Chapter 2. Government R&D

Organization of S&T Policy Making
Main Science Funding Agencies
Science and Technology Agency (STA)
ERATO Evaluation
RIKEN's Frontier Research
Ministry of International Trade and Industry
Atom Technology

In the 1990s, the government is the dynamic growth sector of R&D in Japan. The Japanese government increased its share of overall R&D support from 16 percent in 1990 to 20 percent in 1994 (table A-4). In 1995, the Japanese government provided two supplemental budgets for R&D that together added $3.5 billion[9] for funding competitive research in universities, improving university research facilities, computer networks, and large-scale R&D equipment and facilities.

The Basic Plan for S&T of 1996 suggests that the government allocate 17 trillion yen (equivalent to approximately $74 billion in constant dollar terms) to R&D from 1996 to 2000. For Japan to meet the suggested investment of $74 billion would require an average annual growth rate of around 10 percent in government R&D investments, far higher than past annual funding increases. Historically, Japanese government R&D expenditures have grown about 4 percent annually, from $4.8 billion in 1975 to $8.9 billion in 1990. However, Japan's government R&D investments accelerated somewhat in the more recent period between 1990 and 1995, at 5 percent annually. Additionally, in 1996 the Japanese government increased its R&D budget by 12.5 percent, reaching $12.2 billion.

Using the 1996 government budget as a base year, a linear growth in the government R&D investment required to allocate $74 billion would be an increase of approximately $1.5 billion each year from 1996 to 2000 (table 4). The suggested amount of $74 billion would represent a sizable increase (35 percent) over the amount spent in the previous 5 years, approximately $51 billion in constant dollars from 1991-95. The Cabinet-approved 1997 R&D budget represents a 6.8-percent increase over 1996. The rate of increases in the 1996 and 1997 budgets together effectively meet the required level of growth. If Japan continues the trends of the 1996 budget and the approved 1997 budget, the year 2000 budget, or shortly thereafter, could meet the doubling goal.

table 4

In contrast to the trend in Japan of increasing government support for R&D, the 10-year trend for government sponsored research in the United States has been a declining real budget. At the peak of public funding for research in 1987, the U.S. Government invested $57.9 billion constant 1987 dollars. In preliminary data for 1997, U.S. government investment decreased to $47 billion (figure 8). Over this same time period, the U.S. Government decreased its share of overall R&D support from 46.2 percent to 33.6 percent (table A-4).

figure 8

Although the U.S. Government spends more on R&D in absolute terms, relative to the size of its GDP, the Japanese government has outspent the U.S. Government in non-defense R&D for more than a decade (table A-3). In 1975, governments in both the United States and Japan were investing the equivalent of about 0.5 percent of their respective GDPs in nondefense R&D. Government funded civilian R&D as a percent of GDP declined both in Japan and in the United States throughout the 1980s, followed by an increase in support of nondefense R&D in the 1990s. However, from 1990-96, Japan had a higher annual rate of growth (5.9 percent) in its government budget for civilian research than the United States (1.5 percent). By 1996, Japanese government civilian R&D rose to 0.54 percent of GDP; that of the United States represented 0.40 percent of GDP (figure 9 and table A-3).

figure 9

The Japanese government spends the largest portion of its R&D budget—an estimated 51 percent—on the general objective of "advancement of knowledge," a general function which includes general university research funds (GUF), and should not be equated with basic research (figure 10). The U.S. Government funds the majority of its R&D budget on defense—55.3 percent in 1994—down from 67.5 percent in 1985. The second most important objective of government R&D funding in the United States is health research, approximately 17 percent.

figure 10

The emphasis of the Japanese government on various R&D objectives has been relatively stable throughout the past decade, except for an increase in energy, advancement of knowledge, and defense.

The relative share of agricultural research declined to compensate for these increases (table A-8). In this same time period, the U.S. Government increased its support of general science and health research, and decreased its support of defense research.

Organization of S&T Policy Making

The four most important organizations for S&T policymaking in Japan are the Prime Minister's Council for Science and Technology (CST), the Science and Technology Agency (STA), and two ministries: the Ministry of Education, Science, Sports and Culture (herein referred to as Monbusho), and the Ministry of International Trade and Industry (MITI). CST, composed of cabinet ministers and agency heads (e.g., Finance, Monbusho, STA, and the Economic Planning Agency), as well as representatives from universities and industry, advises the Prime Minister's office on long-term research goals. The role of the CST is becoming more important. In 1992, the CST provided the framework for the current S&T policy called, "General Guidelines for S&T," reemphasizing the importance of focusing on basic research. CST, chaired by the Prime Minister, with a secretariat within STA, is attempting to coordinate science programs across all ministries, and is considering the possibility of integration of science programs across government agencies. Such integration of programs may be discussed in future plans (Ito, 1996).

One difference between Japan and the United States in funding R&D is that the above mentioned main Japanese science agencies use intermediate organizations to carry out their funding and selection of proposals. (U.S. science agencies, such as the National Science Foundation and the National Institutes of Health, fund the research community directly.) For example, Monbusho administers its fellowship programs for doctoral students, postdoctorate researchers and international exchange of scientists, as well as its newly initiated "Research for the Future" funding program, through the Japan Science and Technology Corporation (JST).[10] Several STA programs are implemented by the Japan Development Research Corporation (JDRC). The Agency of Industrial Science and Technology (AIST) and the New Energy and Industrial Technology Development Organization (NEDO) carry out a variety of MITI's science programs. Thus funding targets for a program can sometimes be described in terms of the implementing agency. For example, under the S&T Basic Plan for 1996, "There is a target of funding 10,000 JSPS fellowships[11] for postdoctorates and graduate students by the year 2000." (See section on Academic Research.)

Within the Science and Technology Basic Plan of 1996, the S&T structure has remained relatively unchanged, but component parts are expected to work together (figure 11). There is greater cooperation among Monbusho and the other main science agencies. Plans presented to the Ministry of Finance between June and August of 1996 dealt with technical improvements and means to remove barriers among the three major science ministries. For example, for the first time, STA and MITI may directly fund research projects in universities. This level of cooperation between agencies is a new development (Ito, 1996).

New funds have been provided to the main S&T agencies for new basic research programs, often referred to as "basic strategic research" (table 5). This term for basic research reflects Japan's view of its need to invest in basic research as an overall strategy; it does not refer to targeting certain areas. In addition to these new programs, science and technology agencies are implementing a range of policies which give priority to the following four areas:

table 5

Ministerial R&D budgets are composed of: (1) funds for ongoing research-related operations and personnel costs of government research institutes and higher educational institutions, and (2) funds for research promotion-primarily in the form of external grants.

figure 11a
figure 11b

Monbusho and STA have the largest budgets, accounting for 49 percent and 26 percent, respectively, of the government's total R&D budget (figure 12). MITI is the third largest science funding agency (12 percent), followed by the Defense Agency (6 percent), the Ministry of Agriculture (4 percent), and the Ministry of Health and Welfare (3 percent). Most of Monbusho's funds are committed to the ongoing operations of the university system, including general university funding, as previously described.

figure 12

Main Science Funding Agencies


With approximately one-half of the Japanese government R&D budget in 1994, Monbusho is responsible for promoting science at all levels of education, as well as funding university research, attached laboratories, and national inter-university research institutes. Several of the latter joint-use laboratories were established by Monbusho in the second half of the 1980s to upgrade university research facilities, and to provide access to researchers from all national universities. These include the Institute of Statistical Mathematics, the National Astronomical Observatory, and the National Institute for Fusion Science.

In the 1970s, Monbusho financed the building of new universities and research facilities, such as Tsukuba University and the National Laboratory for High Energy Physics (KEK). The science policy in the 1980s was focused on halting further expansion of universities and decreasing budget ceilings (universities would get less than the previous year). The expansion of facilities' budgets in the 1990s is primarily to upgrade obsolete university facilities rather than build new ones.

Supplemental budgets in 1993 and 1995, requested from the Diet (Japanese Parliament), provided large increases for research facilities as economic stimulus packages. In 1993, the supplemental budget of 560 billion yen (2.4 billion dollars) was used for much-needed improvements in university equipment and facilities. In 1995, the supplemental budgets for science, amounting to 680 billion yen (almost 3 billion dollars), allowed agencies to shorten the construction period of "big science" facilities and to provide additional fellowships for foreign and domestic scholars.

The Central Council for Education, Monbusho's most important body for forming education policy, recommended revitalizing Japanese universities with major funding increases. In 1995, Monbusho increased its competitive grants-in-aid program to university professors within national universities to approximately $440 million. These additional funds allowed them to apply for, and obtain, larger competitive grants than previously had been available for university researchers. As reported in the 1996 national R&D survey, these competitive grant funds in 1995 were in addition to the more than $3 billion Monbusho provided for base salaries and university research at these institutions through formula funding (table 6).

table 6

Monbusho uses the Japan Society for the Promotion of Science (JSPS) to administer its fellowships for doctoral students and postdoctorate researchers, as well as the new Research for the Future program for university research.

As with all the major science organizations in Japan, Monbusho has advisory boards with representatives from the science community in both industry and academia. For example, Monbusho's University Council provides advice on matters concerning universities through several working groups. One of these groups provided a forecast of S&E personnel needs in 1991 at the height of Japan's economic growth period, that was very optimistic about industry's future demand for highly trained personnel. This report of the University Council was the basis for Monbusho expanding graduate programs and increasing the limits on graduate student admissions.

Science and Technology Agency (STA)

The STA is responsible for funding a number of national laboratories, such as the National Aerospace Laboratory and the National Institute of Science and Technology Policy. (See figure 11 for the list of laboratories under each science agency.) They also fund big science facilities such as the RIKEN Ring Cyclotron and deep-sea research vessels, as well as competitive grants-in-aid funding programs. The STA also contributes to formation of government science policy through detailed surveys of research areas, and produces an annual "White Paper on Science and Technology," which presents the results of these surveys.

Throughout the 1980s STA provided large funding and sophisticated equipment to its national laboratories, and they also experimented with a new funding mechanism for interdisciplinary basic research in strategic areas. This program of Exploratory Research for Advanced Technology (ERATO), managed by JST, sought to improve the quality of research in the Japanese system by bringing together university professors, industry researchers, and scientists from national laboratories. Under this scheme, researchers undertake a project for a 5-year duration. These programs required slowly changing the culture of science to move toward joint projects between industries and universities (Koizumi, 1993). (See "ERATO Evaluation" conducted by the NSF-supported Japanese Technology Evaluation Center (JTEC).)

STA's funding of competitive grants programs to promote basic research increased by more than $100 million from 1990-96. STA's list of 10 such programs (table 7) includes both established programs introduced in the 1980s, such as ERATO and the Frontier Research Program, as well as the new Strategic Basic Research Promotion. Most STA research funds are channeled through a number of public research corporations, such as the JST and the Institute of Physical and Chemical Research (RIKEN). (See RIKEN's "Frontier Research.") The International Joint Research Program, begun in 1989 and managed by JST, funds international collaborative basic research with other countries. Another STA funded program, begun in 1991, is the so-called Precursory Research for Embryonic Science and Technology System (PRESTO). This program is to fund a single individual for basic research, rather than the scientific teams funded in the above programs. While the ERATO projects were strictly off-campus, the 1996 basic research programs of STA and MITI can now directly fund university professors and build up the excellence of their university laboratories.

table 7

ERATO Evaluation

A U.S. panel of experts commissioned by JTEC to evaluate basic research under the ERATO program concluded that the research performed under ERATO is of high quality, with several projects leading to the development of world class research. ERATO introduced over a dozen new research directions in Japanese research. For example, under one ERATO project, a Japanese research group first realized that scanning tunneling microscopy (STM) could be used to pick up atoms one by one, to control matter on an atomic scale. A spinoff of this project is the Joint Research Center for Atom Technology (JRCAT), a 25-trillion yen, 10-year effort funded by MITI. The Director of JRCAT noted that, "ERATO was the precedent for this research" (Maruyama, 1996).

Besides carving out new research directions, ERATO also introduced new funding mechanisms of fixed-term projects and research structures in the Japanese S&T system that set a precedent for several other "ERATO type" programs that followed. ERATO funding mechanisms allowed STA to fund university professors for a 5-year commitment to a project; some professors took a leave of absence to head an ERATO project; others maintained their teaching while conducting the off-campus research. The new research structure allowed university and industry researchers, as well as scientists from national laboratories, to collaborate. Many ERATO projects were carried out within the Tsukuba Research Consortium near Tokyo. The 14-year experience with ERATO has increased mobility of researchers while providing 5-year fixed-term projects for young scientists.

RIKEN's Frontier Research

As one of the most highly regarded research organizations in Japan, RIKEN conducts multi-disciplinary research, often in cooperation with international institutions and visiting foreign scientists. RIKEN, founded in conjunction with the Japan Atomic Energy Research Institute in 1917, was modeled on the German research institutes (now Max Planck Institutes). In physics, RIKEN is constructing a large-scale synchrotron radiation facility named SPring-8 (Super Photon Ring) in the Harima Science City in Hyogo Prefecture, about 35 miles west of Osaka. Upon completion in 1998, it will be the world's largest ultrahigh-brilliance X-ray synchrotron radiation facility, and available to researchers from Japan and abroad. In addition, through RIKEN, Japan will contribute to the $20-million collaboration in spin physics research at Brookhaven National Laboratory's Relativistic Heavy Ion Collider (RHIC) in the United States. When commissioned in 1999, RHIC will be the world's highest energy-collider of heavy ions, and also the highest energy collider of spin-polarized proton beams for physics research.

In 1986, RIKEN initiated the Frontier Research Program to support long-term basic science in such areas as brain research, new materials, and photodynamics. In 1994, there were 500 permanent researchers and nearly 2,000 visiting scientists. Among the priorities in the Japan science budget for 1997 is 10 billion yen ($43 million in constant dollars) for brain research, including brain mechanisms of mind and behavior, information processing, and neuronal functioning. RIKEN's facilities and expanded research programs have implications for U.S. collaborators. The Frontier Research Program recruits personnel from overseas as well as from Japan, and a number of non-Japanese team leaders hold supervisory positions (STA, 1995).

Ministry of International Trade and Industry

MITI played a key role in technology policy during the reconstruction of the Japanese economy after World War II. Technology policy began with a focus on meeting the requirements for industrial development in the 1950s and later shifted primarily to promoting international trade in the 1970s. MITI currently supports its own programs of industrial research and development through the Agency of Industrial Science and Technology (AIST). But in addition, MITI is supporting basic research and a more open exploration of possible innovations than had previously occurred. In 1995, the New Energy and Industrial Technology Development Organization (NEDO), which is affiliated to MITI, introduced a new grant scheme for large grants to researchers at university and government laboratories.

As part of Japan's effort to support future innovation through basic science, MITI created the National Institute for Advanced Interdisciplinary Research (NAIR) through the reorganization of existing laboratories, as well as the National Institute of Bioscience and Human Technology. NAIR is to contribute to the generic technology base for future industrial S&T. AIST has restructured R&D into the Industrial S&T Frontiers Program, which aims at technological breakthroughs by linking industry, academia, and government. Some of the research themes under this program are superconductivity, biotechnology, new materials, electronics, machinery, and human sensory perception. Under MITI's Frontiers Program, projects such as the Joint Research Center for Atom Technology (JRCAT) are experimenting with mechanisms for mixing scientists from universities, national institutes, and industry. (See "Atom Technology.")

MITI is also attempting to revitalize the research conducted at its national laboratories. For example, in 1993 MITI consolidated several of its life science laboratories into the National Institute of Bioscience and Human-Technology, and then hired a leading university researcher as its Director-General. Through fundamental reforms in these national laboratories, MITI has the goal of encouraging biotechnology research that would match international standards.

Atom Technology

Under MITI's Industrial S&T Frontiers Program, the Joint Research Center for Atom Technology (JRCAT) has gathered approximately 100 scientists from a consortium of industries, national laboratories, and universities to conduct basic interdisciplinary research in atom technology. This technology is based on the invention of scanning tunneling microscopy (STM) by Rohrer and Binnig of IBM Zurich Research Laboratory. By this technique, it is possible to observe and transfer individual atoms. This frontier S&T area in material science spans work in new materials, electronics, biotechnology, and chemistry. The goal is to develop innovative structures and materials through manipulation and control of individual atoms and molecules. It is expected to have engineering applications related to catalytic reactions, defects and impurities in semiconductors, and electrode/colid interfaces (Maruyama, 1996).

A consortium of 30 Japanese firms, including Fujitsu, Hitachi, NEC, Sharp, Sony, and Toshiba, are contributing researchers (and a small annual fee) for participation in this program. JRCAT, located in MITI's National Institute for Advanced Interdisciplinary Research (NAIR) within Tsukuba, also receives 26 researchers from national laboratories in close proximity, mainly the Electrotechnical Laboratory and the National Institute of Materials and Chemical Research. The JRCAT budget is 25 trillion yen for 10 years ($1.1 billion in constant dollars, or approximately $100 million per year), mainly funded by MITI. Funding received directly from MITI (10 percent) has many restrictions. The majority of MITI funding, through NEDO, gives the project the flexibility to hire university professors to lead key groups in this research, as well as postdoctorate researchers and technicians. The funding also provides for travel to one international meeting per year for each of the 100 researchers, for small workshops on specific topics, and for organizing and hosting a large international conference on atom technology. Postdoctorates include young researchers from Japanese industry and universities, as well as young foreign scientists.

For example, one experimental group of JRCAT is exploring new electronic materials and related physics for development of atom technology. The exploratory materials are 3rd transition metal oxides and organic molecular systems with specialized functions. This work will contribute to critical-state phase control on solid surface and solid materials. Results, reported in Nature in 1995, include a new oxide of manganese with interesting properties. By applying magnetic field resistance, the electrical conductivity of the material increases as much as 10 orders of magnitude. In the United States, the American Physical Society (APS) meeting in 1995 held a new session on this effect. A conference on this topic, colossal magneto resistance (CMR), was held in Tsukuba in 1996.


[9] This is in addition to the Japanese government's $10.9 billion budget for science in 1995.
[10] The Japan Development Research Corporation (JDRC) was recently combined with JICST to form the Japan Science and Technology Corporation (JST).
[11] The fellowship total of 10,000 will include positions funded by JSPS (the largest number) and by STA (through JST) and MITI (through NEDO).