NATIONAL SCIENCE FOUNDATION
TOKYO REGIONAL OFFICE

May 15, 2000

Report Memorandum #00-08

A Radical Restructuring of 
Japan's Postwar S&T Policy and Institutions: The Politics and Rationality of the New Century

A series of implementing laws for a major reorganization and reform of the Japanese Government, enacted by the Diet in July 1999, are scheduled to go into effect on January 1, 2001.  (See Report Memorandum #99-11, dated October 7, 1999).  The following paper by Yong S. Lee, Koichi Kitazawa, and Shigeru Nakayama, reproduced here with the permission of the authors, provides detailed information both on the details of the reorganization itself, and on the policy and political background that led to its enactment and to the November 1995 enactment of the Science and Technology Basic Law.  (See Report Memorandum #96-21, dated August 28, 1996, and Report Memorandum #98-06, dated April 7, 1998). 

Yong S. Lee, Professor Political Science and Iowa State University, was in Japan on a JSPS Postdoctoral Fellowship from January through March 2000 and was hosted by Koichi Kitazawa, Professor in the Department of Advanced Materials Science at the University of Tokyo.  Shigeru Nakayama is Professor Emeritus of History at Kanagawa University. 

Inquiries can be directed to yonglee@iastate.edu.

Contents

  Acknowledgement

I.            Purpose of the Study

II.          Theory and Method
                    Policy Discourse as Paradox
                    Methods of Observation

III.         New Science Policy
              How the Debate Began
                    Economic Rationality
                    Political Rationality
                    Bureaucratic Rationality
                    Organizational Dynamics
              New Policy Framework

IV.         Centralizing the Policy Process 
              Institutionalization of the Prime Minister's Office
                    Cabinet Secretariat
                    Cabinet Office
                    National Council for Arts, Science and Technology
              Merger of STA with Monbusho 

V.          Granting Autonomy to National R&D Institutions
              The Concept of "Independent Administrative Agencies"
              Independent National Universities
              Independent National Research Institutes
                    Example of MITI Labs
                    Example of RIKEN Labs

 VI.        Concluding Observation 

Endnotes 
References

Acknowledgement

This study was made possible by a Short Term Invitational Fellowship from the Japan Society for the Promotion of Science (JSPS) awarded to Yong Lee in the spring of 2000 to collaborate with Koichi Kitazawa, Tokyo University, and Shigeru Nakayama, Kanagawa University. The study was also supported by a grant from the Science, Technology, and Society of Iowa State University in the summer of 1999. The Division of International Programs of the U.S. National Science Foundation (NSF) and NSF's Tokyo Regional Office provided valuable assistance in coordinating this international collaborative project. Finally but not least, Iowa State University provided Yong Lee with a faculty improvement leave for the spring of 2000, making it possible to carry out the study. We express our gratitude to the members of Japan's scientific community and the officials of Japan's government who participated in this study. They shared their personal views, concerns, and insights with us without which this study would not have been possible. The analysis and conclusions, however, do not represent the views of the participants, nor the views of the sponsors. All errors, if any, belong to the authors.

A Radical Restructuring of Japan's Postwar S&T Policy and Institutions:

The Politics and Rationality of the New Century  

PURPOSE OF THE STUDY  

In 1995, as the economy was still in deep recession, Japan's Parliament did something unthinkable. On its own initiative, it enacted the "historical" Science and Technology Basic Law announcing that it would increase government research and development (R&D) expenditures to 17 trillion yen (roughly $150 billion) over the next five years.^[1] The increase meant that the Parliament would "double" the present level of government R&D expenditures by 2000. In addition, the Parliament also hinted at a possibility that within the next 10 years or so (by 2006) it might raise government R&D expenditures to a ratio to that of G-7 nations.

Understandably, the world's scientific community was ecstatic to hear the news.^[2] For long the world's scientific community was critical of Japan's postwar R&D policy, which emphasized the "D," leading to a relative neglect of basic research. Right or wrong, the criticism was that in spite of a huge technology-based economic success in the postwar era, Japan was not contributing much to the advancement of knowledge. But when the initial excitement had subsided, many began to ask, "Is Japan serious?" and began to monitor Japan's complicated budgetary politics. The latest news, as the world enters a new millennium, is that Japan's Parliament has kept the 1996 promise. Moreover, the Obuchi government has set in motion a second five-year plan preparing for further increase of R&D expenditures.  

Taking a closer look, the story of \17 trillion is not the most important news to Japan's scientific community--albeit its public relations effect. Obviously, it is important to know how much and where a wealthy nation such as Japan is investing its R&D money. After all, Japan's economy is the second largest in the world, as are it's national R&D expenditures. But this "money talk" diverts attention away from the larger and more fundamental structural revolution shaping in Japan. While leveling up the government share of R&D expenditures, Japan has undertaken a radical restructuring of both its postwar science policy and institutions that in contemporary vernacular amounts to a "reinventing" of Japan's 99 national universities and 86 national research laboratory complexes. Moreover, the structural revolution realigns the entire administrative state, from the Office of the Prime Minister to the mighty national bureaucracy. This is a large and daunting task. Masao Ito, former president of Japan Council of Science and now Director of Japan's Brain Science Institute, calls this change the third most important revolution in the history of Japanese science. The first revolution he refers to the Meiji era starting in 1868 in which Japan began to invite in Western scientists and engineers sewing the seed for the later industrial revolution, and the second to the postwar Japan rising from the ashes to construct a technology-based economy.^[3] The structural reform undertaken at this point is envisioned as a step toward recreating Japan as a nation joining the rank of leading nations in the exploration of the frontiers of knowledge. 

 The change advances several theoretically interesting hypotheses that can be tested empirically.^[4] But for now, it is necessary that we describe the shape of Japan's new science and technology (S&T) system that is emerging at the turn of this new century. The purpose of this paper is to do just that by adding our own interpretations to the emerging new S&T framework and to the underlying politics and rationality. In addition, the paper describes the new policy process and institutional configurations for government-supported R&D. Since restructuring is still in progress, much more remains to be seen. Still, sufficient contours of the new system have already been unveiled for public examination. Of interest to this study is why and how Japan is taking this particular path at this particular time in history. This requires discussion. In this article we speculate on various factors, political and rational, that appear to underscore the restructuring effort.

THEORY AND METHOD

Policy Discourse as Paradox

Restructuring a large S&T system such as the one in Japan ($30 billion a year) can be a complex and unwieldy process since it would invariably galvanize a conflicting array of institutional self-interests. This kind of complexity always tries rational thinking. The rationality framework urges us to think rationally--in terms of objectives and means by which to achieve them efficiently. "I hate politics," people moan, largely because the language of politics often frustrates their familiar mental construct of rationality. Politics is too messy to make a good use of textbook reasoning skills.

In reality, there are not many things in life that obey the rationality rule. Certainly, not in the political community. Even in the subatomic world (e.g., electronics) scientists have long abandoned the classical rationality scheme in favor of viewing the world in a new light. Only when they abandoned the rationality framework did scientists begin to understand the mysterious workings of "complementarity" in nature.^[5] The nature of light, for instance, defies a rational description. This is because light behaves like a particle when it is looked at as a particle, but it also behaves like a wave when it is looked at as a wave. In nature, therefore, "what you see is what you get."

Paradoxes of this kind abound in public policy discourse. To cite a classic example from the social sciences, under certain circumstances (e.g., the arms race) rational thinking inadvertently leads to the worst possible result (e.g., nuclear war).^[6] As a matter of fact, it is not easy to understand how a democratic government really makes controversial policy decisions without learning the language of paradox. In this article we argue that much of what goes on under the banner of government reorganization in Japan cannot be described intelligently without resorting to the language of paradox. This is not unique to Japan. In a liberal democracy where agreement is key to collective action, stakeholders often agree for different reason, they may agree in generality but disagree in specifics, and they may waver between decisions. This is actually a blessing in disguise since this kind of contradictory behavior actually makes it possible for stakeholders to compromise among themselves in pursuit of collective action, leading to outcomes more favorable to all parties. In reality, the dynamics of collective action (public policy) look a lot closer to paradox than to rational thinking. Lindblom could not have been more correct when he observed that democracy is a system of "muddling-through."^[7]

And yet, scholars of public administration have long searched for a good (rational) way of organizing the administrative state. A classic example is the work of Gulick who believed that the principal purpose of organization and reorganization should always be to maximize efficiency.^[8] To do this, the administrative state should follow one of five structural principles: organization by purpose, organization by function, organization by geography, organization by process, or some combination thereof. For some time this dictum was an administrative orthodox and was accepted widely as a way to structure the modern government.

This orthodox, however, was soon challenged by young Herbert Simon in his doctoral dissertation, which later was published as ­Administrative Behavior­.^[9] Simon's argument was that Gulick's organizational principles were in fact not principles at all, but rather a collection of anecdotes. Each of these principles, argued Simon, is competing against one another: the principle is useless unless the conditions are specified. So went the debate.

In the United States, Congress occasionally establishes an administrative reform committee to evaluate the efficiency of public administration and make recommendations for reorganization. Intrigued by many conflicting reorganization strategies often employed by the federal government, Seidman undertook a study to investigate how efficiency principles are applied in federal reorganization efforts.^[10] Seidman's finding dealt a tragic blow to Gulick's thesis. What Seidman discovered was that for the most part federal reorganization efforts had very little to do with the principle of efficiency. To his dismay, Seidman found that politicians and administrators invoke the language of "efficiency" largely as a cover for political objectives. In all likelihood, Seidman's study went to an extreme; however, he made his point. Reorganization of the administrative state is essentially a political process centering on "who gets what, when, and how."^[11] This does not mean that efficiency is irrelevant. What it suggests is that efficiency and politics often are the proverbial "strange bedfellows."

Recently Stone advanced an argument that underpins the orientation of the present study: that collective action of which the reorganization of government is an example of collective action is essentially a policy paradox.^[12] To begin, modern liberal democracies embrace many aspirations, including freedom, liberty, justice, equality, equity, efficiency, and security. “These are ‘motherhood issues' . . .everybody is for them when they are stated abstractly," observes Stone, "but the fight begins as soon as we ask what people mean by them.”^[13] Thus, policy conflict spreads when one aspiration (e.g., equality) is promoted unavoidably at the expense of another (e.g., efficiency). Since agreement is central to collective action, policymakers, when defining policy problems and articulating alternatives, are compelled to rely on the symbolic use of language—e.g., template stories, synecdoches (extreme examples), metaphors (comparative images), and ambiguity. A template of the policy story may begin something like this: "In the beginning, things worked very well. But their usefulness has been outlived. In fact, things are getting very bad. Something must be done." Similarly, policymakers invoke synecdoches (e.g., "welfare queen") or other colorful metaphors to characterize the larger policy problem to which they wish to address.

The most versatile tool in the lexicon of symbols, observes Stone, is the language of "ambiguity." Policymakers relish this language because it enables them to transmit different meanings to different people.^[14] "Some people think that this is a good thing; others, a bad thing, " argues Stone,  "but whatever one's assessment a type of policy analysis that does not make room for the centrality of ambiguity in politics can be of little use in the real world."^[15]  Stone reminds us that the language of ambiguity allows policymakers to aggregate support from different interest groups, ultimately transforming conflicting individual and group intentions into collective purposes. In short, the language of ambiguity allows policymakers "to unite people who would benefit from the same policy but for different reasons."^[16]

We find that Stone's "policy paradox" is particularly useful for the description of reform underway in Japan. Without this practical orientation, the reform efforts are too unwieldy for a coherent description. The reform looks like a rationality project if it is looked at as a rationality project; it also looks to be politically motivated if it is looked at as a struggle for power. We need not make a grand choice. Any attempt to define the reform as a rationality project will miss the real reasons underlying various reform decisions. By the same token, were we to argue that everything is politics, we are likely to miss an important point of the emerging national aspiration.  The language of policy paradox reassures us that these apparently competing hypotheses are actually complementary to one another, and together, they provide fuller explanation. One of the contested debates in the reform process, for example, is over the proposal to restructure the national R&D system as an independent administrative regime. Is this proposal a means by which to create flexibility and competition or a political ploy by which to downsize government? Which is real? As far as our study is concerned, there is no one correct answer here. And yet, should we look for one authoritative answer, we would be disregarding the cardinal principle of ambiguity in collective action. 

Methods of Observation

The method of observation in this study is in-depth expert interviews. The interviews are also combined with an analysis of documentary evidence. The interviews have been conducted with leading members of Japan's scientific community and with government officials who are directly or indirectly involved in the reform process. Represented in the interviews are the members from RIKEN (Institute for Physical and Chemical Research), KEK (High Energy Accelerator Research Organization), Japan Engineering Academy, national universities, the National Personnel Authority, STA (Science and Technology Agency), NISTEP (National Institute for Science and Technology Policy), the Ministry of Education (Monbusho), the Ministry of International Trade and Industry, and the Prime Minister's Office.^[17] Although the interview began with a carefully drawn list of names, the actual interview process has been "snow-balling," with the list expanding at the suggestion of the previous interviewer. Because of the complexity involved in formal and informal reform decisions, some interviewees have been visited more than once. In addition, we made an effort to cross check our interview results against the observations of others, including the NSF Tokyo Office, and the Science Magazine's Tokyo Office.
      The interviews have been carried out in two parts. The first part was conducted during the summer of 1999 at which time the questions were focused on the enactment of the 1995 S&T Basic Law and the adoption of the 1996 S&T Basic Plan.^[18] For this part of our interview, efforts were made to understand the political dynamics underscoring the development of the new S&T policy. The second part of the interviews was conducted in the early spring of 2000.^[19] The interviews at this time were focused on institutional change and the emergence of a new R&D structure. For each interview, we prepared a list of questions relevant to the official position and experience of each interviewee. On average, the interviews lasted about two hours each, and upon completion, the content was transcribed within the same day. All but a few interviews were conducted in English. At times, Kanji (Chinese characters) were used to clarify the subtleties of "translated" English titles and nomenclatures.
      While Lee alone conducted all interviews, Kitazawa collaborated with him throughout the interview process by planning for interviews and clarifying interview results. After each interview, Kitazawa and Lee engaged in lengthy discussion on the interview result, placing it in larger perspective. This discussion generally shaped the next round of interviews. To ensure the reliability of the interview results, extensive efforts were made to cross check the information.^[20]
       In addition to these interviews, we examined relevant institutional records, internal discussion papers, and official policy documents. The government ministries, national universities, and research institutes have provided us with generous access to all these documents, in Japanese as well as in English.^[21] 

NEW SCIENCE POLICY

How the Debate Began

In Japan's national policy process, the Parliament enacts laws and authorizes the government (meaning the executive branch) to work out the details. Normally in this process, the government initiates policy agendas, and the Parliament generally take action on government proposals. In the case of the 1995 S&T Basic Law (hereinafter Basic Law), it was the Parliament that took the leadership in drafting the bill. After its enactment into law, the Parliament authorized the government to formulate the S&T Basic Plan (hereinafter called Basic Plan) for implementation.^[22]  While the Basic Law received support from all political parties, the legislative leadership came from the Liberal Democratic Party (LDP) under Prime Minister Murayama, chairman of the Social Democratic Party.
       In July 1996, Prime Minister Hashimoto's government completed the formulation of the Basic Plan, recommending it to the Parliament for approval. The Basic Plan recommended that the government "double" the R&D investment by year 2000, and the Parliament endorsed the Plan. Immediately thereafter, Hashimoto established the Council for Administrative Reform, chaired by himself, to recommend a bold new plan for reorganizing the government. In 1998, Keizo Obuchi became Prime Minister, and he embraced the work of the Council with enthusiasm. In June 1998, Obuchi submitted the Council's reform proposal to the Parliament outlining what has turned out to be a radical restructuring of the national R&D system. The Parliament promptly approved the reform proposal by enacting it into law titled "The Basic Law for Government Administrative Reform." As we will discuss shortly, the Reform Policy introduced the concept of "independent administrative agencies" to national research laboratories and national universities.^[23] Meanwhile, the Obuchi government has set in motion an informal process for the second 5-year plan for science and technology, from 2001 to 2006. The aim of this second round of planning is to raise the percentage of government R&D expenditures to a level comparable to those of leading nations, and to determine Japan's S&T policy goals more strategically. According to the reform schedule, the reorganization of government will go into effect April 1, 2001. The final status of national universities, however, will not be determined until early 2003. 

Economic Rationality

Why would Japan's government want to change its postwar S&T policy framework? For the entire postwar era, Japan's S&T policy has been enormously successful. For nearly two decades in the 1970s and 1980s, Japan's technology-oriented research enterprise was the envy of the world. The United States was losing competitiveness, and this even precipitated a minor trade war between Washington and Nagatacho in the late 1980s.^[24] Why would you change a system that works well? The answer is simple: as Japan's science policy community sees it, the system does not work well; it had worked well in the postwar era, but this is no longer the case.^[25] The long recession in the 1990s is symptomatic of policy failures and serves as a warning signal that Japan's economic prosperity may be coming to an end. Underlying this analysis is the observation that the center of Japanese society (its inner strength) may be hollowing out, and that Japan's industrial competitiveness may have plateaued. Japan, as a nation, may not be able to survive the new global economy unless it quickly re-orients the S&T framework that promotes its own technological breakthroughs.
      In essence, this argument is a replay of Vannevar Bush's familiar but controversial dictum: "A nation which depends upon others for its new basic scientific knowledge will be slow in its industrial progress and weak in its competitive position in world trade."^[26]  The underlying assumption is that there is a so-called linear relationship between basic scientific research and technological innovation. "Basic research," in Bush's words, "is the pacemaker of technological progress."^[27] Elsewhere in his writing, Bush articulated this linear relationship: 

Progress in the war against disease depends upon a flow of new scientific knowledge. New products, new industries, and more jobs require continuous additions to knowledge of the laws of nature, and the application of that knowledge to practical purpose. Similarly, our defense against aggression demands new knowledge so that we can develop new and improved weapons. This essential, new knowledge can be obtained only through basic scientific research.^[28] 

In the postwar literature it has long been argued that the Bush theory did not apply to postwar Japan, or even to the United States.^[29] But the Bush dictum has arguably returned to Japan. After all, at the advanced stage of economic development in which Japan now stands, Japan's science policy community argues that basic and fundamental research is essential for technological innovation and economic growth. From interviews and perusal of official pronouncements we are persuaded that the Bush dictum is a predominant argument underscoring the restructuring of Japan's postwar S&T framework.
      The origins of this argument require discussion. In the late 1980s the Japanese society began to show the signs of growing frustration within the Japanese public owing to both political corruption in high places and economic slowdown. When the stock market crashed in 1992, throwing the economy into disarray, the Japanese public saw it as the beginning of the larger systemic problems which the Japanese refer to as kuhdohka, or the "hollowing out of the center." The fear of kuhdohka began to permeate the public, and policy makers were forced to look for a fundamental answer, not a quick fix.
      The kuhdohka argument holds that Japanese society is aging rather rapidly, with a corresponding decline of the youthful and productive population. The statistics compiled by the Ministry of Health and Welfare shows that the age bracket of 65 years old or older is on a rapid increase. This age bracket represented about 7% of the population in 1970, but the percentage increased to 15.4% in 1997 and is projected to increase to 28% in 2020.^[30] Worse yet, as expressed by our interviewees, the younger generation does not share the same work ethic as the older generation. The young generation argues that their parents worked hard to make the country rich but their hard work has not translated to a quality of life enjoyed by other advanced nations. What the older generation fears now is that if the younger generation abandons their parent's work ethic, the quality of life in Japan will seriously deteriorate.
      The kuhdohka argument also holds that Japanese industries are losing their inner strength. The comment we hear consistently from our interviewees is that Japan's innovation system has reached a plateau. Sadly, the once mighty industrial machine has reached its limit and has now become confused. One interviewee illustrated the situation this way:

In the postwar era, the nation's vision for industry was to 'catch up' with the West technologically. The S&T strategy then was for industry to import science and technology from the United States and other western nations. Japanese are good at engineering. Through technology transfer and engineering, Japan emerged among the world economic power. So far so good. Japanese industry has caught up with, and even exceeded in some areas, the U.S. industry standard. Japan's industry now is in the big league, but this big league runs on different rules. You cannot copy from others because you are already on the cutting edge. You must create your own knowledge and produce technological breakthroughs. But Japan's S&T framework and industrial strategy have not been structured to compete in this new playing field. Unless the entire national innovation system is overhauled and begins to produce "home-grown" scientific and technological breakthroughs, Japan's industrial machine will not be able to perform new miracles.^[31] 

Can Japan meet this challenge? The response we get from our interviewees is, "Maybe it can . . . if there is a radical change, but not as it stands now." For too long in the postwar era, they argue, Japan's government has neglected to pay proper attention to fundamental research. One cannot blame industry for this failure because this is the government's responsibility. To understand the nature of this argument, we look at Japan's R&D investment patterns (Figures 1 and 2).
       Figure 1 compares the percentage of national R&D expenditures to GDP (Gross National Product) among five nations (the U.S., Japan, Germany, France, and Italy). According to the data from 1981 to 1995, Japan has been spending the largest proportion of GDP on research and development. A caveat is that this data, assembled by STA (Japan's science and technology agency) from several sources, including NSF and OECD, is but a rough sketch. The tax laws of each government are different, and some, like that of Japanese government, are extremely complicated. Moreover, the classification system of R&D, particularly the concept of "development," varies from country to country.^[32] And yet, the sheer magnitude and consistency of the investment pattern afford credence to the hypothesis that Japan, as a nation, may have been investing proportionally more than other leading nations in research and development. 

But this macro picture does not tell the whole story because the national R&D expenditures include both private and public investments. And the nature of R&D investment is different between industry and government. The bottom-line of industry R&D is to create new competitive products. In other words, the industrial firm is ultimately responsible for the free market. By contrast, the modern government has come to support basic and fundamental research that benefits society as a whole. Thus, to get a rough idea of where national R&D resources are allocated between basic research and applied and development research, it is necessary to look at the balance of national R&D expenditures between industry and government.
       Figure 2 graphs the percentage of government R&D to national R&D expenditures among leading nations. The data in the graph include all government R&D expenditures, including defense-related research.^[33] Comparison shows that the percentage of government R&D expenditures is the lowest for the Japanese government. The distance between Japan and the United States is also large. In 1997, for instance, the NSF data shows the ratio of Japan's government R&D expenditures to the total national R&D expenditures to be at 20.5%, and that of the U.S. government at 34.5%. This picture is a sharp contrast to the one in Figure 1 in which Japan is placed on the top of the national R&D scale. Japan may have been underinvesting national resources in basic and fundamental research. This leads to an inevitable conclusion: the government of Japan must increase its R&D expenditures hoping that the increased investment will engender technological breakthroughs.
       In sum, this conclusion appears to have become a powerful intellectual force shaping the new S&T policy embodied in the Basic Law and the Basic Plan. The Basic Plan explains the reasons for new S&D policy: 

The government will actively promote basic research for the following reasons. Achievements in basic research aimed at discovering new laws and principles, creating original theories, and predicting and discovering unknown phenomena are in themselves valuable as intellectual assets to be shared by all humans; the results of basic research contributes to the advancement of culture and at the same time provides dreams and hopes for people. Such new achievements at times radically change technological systems or even creates completely new technological systems, bringing about various positive spin-offs for society. Deep understanding of the human race and nature constitute the very foundation of human development, based on harmony between man and the environment.^[34] 

This rationale is remarkably similar to the one used a half century earlier by Vannevar Bush in Science--the Endless Frontier. A puzzling question is, why did the government of Japan wait until 1995 to recognize the centrality of basic and fundamental research? Why did the government not take action in 1992 when the stock market plunged deeply? In 1992, the Council for Science and Technology recommended that the government "double" its R&D funding as quickly as possible. Why did the government not heed this advice? The government could have re-oriented postwar S&T policy even earlier in the 1980s when all indicators suggested that Japan may be entering a decline in the business cycle. The point of these questions is that the formulation of new S&T policy in the mid-1990s may not be fully explained in reference to economic rationality. 

Political Rationality

To more fully appreciate the complexity of Japan's new S&T policy, it is necessary to look at the political context in which the Parliament passed the Basic Law in 1995. The argument we advance here is that had there not been a political upheaval in Japan's electoral politics from 1993 to 1995, it is conceivable that the new S&T policy may not have come into existence, at least not in its present form.
      In Japan's parliamentary democracy, as alluded earlier, the function of policy development is practically delegated to the government--that is, the national bureaucracy. The parliament generally acts upon government initiatives. The Parliament debates some controversial issues, but spends little time on legislative debate for others. This conjures up the image of Japan's Parliament being a pro forma institution insofar as policymaking is concerned. In this environment, the policymaking responsibility has been shifted to the national bureaucracy, and the national bureaucracy is very careful not to disrupt the status quo politics.
       This status quo political environment, which was in place until most recently, meant that high-level bureaucrats had to be sensitive to the needs of the ruling Liberal Democratic Party (LDP), which had controlled the government since it had come to power in 1955. In Japan, these status quo politics and the cozy relationship between LDP and the national bureaucracy are called the '55 system, referring to the year when LDP came to power. Under the '55 system, a responsible bureaucracy required that bureaucrats act in the interest of the ruling party; they needed not waste much time listening to opposition parties, newspaper reporters, and the public. One interviewee expressed his frustration this way: 

The Diet was not functioning. The Diet was exercising no legislative authority. It only took one or two legislative initiatives a year. All legislative initiatives were coming from the bureaucracy. And the bureaucracy was not willing to take the risk. People were frustrated, and they knew the old system was not working. But there was no other alternative.

With a new century approaching, the kuhdohka symptoms were intensifying, yet the status quo politics under the '55 system offered no room for bold policy initiatives. Although somewhat of a fluke, a crack was opened in the '55 system; this good fortune, according to Curtis, came in the form of a revelation by Asai Shinbun on June 18, 1988 of the historical "Kawasaki City Scandal" involving LDP leadership.^[35] It turned out that the Kawasaki scandal was just an indication of the larger political corruption in the LDP and high government offices. In 1989, the LDP had lost a majority in the Upper House. In 1993, owing to public dissatisfaction with the '55 system, the LDP lost control of the Lower House. For the first time since 1955 the LDP was forced to turn the government over to a coalition of six minority parties: the '55 system came to an end.
       The sudden demise of the '55 system set the stage for both the uncertain politics of the coalition government and the political instability that followed.^[36] In the 1994 election, the LDP still had a poor showing, giving way for yet another coalition of minority parties to control the government. Prime Minister Hata's coalition government, however, lasted less than four months. Finally, in the summer of 1994, the LDP returned to power in an unusual alliance with the Social Democratic Party, its ideological enemy, and elected the SDP Chairman, Mr. Murayama, as Prime Minister. In this precarious position, LDP leadership was under unprecedented pressure to demonstrate policy leadership to the electorate. It was in this charged atmosphere that the LDP, under its Socialist Prime Minister, invented the new S&T basic bill and passed it unanimously.
       According to the insiders' account, Mr. Kato, then the LDP Party Secretary, rallied the LDP troops to initiate new legislative proposals and regain public confidence. This presented an opportunity for Mr. Omi, Chair of the House Committee on Science and Technology, and a former Science & Technology Agency bureaucrat, to craft the bill which became the S&T Basic Law. It is widely known that while at STA, Koji Omi had strongly argued for the promotion of basic and fundamental research.
       A perusal of the Basic Law suggests that the promotion of basic research is its central theme.^[37] Article 1 of the Basic Law begins by stating that "the objective of this law is to achieve a higher standard of science and technology" and that the new policy will promote the "balance" and "harmony" among basic research, applied research, and development." The most crucial language is stated in Article 9, which states that the government shall establish a basic plan for the promotion of S&T. The significance of this provision is that the Parliament, de facto, authorizes the Council for Science and Technology (which is staffed by the STA) in the Prime Minister's Office, to formulate new S&T policy for the new century. Article 9 mandates that the Basic Plan develop a comprehensive plan for "the promotion of basic, applied, and development research, including technology development," and take concrete steps to improve "R&D facilities and equipment," and emphasize "information-intensive R&D activities." Finally, Article 9 states, "In order to secure necessary funds for the implementation of the Basic Plan, every fiscal year the government shall take the necessary measures for the effective implementation of the Basic Plan by including the necessary fund in the budget within the limits of national financial status."^[38] 
       In 1996, the LDP again took control of the government by aligning with two other minority parties. Under LDP Prime Minister Hashimoto, the government completed the Basic Plan, submitting it promptly to the Parliament with a request of \17 trillion spread over five years.
       Did the members of Parliament all agree with the purpose of the Basic Plan and with the price tag of \17 trillion? The Basic Plan was adopted with unanimity, but for different reasons. According to the insiders' view, the Hashimoto government was under great pressure to develop an economic package by which to stimulate the sluggish economy. In Parliament there was also a strong consensus on the urgent need to act. But, not surprisingly, the parties disagreed on which sectors of the economy to stimulate. At this point, some supported the S&T spending bill hoping that it would stimulate the economy, while others supported it with an understanding that the money would help rebuild the long neglected national R&D system.  

Bureaucratic Rationality

From the perspective of bureaucratic planning, the new S&T framework advanced by the Basic Law and the Basic Plan is not entirely new in substance; it is actually a continuing evolution of science policy in the scheme of the bureaucratic planning process.^[39] As evidence, the important elements of the new policy have already been articulated in the previous S&T planning documents. In 1959, the government embarked on a continuing S&T planning process. This began with the creation of the Council for Science and Technology (CST) in 1959 within the office of the Prime Minister. The CST, nominally headed by Prime Minister, consists of 10 council members (4 Ministers, 5 executive officials, and the president of Japan's Council of Science) with a small number of staff in its secretariat borrowed from the Science and Technology Agency. The CST receives additional technical support from relevant STA bureaus and the National Institute for Science and Technology Policy, STA's independent policy analysis unit.
       The CST meets once or twice a year to make policy recommendations to the Prime Minister. For specific policy analysis the CST relies on its work committee, the Committee on Policy Matters. This Policy Committee, in turn, is authorized to create ad hoc research committees to do additional work. Based on these subcommittee reports, the CST develops a long-term comprehensive S&T plan every ten years or so and recommends it to the Prime Minister. Thus, each policy recommendation is built upon the previous one with minor variations to ensure the continuity of science and technology policy.  The following benchmark recommendations are illustrative of the continuing planning process: 

To understand the genesis of the new S&T framework it is particularly important that one pays attention to the 1992 recommendation. As close examination will show, the 1992 planning document had already hinted at the need for a drastic shift in S&T policy for the new century. Further, the 1992 planning document recommended the government to "double" the R&D expenditures as quickly as possible. As a matter of fact, the Basic Plan acknowledges that the major R&D policy proposals have been built upon the planning document of 1992. The point of this discussion is merely to underscore the reality that the new S&T framework is actually part of a continuing bureaucratic project. 

Organizational Dynamics

What can be concluded from this multifaceted discussion?  Clearly, the shaping of Japan's new S&T policy is not purely a rationality project, nor entirely a bureaucratic project, nor a singular pursuit of political objectives, but rather it was all of these. The language of economic rationality routinely invoked in public speeches and official documents is what the discipline of collective action calls the "policy argument." At its core it is a rhetorical device attempting to portray collective action problems in rational terms. But ideas do not move the political community; collective action does. And only collective action affords legitimacy. In the case of Japan's new science policy, the impetus for collective action came in disguise--in the form of political instability, recession, and more generally, kuhdohka. Finally, it was the national bureaucracy, though disjointed, that breathed substance into the new policy. Suddenly, in 1995, all came together.
       In the business of government, the initial policy decision (collective action) is only the beginning. The collective decision must be transformed into organizational action if it is to realize its collective purpose. It is a mystery that once a collective decision has been institutionalized, the new system develops a life of its own--regardless of how it began.^[41] In a sense, this process is what scientists refer to as "self-organization."^[42] Self-organization makes transformation possible because without it, collective action can unravel into primordial (tribal) divisions. This mysterious self-organization is repeated in all evolutionary changes.^[43] On reflection, self-organization is not a mystery at all; it repeats like nature's template. We witness it through large-scale social changes such as the French Revolution, the American Revolution, and the Meiji Revolution. We also witness it at a micro-level when we put together a new work team in an organization. Transformation begins when the new system creates a life of its own in a cycle of evolution. Japan's new science policy is expected to do the same. 

New Policy Framework

In aggregation, what are the major characteristics of the new S&T policy? How different are they from those of the postwar policy? Are there parallels between this framework and that of the United States? To answer these questions it is necessary to take a quick detour through the lexicon of science policy classification. In the current usage, science and technology policy is classified on the basis of how government allocates its R&D funds among three areas: basic research, applied research, and development. According to NSF's Science and Engineering Indicators, research is classified by research of research objectives:

The objective of basic research is to gain more comprehensive knowledge or understanding of the subject under study, without specific applications in mind. Applied research is aimed at gaining knowledge or understanding to determine the means by which a specific, recognized need may be met. Development is the systematic use of the knowledge or understanding gained from research directed toward the production of useful materials, devices, systems, or methods, including the design and development of prototypes and processes.^[44] 

This classification is in keeping with the tradition of Science--the Endless Frontier, a report by Vannevar Bush to President Roosevelt at the end of World War II. Bush wrote: 

Basic research is performed without thought of practical ends. It results in general knowledge and an understanding of nature and its laws. This general knowledge provides the means of answering a large number of important practical problems, though it may not give a complete specific answer to any one of them. The function of applied research is to provide such complete answers. The scientist doing basic research may not be at all interested in the practical applications of his work, yet the further progress of industrial development would eventually stagnate if basic research were long neglected.^[45] 

While this classification system of basic and applied research is still widely used as a basis for comparing science policy among nations, the usefulness of this so-called "linear classification" has been a subject of controversy. This classification system is called a linear system because it assumes a linear progression from basic research to applied research and development. The principal criticism is that the classification fails to deal with interaction between the categories. In the real world of research, much of the relationship between basic and applied research is open and interactive rather than discreet and sequential. Recently, the late Donald Stokes advanced an alternative taxonomy addressing the missing interactive link between basic and applied research. We find that Stoke's taxonomy is practical and relevant to the discussion of Japan's science policy.^[46]  Figure 3 is Stoke's taxonomy developed on two dimensions: advancement of fundamental understanding and consideration of use.^[47]
       Briefly, the Bohr quadrant (symbolizing Niels Bohr's research) represents the concept of "basic research" as defined by Vannevar Bush in Science--the Endless Frontier.  The Pasteur quadrant (symbolizing Lewis Pasteur's research), however, captures the interactive dimension of basic and applied research that addresses both the advancement of knowledge and the consideration of use (application). If research is carried out primarily for use (application) without considering the advancement of knowledge, it is represented by the Edison quadrant (symbolizing Thomas Edison's research). While Stoke's model requires rigorous empirical testing, it removes the old controversy of classifying research (and academic disciplines) into a dichotomous world.

             If we are to look at science policy from the vantage-point of Stoke’s taxonomy, it appears that Japan’s new S&T framework is oriented toward the Pasteur quadrant, or "use-inspired basic research." If this is in fact the correct  characterization, the new policy is a significant departure from the postwar policy which emphasized technology development, or the Edison quadrant. Perusal of the Basic Plan shows that the new S&T policy puts priorities on the advancement of knowledge with emphasis on consideration of use. In the preamble, the Basic Plan makes this point clear: 

Japan is not only facing globalization and intensive economic competition with other countries, but also is aging at an unprecedented pace. People are deeply concerned that Japan may be heading toward a crisis in which industry is hollowing out, social vitality dissipating, and the living standard deteriorating. Moreover, the future existence of humans as a whole is confronted with global environmental issues, food problems, and energy and resource shortage. . .Expectation for basic research is very high; it is not only a basis of innovative development in science and technology, but such development in itself is valuable as intellectual property for all humans to share. Japan, in particular, is expected to boldly challenge unknown fields in science and technology.^[48] 

       The Basic Plan acknowledges that during the postwar era, the government neglected to promote basic and fundamental research. Yet, it does not argue completely in favor of pure basic research to the exclusion of application. Throughout the document, the Basic Plan stresses the centrality of balanced R&D emphasis among basic, applied, and development research: "The government will give consideration to the fostering of balanced R&D ability and to the harmonious development of basic, applied and development research."^[49] The implementation phase makes the case even stronger for "use-inspired basic research."  As part of the big push for basic research, in 1997 Japan established the Brain Science Institute at Institute of Physical and Chemical Research (RIKEN). The establishment of this research institute is premised on the idea: "Brain is one of new frontiers which has a good deal of potential, and its research is expected to lead to such results as improvement of medical science, creation of new technology and new industries and enhancement of quality of social life."^[50] Another example is the Genomic Science Center at RIKEN established in 1998. The same use-oriented theme continues with the Center: "Research on genome, gene and protein . . . the origin of biological organisms, is expected to elucidate functions and structures of genome, and to produce many new technologies in various areas such as medical treatment, preservation of environment, food production and so forth. Research on genome science can create brand-new industries based on such new technologies."^[51]
       Japan's new policy orientation presents a marked difference from that of U.S. postwar science policy which emphasized pure basic research.^[52] Academics in the United States, supported by the National Science Foundation, held it as a matter of "social contract" that the federal government supports pure basic research with full autonomy granted to researchers with an expectation that the benefits will diffuse broadly in society.^[53] The 1980s saw a growing criticism against this ideological interpretation forcing the science policy community to reopen a debate on science--the endless frontier.^[54] Under a confluence of economic and social pressures in the 1980s, U.S. postwar science policy began to emphasize the relationship of science to society, particularly, technological innovation.^[55] In Unlocking the Future: New Science Policy for the Next Century, the U.S. House Committee on Science implores, "Science must maintain a solid relationship with the society that supports it." The Committee goes on to stress, "Through better communication or something else, strong ties between science and the American people are paramount."^[56] The development of science policy in the United States and Japan during the last twenty years hints at the possibility that the U.S. and Japan may be converging toward the "use-inspired," Pasteur's quadrant.^[57] 
       The sustainability of Japan's new S&T policy depends on two contingencies. One is whether the government will keep its promise by increasing R&D investment as planned. The other is the effectiveness of the proposed administrative reform. As we discussed earlier, the Administrative Reform Council created by Prime Minister Hashimoto in 1996 put forward a radical restructuring plan in 1998. In July 1998, the Parliament passed the reorganization bill into law. We now turn to the institutional reform. 

CENTRALIZING THE POLICY PROCESS

At present, administrative reform is well on its way toward the target set by the Parliament: April 1, 2001. While the reform applies to the entire administrative state, the focus is on the S&T policy system and the national R&D enterprise. This includes the restructuring of the Prime Minister's Office, the merger of Ministry of Education  (Monbusho) and Science and Technology Agency (STA), and the reorganization of national research institutes and national universities. The reorganization of national universities is to be completed by April 1, 2003.^[58]
       Given the enormity of change, the full shape of the new enterprise will not be known until April 1, 2003. Even so, the principles underlying the reform have been clearly established--although the underlying motives are debated endlessly. In a typical Japanese style, some ministries are apparently ahead of schedule as if "the early birds catch larger worms." This makes it possible for us to draw the contours of the emerging structure. In this section, we describe the shape of the new system by weaving the reform principles central to R&D institutions. We discuss why these principles have been proposed as they have, how they are practically applied, and what difficulties may lie ahead. Again, the motives for government reorganization are complicated, and the reform objectives do not add up to a coherent set. The relevant reform principles include: 

A brief discussion is in order. Of these reorganization objectives, "downsizing government" is an element that complicates reform efforts. The rationale is that the government has grown excessively large and rigid. The policy argument is, "The system is expensive, inflexible, and inefficient."^[60] "Inflexible" is understandable, "expensive" is debatable. A closer examination will show that this objective is politically motivated. Japan's Statistical Yearbook 2000 estimates the number of employees of the central government in 1997 at 65 per 10,000 of population, and the employees of local governments at 259 per 10,000 of population.^[61]  The combined ratio is 323 public employees per 10,000. These ratios are not high at all when compared to the United States. In 1995, the U.S. Census Bureau estimated the employees of the federal government at 110 per 10,000, and the employees of state and local governments at 633 per 10,000.^[62] The combined ratio in the United States is 743 public employees per 10,000. Thus, the U.S. ratio is more than twice as large as in Japan. In addition, the U.S. ratio does not include postal workers, whereas the Japanese ratio includes them.
       The Obuchi government has decided to downsize government personnel by 25% effective April 1, 2001. The idea of personnel reduction goes back to Prime Minister Hashimoto. In 1996, Prime Minister Hashimoto proposed to reduce government personnel by 10%. This was in response to the rising government deficit and public distrust of the Nagatacho government. When Mr. Obuchi became Prime Minister in 1998, he promised to reduce government personnel by 20%. Mr. Ozawa, a coalition partner representing the Liberal Party, demanded yet a higher reduction of 25%. Obuchi caved in to the 25% target.
       The second, third and fourth objectives are largely a derivative of the first political-rational objective. It would be a mistake to dismiss the importance of these objectives just because they smack of politics. These objectives do have merits of their own since, if properly utilized, they can serve as a powerful means to promote efficiency and flexibility.
        The last two (5 and 6) are related to one another. These objectives are to strengthen the power of Prime Minister so he may be able to control the national bureaucracy. As we will discuss below, the Prime Minister of Japan needs a professional staff of his own. This objective reflects a sense within the Parliament (or the ruling coalition) that the elected body must do, or at least give an appearance of doing, a better job of controlling the national bureaucracy. As it stands, the bureaucracy is powerful yet disjointed. Insofar as S&T policy is concerned, a continuing bureaucratic rivalry between Monbusho and STA is an impediment to the efficient use of R&D resources.
        For the purpose of discussion, we present a rough sketch of the new governmental structure (Figure 3 below) expected to emerge under the reform policy. This diagram draws from our interviews and from government working papers.^[63] The boxes include only the components that are directly related to the S&T policy regime and R&D institutions. The shaded areas and dotted lines mark the structural changes targeted by the reformers, and we focus attention to these lines and boxes. First, we look at the changes introduced in the Office of Prime Minister, which include the Cabinet Secretariat, the Cabinet Office, and the National Council for Science and Technology.  Second, we look at the merger of STA with Monbusho. Third, we look at the changes in the national R&D system, first, with national universities and second, with national research institutes.   

Figure 4  
                     The Structure of Japan's New S&T Policy Process and Institutions

Institutionalization of the Prime Minister's Office

The 1998 reform act authorizes the Prime Minister to strengthen the structure of assistance to his office by creating a Cabinet Secretariat (CBS) and a Cabinet Office (CBO). The Cabinet Office includes a total of seven divisions, one of which is responsible for S&T policy. Each of these offices will be directed by a powerful policy council accountable to the Prime Minister.^[64] The council for science and technology is tentatively named the National Council for Science and Technology (NCST).
           To begin, it is important to note that these offices are statutory bodies amounting to an "institutionalization" of the Office of the Prime Minister. If the U.S. experience were any guide, this "institutionalized premiership" has profound implications for the future of Japan's government. Japan's parliamentary democracy is structurally different from the American system of  "separation of powers." And yet, the institutionalization of the Prime Minister's Office is similar to the Reorganization Act of 1939 in the United States in which Congress granted--reluctantly-- the President with authority to establish the Executive Office of the President. The rationale used for granting this reorganization authority was simple: "The President [needed] help."^[65] The rationale used by the reformers in Japan strikes the same cord: "The Prime Minister needs a powerful planning and coordinating body that shall provide direct assistance and support to him."^[66]
        What does the institutionalization of the Office of the Prime Minister imply? In the case of the United States prior to 1939, the President had very little staff support of his own within the White House. The presidents always borrowed staff from other executive departments.^[67] "The situation of Japan's Prime Minister has been no different," observed one interviewee.^[68] In fact, the situation facing Japan's Prime Minister at the present time is worse. This is a point not particularly well understood. Although there is a general belief to the contrary, Japan's Prime Minister is not the master of his own government. Given that the tenure of the Prime Minister is grievously short (two years or less), he has neither the "separate" power base (e.g., national election) nor institutional resources with which to run his own government, let alone the time to manage his government. Each Ministry is headed by a Minister (a member of the Ruling Party or the Ruling Coalition) whose tenure is no lengthier than that of the Prime Minister. Moreover, since the political system is a parliamentary democracy, the Prime Minister and his Ministers must also get involved in the parliamentary process. This means that the business of government is necessarily delegated to bureaucrats. Needless to say, the structure severely handicaps the ability of the Prime Minister to provide policy leadership. The institutionalization of the premiership is a belated response to this woeful condition.   

Cabinet Secretariat 

The Cabinet Secretariat to be established within the Prime Minister's Office is staffed primarily by political appointees. As the reformers envisioned, the Secretariat is to function as the "ultimate planning and policy coordinating body" that provides "direct assistance and support to the Prime Minister." Specifically, responsibility includes the planning and drafting of basic policies (external affairs, national security, fiscal management, macroeconomic policies, budget formulation, and personnel policy), and the coordinating of ministries. While no clear guidelines in the law dictate how this Secretariat would function, if the workings of the Executive Office of the American President is any guide, much will have to evolve by trial and error. At least from its description, the Secretariat appears to be some combination of the White House Office and the Office of Management and Budget. Perhaps the best analog is the Domestic Council under President Nixon. If this comparison has some validity, one might expect that the Secretariat may possibly develop into a powerful leadership tool for the Prime Minister. A caveat is that the design embraces a built-in instability owing to the political nature of the office, which might make it difficult to develop necessary professionalism. 

Cabinet Office of Arts, Science and Technology