After the developmental state: civil society in Japan.

The Japanese developmental state catapulted Japan into economic prominence. However, almost just as world attention focused on Japan's distinctive model, the era of the developmental state was drawing to a close. A generation of scholars has ably documented the story of Japan's developmental state by focusing on industrial policy. They chronicled how a strong bureaucracy buffered by insulation from politicians lay at the heart of the developmental state. As Joseph Wong points out in the introductory essay to this special issue, scholars have also argued that the developmental state contained within itself the seeds of its own dismantling. (1) Since the 1960s, formal powers had been stripped from the bureaucracy, leaving it increasingly dependent upon "administrative guidance" not legally enforceable. (2) By the late 1980s, the very success of the developmental state had eroded the powers of the bureaucracy to set industrial policy.

This article is organized around two central arguments. First, the article contends that civil society is an unrecognized element of the developmental state story--both in its success and its decline. A quiescent civil society sector (the organized nonstate, nonprofit sector) characterized the developmental state and was as central to the bureaucracy's political insulation as the distance from politicians. Moreover, the very success of the developmental state brought Japan to a new level of affluence and led to the increasing prominence of civil society organizations. Although small in scale compared to the United States, the activities of these groups are corrosive to political insulation.

The second argument relates directly to this point and has three parts. First, civil society organizations are proliferating in Japan. Second, the type of groups appearing are different, as Japan's civil society becomes more pluralistic, particularly with groups that emphasize their independence from the bureaucracy. Third, civil society groups are beginning to flex their muscles, particularly in forging new relationships with political parties (especially the Democratic Party of Japan [DPJ]) and in certain policy arenas (such as welfare). Without overstating the strength of civil society in comparative perspective, it is undeniable that civil society organizations are growing in number and influence in Japan.

This introduction previews my overall contention. In the succeeding sections, I briefly review the characteristics of the Japanese developmental state with special attention to the underanalyzed role of civil society. Then, I provide an overview of Japan's civil society before tracing out the changes in the size and policy influence of civil society in greater detail. The conclusion speculates on both the reasons behind the decline of the developmental state, as well as what we can expect from Japanese civil society organizations in the postdevelopmental state era.

The Rise and Decline of the Japanese Developmental State

Tracing the idea of the developmental state back to its locus classicus is a useful starting point. In his influential MITI and the Japanese Miracle, Chalmers Johnson somewhat reluctantly outlines four elements of the Japanese model of the developmental state (Johnson 1982). These are: (1) an elite bureaucracy; (2) a political system in which the bureaucracy rules and the legislative branch is restricted in power; (3) "the perfection of market-confirming methods of state intervention in the economy"; (3) and (4) a "pilot organization like MITI." (4)

The methods of state intervention clearly constitute an integral part of this model. Previous analyses have focused on the relationship between firms and the bureaucracy. (5) Of course, by the late 1980s, too, companies were able to obtain their own financing, either internally or from overseas. Companies were also no longer reliant on the bureaucracy for access to overseas technology, which they could either purchase for themselves or, more commonly, create with in-house resources. Japan had incubated strong, globally competitive companies. Other factors had also diminished the scope for industrial policy, including the globalization, growth of Internet technology, and foreign pressure to economically liberalize. Increasingly, the emphasis shifted away from industry-specific policies to economywide reforms. (6)

However, earlier analysis of the developmental state in Japan has primarily focused on the relationship between politicians and bureaucrats. Indeed, the debate on which group really held power in Japan was so consuming that Ethan Scheiner, Michio Muramatsu, and Ellis Krauss half-jokingly prefaced an important rethinking of the issue with the fear that any further discussion of the topic could evoke a primal scream from inundated readers. (7) Yet it is my contention that the political insulation of the heyday of the developmental state has been stripped from Japan's bureaucracy. I focus on three factors. (8) First, scandals have erupted in the previously immaculate bureaucracy itself, as discussed below. Second, savvy politicians have grown increasingly aggressive in policy arenas, and the system of political insulation has therefore broken down. (9) Third, the development of civil society organizations has come to represent a further threat to bureaucratic autonomy. This is all to say that the diminishing capacities of the insulated developmental state have opened up new political opportunities for Japan's once weak advocacy sector to more strongly assert itself in national affairs since the 1980s.

Political Insulation

The political insulation demanded in a developmental state also depends upon insulation from civil society organizations or interest groups. Even though this insulation was never complete (which might be either impossible or dangerous), by and large most interest groups and civil society organizations were constrained in influence over policymaking in Japan's developmental state. (10) Although producer and industry groups were relatively favored, in critical respects the political insulation of a bureaucracy is predicated in some sense on a weak civil society sector. (11) As argued below, Japanese civil society groups have grown in strength and number, but Japan's professionalized civil society sector lags in influence compared to, say, that in the United States, where civil society groups are particularly powerful.

Measuring civil society, or the strength of civil society, is a notoriously problematic exercise. Certainly an argument can be made that Japanese civil society was robust even during the heyday of the developmental state, especially in terms of social movement mobilization. In 1960, for example, student demonstrations, with a backbone of well-organized student groups, against the extension of the U.S.-Japan Security Treaty rocked the nation. Proportionally, as many Japanese participated in Japan's environmental movement as Americans participated in the U.S. environmental movement. (12) Moreover, local civil society groups in Japan such as neighborhood associations flourished throughout this period.

I have argued elsewhere that Japan's civil society sector cannot be thought of simply as weak; it is a mistake to consider Japanese civil society unidimensionally. Rather, a dual structure radiates throughout the subsectors. (13) Japan has a plethora of small, local groups and a paucity of large, professionalized groups. Neighborhood associations (NHAs) are quite plentiful in Japan, for example, but large advocacy groups are especially scarce. These dimensions are made clear by an overview of Japan's civil society. Table 1 and Figures 1 and 2 illustrate that Japan certainly does have many civil society groups; what Japan does not have, in comparative perspective, is many employees working in civil society groups. This means that there are many groups with few or no employees and few groups with many employees. Having many groups, but not many employees, is a defining feature of Japan's dual structure in civil society.

[FIGURES 1-2 OMITTED]

Figure 1 shows the proportion of the total workforce employed by civil society organizations. (These figures exclude workers in education, health, and social services, whose figures vary widely because of state policies, but they do include workers in all other civil society groups.) This proportion represents the professionalization of civil society organizations. Proportionally Japan's 73,500 civil society professionals are fewer than half the number of the next lowest nation (Germany) and less than a third of the average for the developed nations in the figure.

Figure 2 shows the number of employees working in public-interest legal persons (PIPs)--the legal category of civil society groups that includes the largest and wealthiest civil society groups. This is the category in which we would expect to find Japanese civil society's largest employers. Most groups, though, have only a handful of employees. Only a few thousand civil society groups in Japan employ more than ten people.

Figure 3 is based on the Japan Interest Group Survey (JIGS) data set of Japanese interest groups (see appendix) and reveals that large advocacy groups in Japan are far less common than those in the United States.

[FIGURE 3 OMITTED]

Besides the small size of the professional civil society sector in general, another factor intrudes in our consideration of the ability of civil society organizations to influence policy outcomes in Japan. Many groups face constrained independence from the state due to institutional factors such as legal regulations and bureaucratic practices including amakudari (literally, "descent from heaven," referring to the employment of retired bureaucrats in new posts, typically in areas they once regulated). Put simply, it is hard for independent groups to grow large in Japan and just as hard for large groups to remain independent. This is particularly relevant as we consider the relationship of civil society organizations to the developmental state and also again as we consider the reasons behind the spurt in civil society just as the bureaucracy's star wanes.

The mechanics of this pattern can be traced to an institutional arrangement that provides significant monitoring powers (reporting and investigating) and sanctioning powers (various punishments including dissolution of the group) in the hands of a single bureaucratic ministry or agency. Even in the abstract, it is easy to see that if a single agency grants a group permission to form, monitors it, has the ability to punish it, and can even dissolve the group entirely, often without effective legal challenge, then that agency holds significant power over the group. In Japan, the PIP has reporting duties to the competent ministry, which retains the power to investigate the group or even to revoke the PIP's legal status. Moreover, the concomitant tax benefits are not so generous as in other industrialized democracies.

Even worse, the bureaucrats have insisted on continuing "administrative guidance." (14) Backed by sanctioning power, this administrative guidance forces licensees to comply with bureaucratic demands and impairs the independence of the civil society sector. It has been employed in such a heavy-handed way that many observers regard the Social-Welfare Legal Persons, for example, as little more than cheap subcontractors for the government without the independence necessary to qualify them as true "citizens groups" (shimin dantai). Iriyama Akira, director of the Sasakawa Peace Foundation (a PIP), puts it well when he says: "Even those like us who make it through and get permission have to suffer from very severe control and guidance from authorities. If I start to talk about the notorious administrative guidance, it'll take days." (15) Despite the great logistical problems it creates, foreign groups such as the Asia Foundation sometimes choose not to become a PIP precisely to avoid bureaucratic interference. In a nationwide Economic Planning Agency survey of Japanese nonprofit organizations (NPOs), the most common reason cited for not applying for legal status was that accounting and finance reporting requirements were too onerous (61 percent of the groups listed this reason); the third most common reason was the fear that the objective of the NPO, or the content of its activities, could be controlled by the bureaucrats (45 percent). (16) A PIP must submit a report of its annual activity, list of assets, accounts of changes of membership, and financial statements for the past year, as well as planned activity reports and budget estimates for the coming year.

In Japan, it is the permitting agency that is empowered to investigate the PIP. Moreover, the permitting agency can issue directives to the PIP that, if not obeyed, can result in the PIP's dissolution. The agency, moreover, can make on-site inspections and audits. Civil Code article 68(1)(iv) provides that PIPs are to be dissolved if the authorizing agency cancels its authorization for the establishment of the corporation. Article 71 states that the authorizing agency may cancel its approval if the PIP has engaged in activities outside its purposes as defined in the articles of association, has violated the conditions under which it received approval for establishment, or violates supervisory orders issued by the authorizing agency. The Civil Code Enforcement Law, article 25, requires an inquiry by the authorizing agency and also requires the agency to indicate the reasons for the cancellation to affected parties, who then have the right to a legal proceeding and appeal. One notes that a cancellation of authorization is interpreted as a response to changed circumstances, not as a mistake by the authorizers as to whether the PIP was in the public interest to start with. Despite the possibility of appeal, the legal deck is stacked in favor of the authorizing agency--in part because of the considerable discretion attached to defining the public interest. (17) It is not surprising, then, that advocacy groups are particularly small in Japan. They average only 3.4 employees and expenditures of [yen] 36.12 million (U.S.$361,200) a year, only 22.7 percent of the average for all nonprofits in Japan. (18)

As a result, Japan's civil society has had a constrained voice in policy formation. Figure 4 shows how research featured in newspapers in Japan and the United States comes from widely different sources. The most important element, for our purposes, is that civil society is less able in Japan to get its research findings into the public domain or public discourse.

[FIGURE 4 OMITTED]

It is intuitive that there should be a strong causal link between group size and the group's ability to attract media attention. An analysis of the JIGS data set, which includes variables on group size and the number of times the group has been featured in the media, confirms this intuition. Figure 5 presents these results, which show a strong correlation between group size (in terms of numbers of permanent employees) and the media coverage the group can attract.

[FIGURE 5 OMITTED]

Japanese civil society groups have not been prominent in public debates. Although this does not inevitably mean they are not powerful in shaping public opinion or affecting policy outcomes, I do contend that this is the case outside of the producer groups mentioned above (and much evidence supports this view). Consider policy debates on moral issues. In many countries, including the United States, civil society groups, in the form of religious groups, heavily influence the public debate on issues seen as having a moral dimension, such as abortion or euthanasia. This is not true in Japan despite the large number of religious groups (see Table 1). Another example is policy change on women's issues. Essentially toothless in practice, the Equal Employment Opportunity Law (EEOL) of 1986 still nonetheless guaranteed gender equality in the workplace. Although one might have expected women's groups to have taken the lead in forcing this measure onto the policy agenda, it was in fact actors in the bureaucracy who pushed the issue, bringing Japan into compliance with the United Nations antidiscrimination pledge signed by Japan during the 1975 Year of the Woman.

For an environmental example consider the sweeping package of antipollution measures that gave the 1971 Diet the label the "Pollution Diet." Although thousands of environmental groups sprang up between 1965 and 1975 in Japan, they did not play a role in getting the antipollution measures on the agenda. These groups remained local and disaggregated. Similarly, it was not Japanese environmental groups but the bureaucracy that took the lead in the passage of the Kyoto Accord on global warming. Japanese civil society groups have also been weak on issues ranging from whaling to human rights (e.g., fingerprinting of Korean residents). (19)

Growth of Civil Society Organizations

The preceding section demonstrates how Japanese civil society organizations are underprofessionalized in international comparison and that this has constrained their ability to influence public discourse on important social issues. However, if we set aside for a moment comparative statics and instead take a longitudinal view, we see that Japan's civil society organizations have increased in number and size since the 1960s. Japanese activists often point to the 1995 Kobe earthquake as a seminal event catalyzing the growth of Japan's civil society. The tragedy contributed to the growth of Japan's civil society organizations for a number of reasons, (20) but Japan's civil society has been growing before and after the disaster for reasons discussed below.

There has been a steady increase in many legal categories of PIPs. In addition to this expansion, Japan's civil society has also diversified. For example, Yutaka Tsujinaka, Japan's leading authority on civil society, analyzes Establishment Census data from the Management and Coordination Agency to argue that the number of associations in Japan more than doubled from 1960 to 1996, with the greatest increases coming in areas outside of business and labor organizations. (21) Tsujinaka sees this as evidence of the growth and the pluralization of Japan's civil society.

Most of the growth has been in "new" types of organizations, known variously as "citizens groups" or NPOs, both of which are Japanese parlance for groups that specifically prize their independence from the bureaucracy, rather than referring to a particular type of group; precisely the kinds of groups most relevant for a discussion of postdevelopmental state Japan. To be sure, many traditional associational forms stagnated. The number of religious organizations dropped by nearly a third from 1987 to 1997. During the same decade, however, the number of nongovernmental organizations (NGOs) headquartered in Japan that registered with the United Nations nearly doubled. (22) A 2000 government survey found that nearly half of extant citizens groups had sprung up in the 1990s (42 percent), and more than a quarter had sprung up in the 1980s (25.4 percent). Simple arithmetic yields the result that two out of three citizens groups operating in 2000 had formed in the previous score of years. (23) Citizens groups reported the lowest levels of trust of the bureaucracy of any category of group in JIGS, another mark and cause of their desire for independence.

The quintessential "new" type of civil society organization is, however, the NPO Legal Person. This category, after all, did not even exist until 1998, as discussed below, and was designed to be free of the bureaucratic domination that plagued old-line civil society groups. Figure 6 charts the rapid growth in NPO Legal Persons since the law's promulgation in 1999. The bottom line is that we are witnessing not only pluralization but also transformation; growth is particularly rapid for civil society groups that prize their independence from the bureaucracy, and little could be more debilitating to the developmental state.

[FIGURE 6 OMITTED]

The increase in the number of civil society organizations in Japan reflects two distinct types of factors: common international factors, and factors specific to Japan. In part, this growth trajectory reflects worldwide trends and is caused by economic growth, by telecommunications spread, and through the emergence of new norms. In other words, civil society organizations have proliferated not only in Japan but everywhere in the world, particularly in the advanced industrialized democracies. Although I will outline Japan-specific factors that have promoted the development of new and more civil society organizations, it is clear that to understand the growth of civil society worldwide we also need general explanations.

Economic Development

Researchers have demonstrated a clear correlation between the level of economic development and the development of civil society organizations. At the individual or national level, higher income and higher education levels both correlate with increased participation in (for individuals) or numbers of (for states) civil society organizations. Indeed, it is precisely because of this general relationship that we can distinguish Japan as such an outlier, as in the preceding section. However, this also means that to some extent, once again, the developmental state brought about its own undoing. Japan increased in affluence thanks to the success of the developmental state in fostering economic growth. The very spread of material affluence (and educational opportunity), arguably the reason for being of the developmental state, itself contributed to the spread of civil society organizations. (24)

International Factors

International factors were also at play. As civil society organizations proliferated and strengthened in one state, they frequently sought allies in other states. Sometimes they simply open up a branch office. Greenpeace and Amnesty International, both originally British organizations, are by now so well established in the United States that it is doubtful that most Americans recognize their origin. Civil society organizations also sought to plant their seeds in Japan but often found the ground (paralleling the experience of corporations) not particularly fertile, in part for the legal reasons discussed above. Greenpeace, for example, is the only nonnative environmental group in the top ten in membership in Japan (at tenth, with 5,400 members). (25) Civil society groups did achieve success, however, through another, less obvious, means: norm change. Civil society groups have succeeded in establishing new international norms, even in notoriously state-dominated areas such as security when the International Campaign to Ban Landmines (ICBL) succeeded in establishing an international norm against the use of antipersonnel landmines.

More and more transnational civil society organizations are active. One reflection of this is the increased participation of NGOs at UN conferences. At the 1975 Mexico City first UN Conference on women, 114 civil society groups gained access to the official conference, and 6,000 people attended the NGO forum held in parallel with the official UN activities. A decade later, at the UN Decade on Women Conference in Nairobi, 163 civil society groups gained official accreditation and 13,500 attended the parallel forum. Ten years after that, at the 1995 Fourth World UN Conference on Women, in Beijing, 3,000 civil society groups were accredited and 300,000 people attended the NGO forum. The human rights field saw similar gains. The 1968 Tehran International Conference on Human Rights admitted fifty-three civil society groups with consultative status. The 1993 Human Rights Conference in Vienna had 248 civil society groups with consultative status and 593 as participants. Among all the transnational civil society groups, environmental groups have increased with the most striking rapidity. Only 1.8 percent of transnational NGOs were environmental groups in 1953, but by 1993 that proportion had risen to 14.3 percent. (26) Fewer than 300 civil society organizations attended the 1972 Stockholm Conference on the Environment. Twenty years later, 18,000 attended the UN Conference on Environment and Development in Rio de Janeiro (the Earth Summit), 1,400 of them as official registrants. (27)

This blossoming participation alters states' understanding of standard operating procedures. One far-reaching norm that civil society groups imposed was that of civil society group participation in international aid and development programs. It is fascinating to see how the establishment of these international norms reflect back on and thus alter domestic policies in Japan. Kim Reimann argues persuasively that Japanese government officials were handicapped by the absence of strong NGO partners in clear violation of "standard practice" and moved to foster partner groups. As she writes, "Because Japan was an outlier among industrialized nations in the 1980s--it was the only OECD country that did not have a visible NGO component.... [I]t was under pressure to make changes that would somehow bring it in line with international practices and standards." (28) Japan's inhospitable institutional framework accounted for the dearth of large international development groups. In Reimann's account, it is striking how international norms and participation in international conferences moved Japanese officials to, at least in this one area, encourage the growth of civil society organizations in the 1990s. A variety of government initiatives fostered new groups, and Japan's international development nongovernmental organization (IDNGO) sector grew from a handful of groups in the 1980s to over 300 today. (29)

Another unexpected international influence came from the massive expansion of Japan's foreign direct investment (FDI) into the United States after the appreciation of the yen in the mid-1980s. It may seem surprising that Keidanren, the big business peak organization, lobbied for loosening regulations on civil society organizations and the NPO Law in 1998. In fact, Keidanren's interest in civil society grew out of the late 1980s and resulted in the formation of the Social Contribution Group of the Social Affairs Bureau. With increased Japanese direct investment in the United States, Japanese corporations were often faced with request for donations or other support from U.S. NGOs. The corporations were often not sure at first how to deal with these kinds of groups, which were not nearly so active in Japan. Keidanren consequently set up a study group on the NGOs, and it later formed the 1 Percent Club (for donations) in early 1990. Keidanren went on to found its Social Contribution Promotion Committee in May 1990. Then, in April 1991, the Social Contribution Group was formed. In the aftermath of the Hanshin earthquake, this group was active in coordinating relief donations of money and material from corporations, channeling them to NPO and volunteer groups. This group, somewhat isolated though working closely with civil society organizations, became an advocate for the views of the civil society groups within Keidanren. They also had wide scope for this advocacy, as they were the acknowledged experts on a subject that had hitherto attracted little attention within Keidanren. (30) International involvement, this time through FDI, had again come back with unexpected consequences for the promotion of Japan's civil society organizations. In both instances--Reimann's argument above and Keidanren here--international involvement served as a transmission belt for ideas about reworking the relationship between state and society. This impact occurred precisely because of the international growth of civil society.

Domestic Factors

This general international trend played out in Japan, as is to be expected in any country, through a variety of distinct features. It is as important to note that the growth of civil society in Japan is consistent with established patterns in advanced industrialized societies as it is to point out that for its economic level it remains underprofessionalized. To move forward with the argument, however, in Japan, three domestic factors account for the continued expansion of civil society organizations in the 1990s.

First is regulatory change. The 1998 NPO Law liberalized conditions under which groups could form and operate in Japan. In 2001, revisions of the tax code created a subcategory of these groups with tax benefits for charitable contributors. (31) The pace of these regulatory changes is accelerating, and new tax provisions have already been revised. More important, in 2002 the government began holding discussions with an advisory council (shingikai) of academics and NPO activists on a wholesale reform of the legal framework for civil society groups. The government released its interim report at the end of March 2004. (32) The rise of the NPO Legal Persons was highlighted above. A second factor is Japan's social and demographic change. (33) As the aging society puts pressure on Japan's limited welfare state, civil society organizations are seen as cheap and essential vehicles to provide needed social services. These newly attractive civil society organizations will likely be increasingly promoted. Indeed, this was one factor behind the passage of the NPO Law. Third is a breakdown in public trust of the bureaucracy. As the state bureaucracy is seen as less trustworthy, there will be a tendency to look to nonstate actors for solutions to social problems. A series of scandals struck at the essence of the developmental state: the elite bureaucracy. They were insidiously damaging in several ways. For one, the scandals undermined the reputation of the bureaucrats themselves. After all, the Japanese bureaucracy was widely seen as "deserving" their power for two reasons. The bureaucrats comprised the best and brightest, but they were also seen (and saw themselves) as serving a national mission above petty self-interest. The scandals revealed that even the economic bureaucrats were capable of acting venally. If the bureaucrats were not acting in the national interest, many Japanese asked themselves, why let them run the country?

Data from public opinion surveys reflect these moods. (34) A 1994 Yomiuri Shimbun poll found citizens mixed in their views of the bureaucracy. A substantial 27 percent found no "positive aspects" in the bureaucracy. Yet a remarkable 10 percent of those polled thought that there were "no problems" with the Japanese bureaucracy. Twenty percent described bureaucrats as "excellent," and 17 percent found them "obnoxious." The most common criticism to emerge of those unhappy with the bureaucracy was the "cozy relationship" between politicians and businesses (34 percent), reflecting the scandals then emerging (which only grew worse after the poll was taken). (35) Not surprisingly, Japanese were angered by the scandals, and suspicion deepened. A 1998 Jiji Press poll found that the most common reaction to the scandals was anger (58 percent); the second most common (37 percent) reaction was that "there must be many more similar scandals that have not been revealed yet." Only a tiny minority (3 percent) of Japanese felt that "only a limited number of officials are doing such things." (36) Clearly, mistrust of the bureaucracy ran rife in the public. By 1998, a Nihon Keizai Shimbun poll found that an overwhelming majority (76 percent) supported a law to regulate ethics of civil servants, on the grounds that "we can't count on civil servants acting in good faith." (37)

In turn, this loss of public confidence further undermined the bureaucracy by eliminating an important perk of bureaucratic service. Bureaucrats toiled long hours and chose not to enter top private firms, where their salaries would certainly be higher, for the power, surely, and also out of a sense of national service, but many joined, too, for the social respect bureaucrats were accorded. A career as a bureaucrat began to seem less attractive to new college graduates. Moreover, bureaucrats themselves were shaken by the changes. The esprit de corps that formed an important part of the success of the Japanese bureaucracy began to wither. Surveys of the bureaucrats themselves reveal that "bureaucrats have lost self-confidence." (38) By 2002, bureaucrats' perceptions of their own influence had plummeted dramatically. Only 21.75 percent of bureaucrats in 2002 claimed the bureaucracy had the greatest impact in determining Japanese policy. (39)

Linking Civil Society and Politics

The preceding section contends that civil society organizations are expanding in number in Japan and that particularly prominent in this rise are groups that steadfastly maintain their independence from the bureaucracy. This does not necessarily mean that such groups are apolitical. In fact, civil society organizations are forging new relationships with political parties. This symbolizes the changed circumstances of the postdevelopmental state. The ruling conservative Liberal Democratic Party (LDP) is recasting its relationship to its traditional supporter groups, such as the Japan Medical Association, agricultural cooperatives, and construction industry associations. (40) In fact, one of the reasons behind the LDP's interest in liberalizing the regulatory regime for civil society organizations is its desire to use NPOs as vote collection vehicles. Given the declining power of other organizations to deliver votes and the need for a plurality to win a single-member district, NPOs seem attractive--especially as their appeal is greatest among the independent voters who are increasingly crucial to winning elections. On March 30, 2004, the LDP subcommittee (senmon iinkai) on NPOs hosted a symposium as part of an increasing effort to round up the NPO vote for the LDP. The chairman, Yoshio Yatsu, conceded that it would be disingenuous to claim that the party was not hoping to gain votes with these tactics. (41) LDP politicians have long shown a tendency to favor groups as means to gather votes, of course, and they seem to instinctively see NPOs in that light. However, there will be limits to how effectively NPOs will gather votes for the LDP. For one thing, few NPOs support the LDP. Also, the NPO Law itself (due to provisions insisted upon by the LDP) places prohibitions on political activities. More important, NPOs have yet to demonstrate the ability to deliver votes in the manner of other organizations such as Soka Gakkai or labor unions. The experience of Shoji Takahira of the Kodomo Gekijo National Center (an organization promoting the performing arts among children) should give LDP officials caution. The organization had affiliates in every prefecture and boasted nearly 400,000 members. Shoji Takahira campaigned in the 2001 House of Councillors election as a national Proportional Representation candidate for the DPJ but gained only 11,175 votes. This comes in stark contrast to the proven (if perhaps declining) vote-gathering ability of other groups. In the proportional representation section of the House of Councillors' election, the postmasters' interest group delivered 478,000 votes in 2001 and 282,000 in 2004 for its candidates. The Japan Medical Association's political group delivered 227,000 in 2001 and 250,000 in 2004, while for the DPJ the local government union delivered 216,000 votes for its candidate in 2001 and 167,000 in 2004. (42)

The Clean Government Party (CGP) has since its inception in 1967 been closely associated with the civil society organization that gave it birth, the lay Buddhist movement organization Soka Gakkai. It has also nurtured close links to welfare groups, in line with CGP's carving out of the welfare space in the policy arena. Until its demise as a significant party, the Japan Socialist Party (JSP) worked closely with labor unions. It was the birth of new political parties, however, after the split of the LDP in 1993 that is most important for this analysis.

I have argued elsewhere that small, left-leaning parties pushed hard for the passage of the NPO Law in an attempt to secure a viable base for the proportional representation section of the new Japanese electoral system. (43) The successor to these parties and to the JSP is the Democratic Party of Japan, now the major opposition party in what has become a party system dominated by the LDP and DPJ with the significant though smaller presence of the CGP. The DPJ has been active in supporting the new type of civil society organizations, seeing them as natural allies. For example, the DPJ manifesto, which was issued to much ballyhooing in the run-up to the November 2003 lower house elections, contains a section on NPOs and their role in society. Specifically, the manifesto calls for a reform of Article 34 and a vast expansion of tax benefits to nonprofit organizations (an increase of 48,000 percent to cover 60 percent of all NPOs), increasing to 20,000 the number of NPOs providing after-school care, and (rather more vaguely) using NPOs to lower unemployment. (44) NPOs are also increasingly a training ground for DPJ Diet Members and staffers. (45) After the November 2003 election, twelve of the DPJ's 178-strong House of Representatives delegation had a background in civil society groups (including NPOs, NGOs, and citizens movements). (46) This was a dramatic increase, as most (seven) of the Diet Members had won election that November. Three others were newly minted second-term Diet Members. The development of NPOs as training-grounds for DPJ politicians is new and has important implications both for NPOs and political parties.

Moreover, support for these types of groups goes hand-in-glove with the DPJ manifesto's avowed aim of reducing bureaucratic power. This, after all, is the party that issued a major policy document titled "Showing the Bureaucrats Who Is Boss." Thus, a major political party is not only pushing for the expansion of civil society organizations but doing so pointedly at the expense of the bureaucracy. This surely provides an eloquent testament to the fundamental transformation of the political insulation for the developmental state.

Unlike in the past, civil society organizations are more prominent in the media today. At least, groups are mentioned more often in the Asahi Shimbun newspaper. From 1987 to 2000, the number of times that the phrase "civil society" (shimin shakai) appeared over the calendar year rose from 14 to 99. The term "NGO" rose from 38 appearances to 683, while NPO rose from 0 to 381. By contrast, references to labor unions declined slightly, from 394 to 373, while women's groups (fujin dantai) dipped from 30 to 5. Moreover, there is evidence that civil society groups, particularly in the welfare arena, have been able to influence the direction or implementation of policy. (47) In contrast to the external forces that impelled the 1986 passage of the Equal Employment Opportunity Law, women's groups played a more important role in the 1997 revision of the EEOL that legally required sexual harassment prevention at the workplace. (48)

Conclusion

In many ways, the success, not the failure, of the developmental state brought about its demise. Successful industrial policy nurtured powerful companies no longer reliant on government support. Long-term economic growth bolstered an incumbent party whose politicians came to develop the ability and inclination to intervene more aggressively in policy matters. The same economic growth, in Japan as elsewhere, goes hand-in-glove with an increasingly large civil society sector, which strips the bureaucracy of its political insulation. Perhaps the achievement of catch-up--the overriding imperative of Japan's postwar developmental state--should bring about its dismantling. After all, its purpose has been served. And many might argue that the developmental state by its very nature is more adept at catch-up than at get-ahead, and so even in pure economic terms the arrangement has outlived its usefulness. A social perspective could emphasize that the postwar consensus on the primacy of economic growth no longer holds and that Japan now needs to reinvent its social compact. Japanese today will not agree to give bureaucrats the leading role in society in order to foster economic catch-up.

As with the other elements of the developmental state, civil society also is changing. I hasten to add that no one should expect a sea change overnight. Civil society groups require time in order to build expertise, legitimacy, and organizations. Figure 7 shows that most large groups in the United States and in Japan were founded decades ago; it requires time to amass resources.

[FIGURE 7 OMITTED]

Moreover, it would be a mistake to ignore the continuities from and the legacies of the developmental state. Bureaucrats remain relatively quite powerful in Japanese policymaking, professional civil society remains weak, and industrial policy activism has by no means disappeared. However, it is undeniable that Japanese politics in the early 2000s is in many ways clearly distinct from the 1960s, the heyday of Japan's postwar developmental state. These substantial changes are irreversible; there can be no return to the status quo ante. Returning to Johnson's four defining characteristics of the developmental state will help us take stock. The elite bureaucracy remains elite, but it has been tarnished by scandal. The legislative branch has asserted itself increasingly in policy. The "perfection of market-confirming methods of state intervention in the economy" seems dubious in the world of global sourcing of capital or production and the Internet. Perhaps most telling of all, MITI is not even MITI any longer. In 2000, it officially changed its name to the Ministry of Economy, Trade, and Industry (METI).

An underprofessionalized civil society sector in Japan provided the developmental state critical political insulation. Japanese civil society after the developmental state has grown in size. An equally important change in the profile of Japanese civil society is the shift to groups seeking to limit bureaucratic influence over their workings; groups that steadfastly maintain their independence from the bureaucracy have been especially prominent in the surge in civil society. Moreover, the underlying factors that spurred the growth of these groups are likely to increase, not diminish; abetted by the heightened pace of legislative reform, our expectation should be for this growth to continue for a while. In an encouraging international environment, civil society groups have the potential to forge closer links with political parties, exert greater influence over public policy, and amass critical organizational resources. Although the above caveats still apply, we should expect to find that linkages to political parties, coupled with an increase in the size of especially independent-minded groups, position civil society organizations to play a more dynamic role in politics and policymaking in Japan after the developmental state.

Appendix: JIGS Data Set

Analysis of the Japan Interest Group Survey (JIGS) data set is used in this article. The Cross-National Survey on Civil Society Organizations and Interest Groups, a team of six researchers led by Yutaka Tsujinaka and based at Tsukuba University in Japan, collected the JIGS data set. The team conducted an extensive survey of more than 1,600 associations in Tokyo and also in Ibaraki Prefecture in 1997, involving thirty-six questions and 260 subquestions. Assisted by Japan-based experts and local collaborators, the team has also conducted similar surveys in the United States, South Korea, Germany, and China. The team utilized random sampling of telephone book directories (in Japan, the NTT telephone book). Not all groups necessarily have their own telephone line, but this allows the group to also sample groups without legal status. In this way, JIGS is more comprehensive than government data and catches many groups that would otherwise be uncounted. I am grateful to Professor Yutaka Tsujinaka for generously sharing this data set with me.

Table 1 Civil Society Groups in Japan by Type and Number of Groups

Type of Group                                        Numbers

Group A: With Legal Status (a)
  Education Groups                                     6,155
  Social Welfare Groups                               13,000
  Religious Groups                                   183,894
  NPO Legal Persons                                    7,634
  Foundations (b)                                     25,927
  Medical Groups                                      22,838
  Cooperatives                                        23,718
  Political Groups                                    72,796
  Think Tanks (not counted elsewhere)                    449
  Neighborhood Associations (with legal status)        8,691
  Subtotal                                           412,523
Group B: Without Legal Status
  Neighborhood Associations                          292,227
  Children's Groups                                  130,000
  Elderly People's Groups                            150,000
  Other Civic Groups                                 598,000
  Voluntary Groups with Offices                        4,200
  Subtotal                                         1,212,227
  Total                                            1,624,539

Sources: Yutaka Tsujinaka and Hiroki Mori, "Gendai nihon no rieki
dantai: katsudou kuukan betsu ni mita rieki dantai no zonritsu
koudou youshiki" [Survey of contemporary Japanese interest groups
by area], Senkyo 51, no. 4 (1998): 4-15: Naoto Y. Yamauchi, Za
non-purofitto ekonomii [The nonprofit economy] (Tokyo: Nippon
Hyouronsha, 1997); Robert Pekkanen; "Molding Japanese Civil
Society." In Frank J. Schwartz and Susan J. Pharr, eds., The
State of Civil Society in Japan (Cambridge: Cambridge University
Press, 2003): Chikio Hayashi and Akira Iriyama, Koueki houjin
no jitsuzou [The real face of public interest legal
persons] (Tokyo: Diayamondo, 1997); Japanese government documents.

Notes: (a.) Gaining legal status means becoming a "legal person"
under Japan's Civil Code legal system. Several varieties of
"legal person" exist as categories under which various
groups may incorporate.

(b.) This category includes two legal groupings: zaidan houjin
and shadan houjin.

For comments on earlier drafts of this article I would like to thank Akihiro Ogawa; all participants at the conference "After the Developmental State," Seoul, Korea, March 2004; the editor, Byung-Kook Kim; and anonymous reviewers of Journal of East Asian Studies. I am grateful to Jaeyoung Choe for his help familiarizing me with the JIGS data set. I am indebted to Joseph Wong for his comments and direction in shaping this article.

Notes

(1.) Peter Evans, Embedded Autonomy (Princeton, N.J.: Princeton University Press, 1995).

(2.) Chalmers Johnson, MITI and the Japanese Miracle (Stanford, Calif.: Stanford University Press, 1982).

(3.) Johnson, MITI and the Japanese Miracle, p. 317.

(4.) Johnson, MITI and the Japanese Miracle, p. 319. Although I will not treat the subject at length here, Johnson's magnum opus spawned a veritable cottage industry of responses. Indeed, it has been remarked that a generation of Japan political scientists earned their jobs by responding to Johnson. Richard J. Samuels offered one of the major responses, stressing the reciprocal rather than unilateral nature of bureaucrat-business cooperation: Richard J. Samuels, The Business of the Japanese State: Energy Markets in Comparative and Historical Perspective (Ithaca: Cornell University Press, 1987). For reviews of the literature, see Saadia Pekkanen, Picking Winners? TIPs from Postwar Japan (Stanford, Calif.: Stanford University Press, 2003), and Gregory W. Noble, "The Japanese Industrial Policy Debate." In Stephan Haggard and Chung-in Moon, eds., Pacific Dynamics: The International Politics of Industrial Change (Boulder: Westview, 1989).

(5.) E.g., Samuels, The Business of the Japanese State.

(6.) Mark Eider, "METI and Industrial Policy in Japan: Change and Continuity," The Japanese Economy 28, no. 4 (2000): 3-34, 4-6.

(7.) Ethan Scheiner, Michio Muramatsu, and Ellis S. Krauss, "incentives, Institutions, and Bureaucrat-Politician Relations in Japan." Paper presented at the annual meeting of the Association for Asian Studies, San Diego, Calif., 2004.

(8.) In this way, my analysis differs in scope from T. J. Pempel's Regime Shift: Comparative Dynamics of the Japanese Political Economy (Ithaca: Cornell University Press, 1998). Pempel's major investigation of the transformation of the Japanese polity is broader in focus, including "socioeconomic coalitions, political institutions, and public policy profiles" as well as international factors (p. 14).

(9.) Seizaburo Sato and Tetsuhisa Matsuzaki, Jiminto Seiken [LDP rule] (Tokyo: Chou Koronsha, 1986); Takashi Inoguchi and Tomoaki Iwai, Zoku giin no kenkyuu [Research on policy tribe Diet members] (Tokyo: Nihon Keizai Shimbunsha, 1987).

(10.) Classic treatments of constrained interest group politics include T. J. Pempel and Keiichi Tsunekawa, "Corporatism Without Labor: The Japanese Anomaly." In Philippe C. Shmitter and Gerhard Lehmbruch, eds., Trends Toward Corporatist Intermediation (Beverly Hills, Calif.: Sage, 1979). Michio Muramatsu and Ellis Krauss, "The Japanese Political Economy Today: The Patterned Pluralist Model." In K. Yamamura and Yasukichi Yasuba, eds., The Political Economy of Japan, vol. 1 (Stanford, Calif.: Stanford University Press, 1987).

(11.) Typically, these favored groups were economic interest groups (employers groups, agricultural groups, industry associations). As such they are conceptually distinct--either excluded from civil society as market organizations or forming a distinct subgroup within civil society. Under the latter conception, Japan's civil society could be said to have a triple structure: (1) strong local grassroots groups and (2) weak professionalized groups except for (3) strong economic interest groups. The leading Japanese scholar of civil society includes business organizations in his conception of civil society and characterizes Japan's civil society as exhibiting "business supremacy"; Yutaka Tsujinaka, "Japan's Civil Society Organizations in Comparative Perspective." In Frank J. Schwartz and Susan J. Pharr, eds., The State of Civil Society in Japan (Cambridge, U.K.: Cambridge University Press, 2003). This is completely consonant with the above usage of "triple structure."

(12.) Margaret A. McKean, Environmental Protest and Citizen Politics in Japan (Berkeley: University of California Press, 1981).

(13.) The local groups contributed to social capital formation. Participation in such groups correlates with higher levels of trust in surveys and also by itself constitutes an important network for propagating social norms. I also advance an argument for why Japan's civil society has a dual structure. See Robert Pekkanen, "Japan's Dual Civil Society: Members Without Advocates." Ph.D. diss., Harvard University, 2002). Robert Pekkanen, "Molding Japanese Civil Society." In Frank J. Schwartz and Susan J. Pharr, eds., The State of Civil Society in Japan (Cambridge, U.K.: Cambridge University Press, 2003). Robert Pekkanen, "Japan: Social Capital Without Advocacy." In Muthiah Alagappa, ed., Political Change in Asia: The Role of Civil Society (Stanford, Calif.: Stanford University Press, forthcoming).

(14.) This supervision is established by Civil Code article 67. Paragraph 2 establishes a supervision system (kanshi seido) by the competent supervising ministry (shumu kanchou). Article 84 makes further provisions for fines by PIP directors who violate directions by the competent ministry.

(15.) Remarks at conference "Financial Support to NGOs," sponsored by Japan Center for International Education, Tokyo, June 20, 1997.

(16.) Yasushi Noumi, "Kouekiteki dantai ni okeru kouekisei to hieirisei" [Public interest and nonprofit in public interest groups], Juristo 1105 (1997): 50-55.

(17.) Shigeo Hayashi, Koueki Houjin kenkyuu nyuumon [Introduction to public interest legal person research] (Tokyo, Japan: Public Interest Legal Person Association, 1972), pp. 192-193.

(18.) Lester M. Salamon and Helmut K. Anheier, Defining the Nonprofit Sector: A Cross-National Analysis (New York: St. Martin's, 1997); Naosumi Atoda, Takayosh Amenomori, and Mio Ohta, "The Scale of the Japanese Nonprofit Sector." In Tadashi Yamamoto, ed., The Nonprofit Sector in Japan (Manchester, U.K.: Manchester University Press, 1998), p. 105.

(19.) On these issues, see variously (moral issues) Helen Hardacre "Japan: The Public Sphere in a Non-Western Setting." In Robert Wuthnow, ed., Between States and Markets: The Voluntary Sphere in Comparative Perspective (Princeton: Princeton University Press, 1991). In Frank J. Schwartz and Susan J. Pharr, eds., The State of Civil Society in Japan (Cambridge, U.K.: Cambridge University Press, 2003); (women's issues) Susan J. Pharr, Losing Face: Status Politics in Japan (Berkeley: University of California Press, 1990), and Frank K. Upham, Law and Social Change in Postwar Japan (Cambridge, Mass.: Harvard University Press, 1987); (environment) Susan J. Pharr and Joseph L. Badaracco, "Coping with Crisis: Environmental Regulation." In Thomas K. McCraw, ed., America Versus Japan (Boston: Harvard Business School Press, 1986); (Kyoto Accord) Kim Reimann, "Building Networks from the Outside In." In Frank Schwartz and Susan Pharr, eds., The State of Civil Society in Japan (Cambridge, U.K.: Cambridge University Press, 2003); and (Korean residents) Apichai Shipper, "The Political Construction of Foreign Workers in Japan," Critical Asian Studies 34, no. 1 (March 2002).

(20.) Robert Pekkanen, "Japan's New Politics: The Case of the NPO Law," Journal of Japanese Studies 26, no. 1 (Winter 2000); Robert Pekkanen, "Hou, kokka, shimin shakai" [Law, the state, and civil society], Leviathan 27 (Autumn 2000); and Makoto Imada, "The Voluntary Response to the Hanshin Awaji Earthquake: A Trigger for the Development of the Voluntary and Non-profit Sector in Japan." In Stephen P. Osborne, ed., The Voluntary and Non-Profit Sector in Japan (London: Routledge, 2003).

(21.) Tsujinaka, "Japan's Civil Society Organizations," p. 92.

(22.) Yasuo Takao, "The Rise of the 'Third Sector' in Japan," Asian Survey 41, no. 1 (2001): 290-309, 295. Takao provides a convincing description of the rise of independent civil society groups in Japan. See also Karen Nakamura, Signing Deaf in Japan: Deaf Identity, Sign Language, and Politics in Modern Japan (Ithaca: Cornell University Press, forthcoming) and, for a critical view, Akihiro Ogawa, "The Failure of Civil Society? An Ethnography of NPOs and the State in Contemporary Japan" (Ph.D. diss., Cornell University, 2004).

(23.) EPA (Economic Planning Agency), 2001 nen shiminkatsudou repooto [Citizens' activities report 2001] (Tokyo: Ministry of Finance, 2001), p. 7.

(24.) In some ways, this development is consistent with Ronald Inglehart's argument about value shifts in postmodernization: Ronald Inglehart, Culture Shift in Advanced Industrial Society (Princeton, N.J.: Princeton University Press, 1990); although scholars have generally focused on technological changes, regime transformations, and economic development per se as explanations.

(25.) Susan J. Pharr, "Conclusion." In Frank J. Schwartz and Susan J. Pharr, eds., The State of Civil Society in Japan (Cambridge, U.K.: Cambridge University Press, 2003), p. 329.

(26.) Margaret E. Keck and Kathryn Sikkink, Activists Beyond Borders: Advocacy Networks in International Politics (Ithaca: Cornell University Press, 1998), p. 11.

(27.) Ann Marie Clark, Elisabeth J. Friedman, and Kathryn Hochstetler, "The Sovereign Limits of Global Civil Society: A Comparison of NGO Participation in UN World Conferences on the Environment, Human Rights, and Women," World Politics 51, no. 1 (1998): 1-35, 6-9.

(28.) Reimann, "Building Networks from the Outside In," p. 308.

(29.) Ibid., p. 299.

(30.) Pekkanen, "Japan's New Politics"; Tanaka Yasufumi, Keidanren [Social Affairs Bureau], personal interview, Tokyo, October 8, 1997.

(31.) See Pekkanen, "Japan's New Politics" and "Hou, kokka, shimin shakai" [Law, the state, and civil society], on the NPO Law and on the tax revisions. Robert Pekkanen, "A Less-Taxing Woman? New Regulation on Tax Treatment of Nonprofits in Japan," International Journal of Not-for-Profit Law 3, no. 3 (2001).

(32.) Hideko Katsumata, "Interim Report on Public Interest Corporation Reforms Stirs Further Debate," Civil Society Monitor 9 (June 2004).

(33.) Ito Peng, "Postindustrial Pressures, Political Regime Shifts, and Social Policy Reforms in Japan and South Korea." Paper prepared for the conference "After the Developmental State," Seoul, South Korea, 2004; and Peng, The New Politics of Welfare State in Developmental Context: Explaining the 1990s Social Care Expansion in Japan (for UNRISD Project: Welfare States in Developmental Context, n.d.).

(34.) See also Susan J. Pharr and Robert Putnam, eds., Disaffected Democracies: What's Troubling the Trilateral Countries? (Princeton, N.J.: Princeton University Press, 2000).

(35.) N=2091. Source: http://roperweb.ropercenter.uconn.edu, May 25, 1994, Yomiuri Shimbun poll. Accessed March 17, 2004.

(36.) Source: http://roperweb.ropercenter.uconn.edu, April 1, 1998 poll. Accessed March 17, 2004.

(37.) Source: http://roperweb.ropercenter.uconn.edu, March 22, 1998 poll. Accessed March 17, 2004.

(38.) Michio Muramatsu, "An Arthritic Japan? The Relationship between Poltiiciand and Bureaucrats." Asia Program Special Report 117 (Washington, D.C.: Woodrow Wilson International Center for Scholars, January 2004), pp. 26-33, 27.

(39.) Scheiner, et al., "Incentives, Institutions, and Bureaucrat-Politician Relations in Japan," p. 34.

(40.) Ellis S. Krauss and Robert Pekkanen, "Explaining Party Adaptation to Electoral Reform: The Discreet Charm of the LDP?" Journal of Japanese Studies 30, no. 1 (2004): 1-31.

(41.) Asahi Shimbun, June 9, 2004.

(42.) Author interviews with LDP members of the House of Representatives, Tokyo, Japan. Interviews were conducted on July 10, 2002, and June 19, 2004, with two Diet Members. Electoral figures are from Asahi Shimbun, "Toshi mutouhasou to kasanari" [NPOs are the urban unaffiliated voters], June 9, 2004, and Asahi Shimbun, July 13, 2004.

(43.) Pekkanen, "Japan's New Politics."

(44.) Democratic Party of Japan party documents.

(45.) Without providing exact figures, the Asahi Shimbun argued that "the number of secretaries and policy staffers with NPO backgrounds is increasing in the DPJ," in an article entitled "Genba keiken, seisaku ni han'ei" [Experience in the field reflected in policies], Asahi Shimbun, June 16, 2004.

(46.) Calculated from Kokkai Youran [Diet Handbook] various years (Tokyo: Kokusei Jyouhou Sentaa).

(47.) For example, see Peng ("Postindustrial Pressures" and The New Politics of Welfare State in Developmental Context); and Margarita Estevez-Abe, "State-Society Partnerships in the Japanese Welfare State." In Frank J. Schwartz and Susan J. Pharr, eds., The State of Civil Society in Japan (Cambridge, U.K.: Cambridge University Press, 2003).

(48.) As Chan-Tiberghien argues, between 1997 and 2001, other changes also occurred as women's groups refrained their agenda in terms of women's human rights. These changes include the 1999 legalization of the birth control pill (thirty-eight years after the government established a panel to study it), the 1999 Basic Law on Gender Equality and the 2000 Anti-Stalking Law, as well as the 2001 Domestic Violence Prevention Law. Jennifer Chan-Tiberghien, Gender and Human Rights Politics in Japan (Stanford, Calif.: Stanford University Press, 2004).

Robert Pekkanen is assistant professor at the Henry M. Jackson School of International Studies, University of Washington.