2 |
2 |
0 |
14 |
|
Licensing |
6 |
3 |
11 |
11 |
3 |
10 |
44 |
Technology exchange |
0 |
2 |
0 |
10 |
0 |
2 |
14 |
Merger and acquisition |
2 |
8 |
3 |
7 |
5 |
4 |
29 |
Direct investment |
0 |
12 |
8 |
2 |
5 |
5 |
32 |
Consortia |
1 |
1 |
0 |
0 |
0 |
1 |
3 |
Internal venture |
0 |
0 |
0 |
4 |
2 |
4 |
10 |
Production |
4 |
4 |
1 |
1 |
1 |
1 |
12 |
Other |
0 |
1 |
|
2 |
|
1 |
4 |
Total |
47 |
87 |
91 |
105 |
38 |
78 |
446 |
SOURCE: U.S. Department of Commerce, Japan Technology Program, The Role of Corporate Linkages in U.S.-Japan Technology Transfer: 1991 (Washington, D.C.: NTIS, 1991). |
this issue, the "leakage" of U.S. technology through such acquisitions may have only a slight economic impact. In many instances, start-up firms pursue international collaborative ventures because of their need for capital. Policies to reduce this supposed outflow of U.S. technology must address the availability of capital and/or the willingness of managers in established U.S. firms to support small start-up enterprises (overcoming resistance to technologies "not invented here"), rather than attempt to restrict collaboration.37 Smaller firms' "export" of technology through international collaborative ventures rarely means that opportunities for exploitation of these technologies are lost to U.S. firms—in most cases, the U.S. partner in such a collaborative venture does not change its management or its location, and protection from other U.S. firms of its intellectual property is not airtight. The critical agents for the diffusion of these technologies (managers and employees of the small firm) remain in the United States, where they move to other firms, present the results of their research to domestic audiences, and otherwise act to disseminate much of the technology domestically. Indeed, the basis for such domestic high technology concentrations as Silicon Valley in California, Route 128 in Massachusetts, and North Carolina's Research Triangle is the tendency for critical technological assets (mainly people and specialized suppliers of goods and services) to remain regionally concentrated. If the enormous interregional flows of capital of the past 30 years have not diluted and diminished these regional concentrations within the United States, it seems unlikely that international capital flows will do so.
CONCLUSION
U.S. policymakers have yet to address the implications of change in the technological relationship between the United States and Japan. These changes pose a fundamental challenge to much current thinking in the executive branch and Congress, which now all too often proposes that Japanese access to U.S. science and technology be limited. Any restrictions on commercial technology transfer from U.S. universities or firms to Japanese entities, however, could provoke reciprocal restrictions that would harm U.S. competitiveness.
Paradoxically, the U.S. economy appears to have much more to lose and much less to gain by restricting foreign access to U.S. research and technology than at any previous point in the postwar era. The end of U.S. technological hegemony has weakened this nation's control over technology vis-à-vis foreign firms or individuals seeking access to U.S. Technologies,
In the current environment, U.S. firms stand to gain from continued improvements in their access to foreign markets, investment opportunities, and technologies. Achieving these improvements will be hampered if efforts are made to deny access to the U.S. research system.
The actions of many U.S. firms suggest that managers are beginning to pursue strategies designed to improve their access to the Japanese technological research system. Nonetheless, the evidence discussed above indicates that too many U.S. firms still view their Japanese R&D operations as oriented largely toward the domestic Japanese market, and are not working to exploit and transfer technologies from Japan into their global R&D networks. Japan's importance as a source of industrial technology means that U.S. firms must do more to gain access. This will require the expansion and establishment of corporate R&D facilities within Japan, as well as efforts to more closely link these facilities to corporate technology development strategies.
Similarly, the view that joint ventures with Japanese firms in research or product development give away our future must be qualified by an awareness of the potential and actual benefits of well-managed joint ventures for U.S. firms. U.S. managers nonetheless must proceed carefully in cooperating with an actual or potential competitor, and manage their technological and other assets strategically. In most cases, this requires that one maintain or strengthen independent technological and other capabilities, such as manufacturing or knowledge of markets and user needs, improving or sustaining the value of one's contributions to the joint venture. Successful participation in joint ventures requires that senior managers understand their firms' technological and other capabilities and incorporate them into strategic planning. Strategies designed to learn from the joint venture partner must be actively pursued, for ultimately the distribution of the benefits and costs from joint ventures in high technology will swing on how well each party is able to learn from the other.
The growing web of U.S.-Japanese technological linkages among private firms has and will continue to complicate any efforts by one or the other government to restrict access to its domestic research system. The effectiveness of SEMATECH's restrictions on foreign participation, for example, may be undercut by collaborative relationships between U.S. participants in the consortium and such Japanese semiconductor producers as Hitachi (which is working with Texas Instruments on advanced memory chips) and Toshiba (working with Motorola on memory and microprocessor chips). The development of international collaborative ventures among corporations is likely to frustrate attempts by governments to pursue "technological mercantilism"—attempting to restrict outflows of technology by governments in the same way that seventeenth-century European governments restricted outflows of specie. Such mercantilist policies provide powerful
incentives for private firms to collaborate in R&D, marketing, and manufacture in order to improve their access to foreign markets.38
The growth of Japan's technological strengths has raised to high levels of the U.S. and Japanese governments (as well as increasing the level of conflict regarding it) the issue of access by foreign firms to the Japanese research system. This issue figured prominently, for example, in the 1988 negotiations over the renewal of the U.S.-Japanese agreement on scientific cooperation. The Japanese R&D system is difficult for foreign firms to penetrate for reasons that reflect the historic legacy of government policies, as well as differences in industry structure and in the structure of capital markets; these difficulties are not solely a result of current government policies. The complex origins of these structural differences in the organization of national R&D systems and in the ease with which foreigners can gain access to national R&D systems mean that government-to-government negotiations and agreements cannot address all of the causes and consequences of "asymmetrical access."
The structural differences between the U.S. and foreign research systems are such that a strict requirement of reciprocity in access to research facilities is either worthless or infeasible. Assurances by the Japanese government of complete access to Japanese universities, for example, may be of limited interest to U.S. firms, in view of the modest amount of world-class research performed by Japanese university researchers. A "results-oriented" reciprocity requirement that mandated that Japanese firms open their industrial research facilities to foreign researchers could impose a similar requirement on U.S. firms and is scarcely likely to elicit the support of U.S. firms.
Some evidence suggests that the structure of the U.S., Japanese, and Western European research systems may be converging somewhat. As and if the quality and amount of world-class research performed in Japanese universities and quasi-public "hybrid" institutions improve, for example, access to these facilities may become more attractive and important for informed U.S. and European firms. Reduction in the structural dissimilarities of these research systems could attenuate difficulties over reciprocal access, but any such process of institutional change and convergence is
likely to move so slowly that the issue of reciprocal access will remain very difficult for the foreseeable future. The serious impediments to U.S. acquisition of firms in other industrial economies, particularly Japan, are not exaggerated. They will continue to create serious tensions, exacerbating the effects of other structural differences in access to research projects and results, until they are reduced or circumvented.
The interdependent relationship between a scientifically strong U.S. research system and a technologically strong Japanese research system also raises complex issues of balancing national contributions and benefits to the global scientific and technological enterprise. The results of scientific research are increasingly mobile internationally and difficult to ''appropriate" by the discoverer, a characterization that applies less accurately to the results of technology-oriented research. As a result, the possibility exists that the U.S. research system produces global "public goods," which can be exploited by (among others) Japanese firms for private profit. This characterization of scientific and technological research is at best a caricature, and understates the difficulties and costs of transferring and absorbing either type of information, but it captures an important difference between two research systems such as those of Japan and the United States.
The Japanese government has proposed several international scientific research projects (e.g., the Human Frontiers Science Program), in part as a means of expanding its contribution to global scientific research. The HFSP project has progressed quite slowly, however, and Japan's nonfinancial contributions to its advance are likely to remain modest. Significant Japanese participation in international scientific research projects in any but a financial role is likely to be hampered in the near term by the weakness of Japan's basic research capabilities in many areas. Japan's proposed Intelligent Manufacturing Systems (IMS) project, however, focuses on an area (advanced manufacturing process technologies) in which Japanese firms are in a leadership position and to which they could make significant contributions.39 The IMS appears to contain considerable potential benefits for U.S. corporate participants.
Partly because U.S. government officials felt they had not been sufficiently consulted by the IMS project's Japanese sponsors, the U.S. government was initially reluctant to support the IMS proposal. In addition to their concerns over a perceived lack of advance consultation, some U.S. officials felt that U.S. firms would contribute more to the undertaking than they would receive, transferring U.S. technology to Japanese firms. This concern appears to be misplaced, for several reasons. It is based on an
outdated assessment of U.S. and Japanese technological strengths in manufacturing. This approach also attempts to substitute the technological judgments of U.S. policymakers for those of corporate managers. Finally, opposition to U.S. participation in the IMS may reinforce the already distressing tendencies of U.S. managers to ignore external sources of industrial technology. The ambivalent response of the U.S. government to this Japanese proposal for international collaboration on technology-oriented research is unfortunate, and suggests the need for a recognition by U.S. policymakers of Japan's technological capabilities and a more realistic appraisal of the costs and benefits of U.S. participation in international technology development programs.
Intellectual property rights is another area of U.S.-Japan tension and negotiations that may now assume a very different role in this bilateral economic relationship. During the past decade, U.S. pressure has led the government of Japan to improve the protection offered to foreign firms' intellectual property, and U.S. firms such as Texas Instruments have begun to reap significant royalty payments for such key patents as that covering the integrated circuit. Simultaneously, the executive branch and Congress have taken a number of steps to strengthen the domestic protection of intellectual property rights in the United States. As was noted above, however, Japan now is increasingly a technology exporter and is a major patentor in the United States. Stronger international and domestic intellectual property rights protection thus may raise the costs to some U.S. firms of access to the increasingly important flow of technology from Japanese sources.40
As the example of intellectual property suggests, the effectiveness and value of specific technology policy initiatives depend critically on the level of technological development within an economy, both absolutely and relative to other economies. The Japanese government policies that are asserted to have contributed to the technological transformation of that economy now have many admirers and advocates within the United States. Even as some U.S. observers recommend emulation of Japanese research policies and institutions, however, a search is under way within Japan for new institutions to support the indigenous basic research believed necessary to underpin commercial innovation. Japanese cooperative research policies, for example, historically supported the diffusion and utilization of technological and scientific knowledge that was derived from external sources, and supported Japanese firms' efforts to "catch up" with global technology leaders. Within Japan, however, cooperative research rarely served to advance the scientific or technological frontier, a purpose for which it is often
promoted in the United States. Moreover, uncritical imitation of this and other technology policies associated with the period of "catch-up" in the Japanese economy overlooks considerable evidence suggesting that Japanese policymakers are now considering policies, such as public funding of basic research, that have long been central features of the U.S. national research system.
Above all, it is important to recognize that the current complexities in U.S.-Japanese economic and technological relationships are a legacy of successful domestic and international policies. Japan's postwar rise to technological leadership is attributable in part to U.S. policies that assisted Japanese national security and economic reconstruction. U.S. and Japanese citizens alike should be proud of this remarkable accomplishment. Nonetheless, adjustment by policymakers and managers in both the United States and Japan to new technological realities will require fresh thinking on both sides of the Pacific. Failure to adjust to the new environment will result in missed opportunities and unnecessary friction.