Special Topic

Vol.40-No.11 November 1, 2001

     Since the “bubble economy” burst, Japan has been experiencing various changes. One such change concerns the employment system, once praised as a source of economic success but now a target for reform. This includes labor policy. For instance, from the 1970s the major goal of Japanese employment policy was to subsidize private companies that were experiencing difficulties, thereby enabling them to maintain their employment levels. The Ministry of Labour* (MOL, now the Ministry of Health, Labour and Welfare) is currently advocating new policies to promote labor mobility between companies. The regulation of employment practices is also being reduced by deregulation. Partly in response to these changes, the process for determining labor policy has changed. This article analyzes the changes that occurred in the 1990s.

     This pluralistic view resembles the neo-corporatist thesis or the “interest group liberalism” of Lowi (1979), in which interest groups, politicians and bureaucracies dominate their respective policy arenas.

     How do labor unions come into play in this context? Pempel and Tsunekawa (1979) argued that, given the existence of dense institutional networks among interest groups and bureaucracies, Japan is very close to a corporatist regime, except for one omission: labor unions, which are major participants in a corporatist regime. Therefore, these authors concluded that Japan is best characterized as “corporatism without labor.” However, since this pioneering work, studies have appeared demonstrating labor's successful participation in making policy.

     Deliberation councils are frequently used in Japanese policy-making, not only concerning labor, but in other areas also. There were 217 deliberation councils with legal standing in 1996, and more “informal” deliberation councils. In some cases, deliberation councils have simply rubber-stamped government policy, while in others they have played a significant role (Schwarz, 1998; Sone, 1986).

     In the labor policy arena, deliberation councils are very important. Council members include both union and employer representatives, and the councils usually adopt a unanimous ruling. In other words, union representatives have a de facto veto over labor policy, although so do employers. Consequently, the deliberation councils of the MOL became an important field of interest intermediation between related parties (Shinoda, 1989).

     Based on an interview survey of government bureaucrats in 1985, Yutaka Tsujinaka (1987) found that, relative to bureaucrats in general, Labour Ministry bureaucrats tend to have more intense relations with societal actors, to be more exposed to politicians, and to cooperate to a greater degree with societal actors in policy formation. The author also found that MOL bureaucrats are more likely than other bureaucrats to be influenced by the deliberation councils and the interested parties, that is, the labor unions and employers (Kume, 1998).

     However, this does not mean that interests of unions and employers have captured the MOL. Rather, given the competing interests of unions and employers, the MOL did have some autonomy and was able to formulate more economically rational and more market-oriented policy than the Ministry of Agriculture or the Ministry of Construction.*

     Other characteristics of labor policy relate to the partisan aspect of labor policy-making. Pluralist studies of Japanese politics have revealed an increasing influence of political parties in the political process. Against the bureaucratic-dominance approach, or the cultural approach, which emphasizes “consensus,” these studies demonstrated how partisan conflict determined the nature of Japanese politics. Furthermore, these studies argued that members of the predominant Liberal Democratic Party (LDP) successfully increased their influence over policy-making by using their policy-expertise accumulated through on the job training, becoming a “policy tribe.”

     However, labor policy-making is not politicized in this sense. Rather, policy-making has become depoliticized in the 1980s. Although labor policy used to be one of the major focal points of partisan conflict in the 1950s and 1960s, since the 1970s it has become less politicized. This is partly because the demands of labor unions became more “moderate” than in the 1950s, when most labor unions actually pursued revolutionary goals. More importantly, however, under the dominance of the LDP, labor unions, especially moderate private-sector unions, have cultivated communication channels with the government, bypassing the opposition parties. While labor unions increased their participation in the policy-making process, party politics reduced its impact in this policy arena (Kume, 1998).

     The policy tribe plays a less prominent role in the labor policy arena. The pioneering work in this field, Liberal Democratic Party Policy Affairs Research Council (PARC), did not identify any labor policy tribe, although it did note an effort in 1983 to organize a parliamentary group with the aim of promoting the interests of salaried workers inside the LDP (Nihon Keizai Shinbunsha, 1983). Frequently such a parliamentary group is a core organization for the policy tribe. This book concluded that this new group fell short of being a labor policy tribe, but admitted that it might have become one. Inoguchi and Iwai (1987), in the first academic study of the policy tribe phenomena, argued that the labor policy tribe comprises mainly parliamentarians whose careers were in the MOL bureaucracy and ex-chairmen of the PARC labor policy sub-committee. That is, labor policy tribe members are not particularly active in influencing policy-making; rather they are expected to support the MOL.

     Overall, the labor policy arena was not politicized, and interest representation and aggregation were conducted within the institutional framework of the MOL. Here we can see the most “corporatist” policy-making feature of Japanese politics.

3.1 Cases of Labor Policy Formation
     One tentative observation is that the labor policy process seemingly lost the characteristic nature achieved in the 1980s. To support this observation, it is useful to review the postponement of the proposal to reduce working hours in 1993.

     The Labour Standards Law was expected to be revised in 1994 to regulate working hours to “40 hours a week.” However, after the economic bubble burst and the economy weakened, companies lobbied the LDP to postpone the introduction of the regulation (Asahi Shinbum, March 18, 1993). The MOL decided to accept this demand, and called for the Central Labour Standard Deliberation Council (CLSDC) to endorse it. However, union representatives from Rengo (Japanese Trade Union Confederation) not only opposed this proposal, they strongly criticized the procedure. The unions argued that the MOL had downplayed the deliberation council by unilaterally accepting management opinion, and boycotted the meeting. The proposal to postpone the introduction of the 40-hour work week was endorsed by the council, which was then comprised only of management representatives and members representing the “public interest.” However, the latter stated that it was regrettable that the MOL had not had sufficient communication with the members of the Deliberation Council, and demanded that the MOL follow an adequate procedure.

     The case of the Labour Standards Law Revision in 1986 showed a very different pattern, although it also contained a severe conflict of interest between management and labor. In this case, unions advocated new legal restrictions aimed at shortening working hours. A legal approach was pursued because Japanese enterprise unionism, low unionization rates in small and medium-sized companies, and strong inter-company competition made it difficult for unions on their own to achieve reductions in working hours within their companies (Shirai, 1987).

     The MOL, basically concurring with the unions, submitted a draft to the tripartite CLSDC in March 1986, and the CLSDC held a series of deliberation meetings. However, management opposed it, arguing that: (1) reduction in working hours is related to how productivity gains are distributed; (2) the government should not legally regulate working hours since management and labor of an individual company should decide such a matter; and (3) if the government legally imposes the shorter working hours standard, individual companies would have to pay the extra cost of overtime work, thereby suffering a fall in competitiveness. Consequently, the CLSDC could have been deadlocked. Public representatives, scholars, and ex-bureaucrats of the CLSDC took the initiative to break this deadlock. They formulated their own proposal and actively persuaded both management and labor to endorse it. The proposal's main point was that legal working hours should eventually be reduced to 40 hours, but for the time being working hours would be 46 hours, and then reduced to 44 hours. This was a concession to management, because the immediate reduction was to 46 hours per week rather than the 45 hours that the unions advocated. However, the proposal also formally announced the intention to introduce the 40-hour week for the sake of labor. After intensive negotiation, management and labor representatives endorsed the proposal. In this process, the deliberation council functioned as an important and effective forum of interest intermediation, which did not occur in the 1993 case.

     Let us discuss another case. In the 1990s employment practices were deregulated in an effort to adjust the regulation of employment to changes in industry and the economy. For instance, the Worker Dispatching Law, which allowed jobs to be filled by dispatched workers in 26 specific types of work, such as operating office automation (OA) equipment and filing, was revised in 1999 to abolish restrictions on the types of work covered. In 1999, revisions to the Employment Security Law lifted the general prohibition on private fee-charging placement businesses.

     Here, we review the aftermath of the regulation of working hours. The Labour Standards Law clearly prescribed that the 40 hour-a-week principle would be fully introduced in April 1997. The MOL proposed that in order to implement the new regulation smoothly, in the first two years the MOL would use administrative guidance rather than sanctions to obtain company compliance. This proposal was discussed in the CLSDC. Naturally, union representatives criticized this policy, arguing that the MOL would allow management to retain long working hours practices through the backdoor. Union representatives did not openly oppose the policy, because the Director-General for Labour Standards Bureau of the MOL promised that the MOL would not allow illegal practices, and would use adequate guidance to ensure company compliance with the new law. Despite this commitment, however, in the formal deliberations, three out of seven union representatives abstained from voting, which was uncommon in this deliberation council (Shukan Rodo News, December 9, 1996).

     Similar conflicts of interests appeared in the CLSDC deliberations in 1997. The council had been studying the overtime work system, the discretionary work system, the flexible working hours system, and the upper limit on labor contracts. On all issues, unions and management took very different stances. In short, union members demanded legal restrictions, whilst management wanted a freer hand (Shukan Rodo News, February 2, 1998). For instance, labor unions insisted that the upper limit on overtime should be set at 150 hours a year, while management argued that such a regulation was inappropriate on the grounds that the economic situation was unpredictable. Thus, the report of the CLSDC (November 21, 1997) was not unanimous, but listed the minority view of the unions. Consequently, the MOL had to prepare the bill for the Revised Labour Standards Law without having completed “negotiations” between the interested parties. The bill proposed by the MOL to the CLSDC in January 1998 did not receive the support of the unions, although the Council endorsed it.

     Labor unions subsequently sought to influence the labor policy-making process from outside the deliberation council. At that time, the Social Democratic Party was in the governing coalition, and Rengo appealed to the SDP that their demands be reflected in the bill. But it also asked opposition parties such as the Shinshin Party and the Democratic Party to submit a bill of their own incorporating Rengo's demands. In parliamentary deliberations, some concessions were made to union demands. Further restrictions were added to the discretionary work system and its introduction was postponed for a year, and the use of the overtime work system was limited (Nihon Rodo Nenkan, 1999, pp.343-352).

     Hence, we see that the labor policy-making process, in which conflicting interests between labor and management are negotiated within the framework of “corporatist” arrangements, is changing. The conflicts are spilling over into partisan politics.

     It is noticeable that in other policy arenas the role of the deliberation councils also seemed to decline. Furthermore, in the context of administrative reform, the Hashimoto Cabinet decided that the use of the deliberation council should be reduced. Issues such as legalization of the stockholding company, the stock-option system, and the revision of the Juvenile Law were all dealt with by party politicians without the use of deliberation councils. The contrast with changes in the labor policy arena as described above seems stark.

3.2 Three Surveys of Union Leaders
     Data from three surveys of union leaders are available. In 1980, Michio Muramatsu conducted the first survey of 252 interest groups, including 52 labor unions, which asked questions about their activities and preferences. In the summer of 1989, the author conducted a survey of 51 labor unions, using the same questionnaire format and targeting the same unions as in the 1980 Interest Group Survey to investigate any changes in the perception of labor leaders during the 1980s. The targeted unions were four national-level unions and 47 industry-level unions. In 1994 Michio Muramatsu conducted the Second Interest Group Survey using the same procedures, but with some new questions (Muramatsu, Ito and Tsujinaka, 1984; Kume, 1998).

     What kinds of changes occurred in labor politics according to this “panel” data? First, we asked how often the union leaders are consulted by a government agency on policy issues, using a five-scale answer format. In all three surveys, around 60 percent of the unions answered “very often,” “often” and “to some degree,” although the percentage peaked in 1989 (Figure 1). If we disaggregate the data, we can see that government consultation with private-sector unions decreased, while consultation with public-sector unions increased. In addition, Rengo answered “often” in both 1989 and 1994 (it did not exist in 1980.) In summary, the level of administrative consultation with unions has not changed its intensity.

     However, it is interesting to discover that the union leaders' confidence in government agencies declined dramatically in the 1990s. We asked to what extent the union leaders have “trust” in the government agency with which they have their usual contact. While we found they had more trust in the administrative agencies in 1989 than they did in 1980, trust declined in 1994 (Figure 2). This decline occurred among both private and public-sector union leaders: the percentage of private-sector union leaders showing trust was 53.1 percent in 1980, 63.9 percent in 1989, and 30.3 percent in 1994. Corresponding figures for public-sector union leaders were 28.6 percent in 1980, 70.0 percent in 1989, and 40.0 percent in 1994. We can see a general trend among union leaders.

     This data seems to be consistent with the observation of changing patterns in labor policy-making in the 1990s as described above. In the 1980s the unions successfully increased their participation in the policy-making process, one important avenue of which was the deliberation councils. In other words, in the 1980s unions increased their contacts with government agencies through institutionalized channels of participation, and were satisfied with such participation. However, the trust placed by unions in such institutionalized participation declined, as it fell short of the effective forum of interest intermediation. We can also infer important changes in labor policy-making in the 1990s on the basis of the union leader survey data.

4.0 Possible Interpretations
     How can we explain changes in labor policy-making? I propose three causal hypotheses. First, globalization may be behind this change. Globalized market competition in combination with a prolonged economic downturn after the bursting of the bubble economy put pressure on the economy, and placed on it demands for its structural transformation. Many economists and practitioners argued that the once admired Japanese form of capitalism had become obsolete (e.g., Noguchi, 1994; Katz, 1999). Many believed that the Japanese management system had become rigid, reducing the international competitiveness of Japanese industry. Based on this understanding, management began introducing reforms to employment practices and the wage system. Performance-based wage systems in place of seniority wages, and labor mobility rather than “lifetime” employment define a reform agenda.

     This view easily leads to the conclusion that management wants radical new managerial practices and demands new legislation or deregulation at the expense of workers, so much so that the institutionalized policy-making based on compromise and consensus became obsolete. On this view, it is the management offensive that broke the established method of policy-making.

     This interpretation has some problems. It assumes that management in the new globalized market is unilaterally transforming the Japanese political economy, including labor policy. However, management does not have a clear goal for such a transformation, and neither could management implement such a goal unilaterally, as the slow, sometimes stalemated, process of “restructuring” shows. This point is related to the second issue, which concerns labor's preferences and influence. The chairman of the CLSDC, Tadashi Hanami, reflecting the conflict-ridden process of deliberation on deregulating labor standards, commented that union representatives resisted government proposals under the false assumption that “deregulation would necessarily force workers to accept worse working conditions.” (Shukan Rodo News, February 2, 1998). The chairman's point is that deregulation would sometimes promote workers' interests. It is worth noting that unions in the electrical equipment industry are more eager to accept new flexible employment practices. Furthermore, the new globalized economy makes it more important for employers to motivate their workers to help companies survive the increased competition (Thelen and Kume, 2000). This means that globalization may improve the bargaining position of some workers. Globalization has many faces. In other words, globalization influences the national political economy, but does so within political and institutional contexts.

     A comparative perspective provides clues. Political economists have recently paid increasing attention to two national economies that are thought to have adapted to globalization successfully: New Zealand and the Netherlands. In 1984 in New Zealand, the Labour Party came to power with a drastic deregulation policy package spanning the capital market to the labor market. This caused tension between the Labour government and labor unions. Some unions tried to rebuild negotiation channels with the government in a corporatist way, but consensus among unions could not be reached (Bray and Neilson, 1996). What was worse for the unions was that the conservative National Party, which won the election in 1990, took a more radical approach to deregulation, making relations between the government and unions even more conflict-ridden. In any case, the New Zealand economy began growing in the mid-1990s (Kelsey, 1995).

     A similar economic recovery based on deregulation occurred in the Netherlands, which suffered from the “Dutch disease” in the early 1980s. The Dutch government also loosened regulation of the labor market, allowing part-time and temporary employment, whilst cutting back the welfare program. However, the Dutch political process is very different from that of New Zealand. Reform started in 1982, when the government, the unions, and management agreed to implement the Wassener Agreement. This Dutch “Polder Model” is formulated and implemented in the corporatist framework (Visser and Hemerijck, 1997).

     These two cases show that the pressure of globalization can transform national politics in various ways rather than result in a simple convergence. This observation leads us to our two alternative interpretations.

     The second interpretation focuses on the development of party politics. In 1989 the LDP lost its absolute majority in the Upper House, and in 1993 it stepped down from the governing position after losing its majority in the powerful Lower House, ending its dominance that began in 1955. Since then, there has been a series of coalition governments (with a brief one-party minority government), although the LDP returned to government in 1994 and maintained its position as the largest party within the coalition.

     According to most informed observers, the political realignment has not been completed yet, and the future course of party politics remains unclear (e.g., Otake, 1999). However, in the 1990s there was a substantial movement towards a two-party system, although this is not yet complete. The Shinshin Party was established by social democrats and liberal conservatives as a rival party to the LDP, although it failed to achieve its potential. Now the Democratic Party is pursuing the same goal. This is partly because the end of the Cold War blurred the ideological dividing line between left and right, but also because of the introduction of the single-member district system in the elections for the Lower House.

     This change seemed to influence the nature of labor policy-making. In the heyday of the LDP's dominance, it was a safe bet that the LDP would stay in power for the foreseeable future (Miyake, 1986). Thus, it was a rational strategy for the unions to participate in the policy-making process and negotiate formally and informally with the LDP government to realize their members' interests. The LDP had its own reasons for incorporating labor into the policy-making process. It faced declining popular support due to demographic change. Traditional LDP supporters in the agricultural industry decreased in numbers. Thus the LDP tried to be a “catch-all” party by appealing to the old-leftist camp, that is, to labor. An easy target was private-sector labor, which was politically moderate and had a “materialist” orientation. The result was the development of corporatist policy-making, particularly in the labor policy field.

     However, after 1993 union leaders began a serious attempt to build a strong rival party to the LDP, believing such a strategy to be promising in the new political environment. The labor union survey data shows that in 1994 union contact with the LDP declined substantially. In 1994, when the LDP was the opposition party, it approached the unions for support, which the unions were reluctant to offer. When the LDP returned to government with the Social Democrats, the unions tried to establish the Democratic Party as an opposition party to the LDP. In return, the LDP coalition government stopped its regular consultations (seirokaiken), which had been an important institutional channel between the government and labor. In sum, the new party politics transformed the old institutionalized labor politics.

     The third interpretation is related to the second, and focuses on the labor unions. As explained above, labor unions, especially private-sector ones, have fruitfully increased their participation in the policy-making process. In this process, the private-sector unions “sold”to the LDP government their moderate “business unionism,” which is distinct from the ideological leftist union movement. As this strategy was so successful, the private-sector unions achieved some supremacy over the public-sector leftist movement. A rather similar process took place in Sweden in the 1960s, where the export-oriented metal sector established its hegemony within the labor movement (Pontusson, 1992).

     This eventually led to the establishment of a unified national labor union federation, Rengo, in 1989. So this should have meant the private-sector labor movement winning a position of hegemony, and attempting to consolidate its “corporatist” strategy. However, what actually happened was that Rengo changed its policy demands from “production”-oriented ones, such as industrial policy and active labor market policy, to “distribution”-oriented ones, such as reductions in working hours. Furthermore, Rengo became more eager to demand state regulation of the labor market. These new issues tend to provoke political conflicts.

     What is interesting is that in the process of administrative reform in 1997 under the Hashimoto Cabinet, Rengo played an important role in protecting the interests of public-sector workers. We should recall that private-sector unions, which were the driving force behind the labor front unification in 1989, played a leading role in administrative reforms during the Nakasone government. Once unification was complete, Rengo could not play such a role, due to intra-union politics. Furthermore, some Rengo leaders, such as the first chairman Akira Yamagishi, hoped that with this unified power they could develop a more political and purely social democratic strategy against the conservative LDP government, which happened to lose power in the late 1980s. This strategic turn on the side of the labor movement seemingly changed the nature of labor policy-making.

     In Japan, the pressure from globalization became severe after the economic bubble burst in 1990. At that time, a major political discourse started on the issue of how to restructure the political economy. My counter-factual inference is that if the old institutional framework had remained, those challenges could have been managed in a very different way: that is, in a “Dutch-like” rather than a “New Zealand-like” way. In this sense, changing institutional settings are very important for understanding changes in labor policy-making, and probably the political economy.

previous page    next page    MENU

Labour Law and Social Policy

Working Conditions / Human Resources Management