The Development and Structure of Japanese Enterprise Unions
The major reason for writing a paper on Japanese unionism is that much of the debate assessing Japanese enterprise unions was set in the pre-1990s period when the economy was strong and there was a shortage of skilled labour. Thus, whilst some commentators argued that Japanese enterprise unions were little more than an arm of management,  it was difficult to refute the argument that this form of unionism had resulted in real wage and benefit increases for a considerable period of time whilst generally ensuring a high level of job security for large numbers of workers.  Over the past 15 years or so a different economic context has arisen where wage gains have been harder for unions to achieve, where unemployment and irregular employment have increased, and where globalization has made the market for Japanese goods highly competitive. It is time to provide a more contemporary assessment of this form of union organization.
Trade union history and development
The rapid industrialization and economic growth that occurred after the Meiji Restoration in 1868 transformed many feudal workers to wage labour and led to a severe shortage of skilled workers. These conditions encouraged the formation of trade unions, although the reservoir of cheap agricultural labour meant that it would be some years before a viable union movement would emerge. In these early years attempts by workers to win improvements in wages, benefits and working conditions by engaging in strike action frequently met with strong opposition from employers and the government.
Despite numerous attempts to form unions during the 1880s and 1890s, labour remained largely unorganized until the end of World War I. After this time the trade union movement began to develop, although union membership advanced slowly. A number of large-scale strikes occurred in the 1920s and in the early 1930s. By this time over 800 unions had been established, although they had recruited less than 10 per cent of the workforce.  Whilst the majority of unions were organized along industrial or craft lines, about one-third of all unions were organized on an enterprise basis. 
By the mid-1930s, in an attempt to contain strike activity and the growing union movement, the government dissolved all trade unions and absorbed them into the Industrial Association for Serving the Nation. The aim of this organization was to control radical elements in the workforce and was part of a general crackdown on worker opposition to employers or government.  At the enterprise level this body filled the vacuum left by unions and served as a defacto local bargaining mechanism and an employee welfare system. 
Following World War II, under US occupation and a new constitution, trade union membership grew rapidly and major industrial action, led by industrial unions, commenced almost immediately.  Strikes in the early post-war years were often short and peacefully settled, but over time they increasingly met stronger resistance from employers. Militant industrial union activists were dismissed and employers pressed for a more cooperative, enterprise-based union structure. While such unions had existed earlier, they were now strongly supported by management in terms of recognition and provision of facilities.  Workers were generally willing to accept this form of unionism as they were not included in the emerging broader social partnership arrangements and so had few alternative mechanisms to improve their wages and working conditions.
Union types and structure
Since the 1950s, over 80 per cent of union members have been part of an enterprise union which has exclusive representation within the company, although in some companies a weaker, second enterprise union or industrial-type union also existed. . Nevertheless, most unionized enterprises have only one union representing employees  and this type of union accounts for over 95 per cent of all unions.  Local unions may also be formed within the individual plants of the enterprise, although they are directly linked to the enterprise union. This enterprise-based structure meant that unions tended to be found in the larger companies and government agencies where organising large numbers of employees was possible.
The enterprise structure of Japanese unionism has meant that the major objectives of these unions have been the pursuit of economic goals such as job security, increased wages and improved working conditions; goals that have been traditionally pursued through collective bargaining. These goals and the enterprise structure of unions are consistent with the economic activities of companies  and can thus be broadly classified as business unionism. This form of unionism can be contrasted to social-democratic and revolutionary types of unions where unions seek a wider role in society or oppose antagonistic class interests. By adopting such a structure Japanese unions have been able to maintain a high degree of independence from the state and have not generally sought corporatist type arrangements from any of the major political parties. They have, by contrast, sought and encouraged cooperative arrangements with employers as part of their broader market-based strategies.
Most full-time, regular workers in a company are eligible to join the union and this includes front line supervisors and usually managers up to the level of subsection head.  Until recently, union membership did not normally extend to part-time employees, many of whom are women. Thus the typical union member in a Japanese enterprise union is a full-time male employee and the organizing policies of enterprise unions have served to protect this core group of male workers. Union officials are drawn almost exclusively from the membership, with some senior officials on leave from the company and working full-time for the union. 
Enterprise unions normally belong to an industrial federation which, in turn, will be affiliated to a more general peak union organization.  The major national peak union federation is the Japanese Trade Union Confederation (Rengo), which is made up of over 50 industrial union federations. The present day Rengo was formed from an amalgamation in 1989 of the private sector Rengo and unions belonging to three public sector peak union bodies. Rengo itself was the product of a merger in 1987 of unions belonging to five private sector peak union bodies.  The amalgamations have taken place as part of an effort to unite the union movement after many years of division, to provide a single union voice, and to increase trade union membership. Total unity has not been achieved, however, and two other smaller peak union bodies have emerged.
The number of unions in Japan in 2006 stood at 59,019, a decline of over 18 per cent from 1990 and 21 per cent from the record number of 74,579 unions that existed in 1984. At this time the total number of union members was 10,041,000 which represented a decline in membership of 18 per cent from 1990 and 21 per cent from the record membership of 12,699,000 in 1994. This represented a union density (number of union members as a percentage of the total workforce) of just over 18 per cent which is down significantly from the 25 per cent of workers who were members in 1990 and the peak density of 55.8 per cent in 1949. Similarly, all three peak bodies have lost members in recent years . The fall in union membership and density is graphically illustrated in the chart below.
Declining Trade Union numbers and density, 2000-2006
Source: www.eurofound.europa.eu/. ../tn0706028s_1.htm
The loss of members has sparked a number of activities at all levels of Japanese unionism. At the enterprise level, some unions have sought to expand their membership by recruiting more women, part-time workers and managers, as well as workers in subsidiary and subcontracting companies. This strategy was seen as essential if unions were to maintain membership in the wake of an increasing percentage of workers employed on a part-time or temporary basis (see graph below). In a small number of cases, enterprise unions have also extended their coverage to take in workers from other related firms within the corporate group,  while others have allowed members transferred to subsidiaries to maintain their union membership and about a third of all unions have allowed temporary employees to join.  Industrial union federations have also sought to halt the declining membership by merging with other industrial unions and by setting up occupational type trade unions, unions for agency workers and more general community type unions.  Rengo has also established ‘regional unions’ where part-time workers can join as individual members . 
What do Japanese unions do?
Increases in non-regular and part-time workers, 1990-2005Source: OECD Economic Survey of Japan 2006
Japanese enterprise unions have traditionally placed strong reliance on collective bargaining at both the company and industry levels. Collective bargaining is an appropriate mechanism for this form of unionism as the key objectives of economic rewards and welfare benefits can be achieved by trading off aspects of work such as job control, work rigidities and the introduction of new work forms and technology. Underpinning this bargaining has been a tacit understanding that employment security was guaranteed.  This has led to a dual set of employment conditions where workers in unionised enterprises enjoy ’lifetime employment’ and superior working conditions while their counterparts in non-unionised enterprises are subjected to the vagaries of the market.
Collective bargaining in Japan over wages and other monetary conditions developed a unique form in the 1950s. Groups of unions would lodge their wage demands and if these were not met would simultaneously stage repeated, short industrial actions. This system became known as the spring wage offensive or 'shunto' and led to a system of industry and national level coordination of enterprise wages.  Such a coordinated system was important from the perspective of capital as enterprises where strike action was taking place would not lose market share to their competitors and, as the industrial action was short, it would not threaten the viability of the company.  Claims were thus formulated at the industry and national level, with collective bargaining taking place at the industry and/or enterprise level around March each year. Key unions would lead the bargaining and set standards that weaker unions could adopt. The similarity of wage settlements across enterprises and industries demonstrated the influence of union federations and that the wage settlements were based on macro-economic conditions as much as enterprise considerations. 
This system remained intact for the next four decades and was able to deliver real wage and benefit increases for most union members, as well as the working population in general. During the 1990s, with the bursting of the bubble and the economy experiencing very low growth rates and increasing global competition, wage settlements became more diverse and local enterprise considerations more important. By the late-1990s the prolonged period of economic downturn led many employers and some unions to question the relevance of this multi-employer approach  and in 2002 Japanese unions were prepared to accept the end of 'shunto' as the major wage fixing mechanism.  The concept of a socially acceptable wage has now essentially disappeared from the Japanese system of wage determination. 
The demise of 'shunto' meant that the collective determination of wages and working conditions moved to a more enterprise-based system.  Such an enterprise-focused system had, however, been developing from the early 1990s and paralleled the decline in union membership and the economic downturn. Companies had been expanding joint consultative arrangements such that by 1997 nearly four out of five unionized enterprises had such mechanisms in place.  As in earlier years, these activities were concentrated in large enterprises. Joint consultation was further strengthened by the revised Labour Standards Law (1998) which introduced statutory management-worker committees for the revision of working hours. 
As many companies have experienced financial difficulties over the past decade, unions have been forced to trade wage increases for guarantees concerning employment levels. Employers have, since about 1999, also sought to align wage increases with business performance.  This has been particularly the case with the lump-sum bonuses which are normally awarded twice a year. While some employers have also attempted to extend performance criteria to monthly basic wages, it remains the case that this wage component remains strongly linked to age and length of service. There also appears to be greater diversity in scheduled hours of work, although overall unions have not made any major gains in this area in recent years. The increased diversity in hours of work may be partly a result of increased enterprise bargaining following the demise of shunto as a national system but is also due to the 2001 Social Agreement on Employment between Rengo and Nikkeirein, the then peak employer body, which committed the parties to maintaining employment while recognising that workers may need to accept shorter working hours and lower wages. 
Industrial action that flows from collective bargaining is low in Japan by international standards. Working days lost due to industrial action have been falling as have the number of disputes. The reasons for disputes have also changed with disputes over wages giving way to disputes over ‘discharge and re-instatement’ and individual disputes over the termination of employment and the deterioration of employment and working conditions.  In unionised enterprises, typically larger manufacturing companies and government agencies, wages seem to be higher and working hours lower, although the major determinant of these employment conditions are the demand for labour and the economic health of the company.
The development, structure and strategies of trade unions
What factors can explain why Japanese unions have adopted this enterprise focus? The first explanation is that the structure of Japanese unions is a product of certain historical traditions. This can be seen in the high levels of trust and collective identities that exist in the modern Japanese company which can be traced back to the Tokugawa period.  The strong hostility of management towards industrial unionism was important in establishing, the dominance of enterprise structures. However, as stated earlier, only a third of all unions were organized along enterprise lines in the 1930s, and since the 1950s, it is likely that other factors are more important.
The second explanation for Japanese enterprise unionism relates to the political context and the degree of liberalization and autonomy granted to worker organizations. The industrial unions that developed during the US occupation had increasingly engaged in industrial action that was strongly resisted by employers and the state. This occurred at a critical juncture of economic and political events which encouraged the state, the US occupation authorities and employers to restructure unions along enterprise lines. At the same time, the productivity movement was emerging in Japan with its underlying principle of sharing enterprise gains between employers, workers and consumers. This was accepted by workers as a means of ensuring job security, and served to reinforce an enterprise union structure since productivity programs could be developed most effectively at the enterprise level.
This explanation leads to a consideration of the role of employers in shaping union structure. Japanese employers promoted enterprise unions in an attempt to overcome the ‘excesses’ of industrial and ideologically based unionism that they viewed as responsible for the wave of strikes in the late 1940s and early 1950s. Moreover, this was seen as a way to more closely align the interests of workers with those of the company. This led to employers providing resources for the enterprise unions, and these unions, with some guarantees of job security, were prepared to work with management to ensure the success of the company. Thus, while external influences were important, employers’ strong support for this type of enterprise unionism proved decisive in ensuring the continuation and growth of this form.
Employers’ strategies cannot, however, be divorced from the environment in which they operate. This leads to a fourth explanation of Japanese union structure; the economic contexts of competition and lately globalization. Japanese firms operate, and have done so for many years, in a highly competitive domestic market. Through the keiretsu structure Japanese companies have a high degree of vertical integration and are usually focused on a particular industry. This industrial structure has resulted in intense competition between firms. In turn this competition has led to an emphasis on market share, which can be achieved by a unique product, a focus on quality (and to a lesser extent price), and improvements in productivity. All of these are company-based objectives that are assisted by a union structure in which collective bargaining frameworks are enterprise based. These activities were embraced by the leaders of the ‘new’ enterprise unions who saw wider industrial and ideological issues as not part of their responsibilities. For this group of union leaders, industrial or ideological concerns would become the province of the emerging union federations. More recently, although not an explanation of union structure per se, globalization has reinforced enterprise union structures both within Japan and in offshore operations.
Toyota workers rally for wage hikes
This paper examined the structure and activities of Japanese enterprise unions and explained the dominance of this form of unionism. Unions in Japan face many of the challenges that unions in other industrialized countries face. These challenges include decreasing membership, declining relevance to younger workers, a fall in the significance of collective bargaining and increasing employer involvement in the export of jobs overseas.
What then is the future of Japanese enterprise unions? The present climate provides a clear indication of the weakness of the enterprise union model. The demise of shunto, coupled with an increase in consultative activities within enterprises, have served to reinforce this form of unionism. Moreover, there is little pressure for change as can be seen in the inability of Japanese unions to provide a more inclusive union structure by recruiting part-time and women workers. While there are some possibilities for change, as shown by the emergence of several general or industrial type unions, and unions that go beyond one company to service workers in associated companies and networks, to date these initiatives are the exception and do not indicate a wider trend. Of course large companies will continue to be highly unionised, but with the demographic changes taking place in Japan, coupled with cutbacks in pensions that will force more people to take part-time work after retirement, it is likely that an increasing number of workers will fall outside unions and the protection and benefits they can provide.
John Benson is Professor and Head, School of Management, University of South Australia, Australia. This paper is a condensed version of a chapter in a recently published book on trade unions in Asia. The full chapter can be found in Trade Unions in Asia: An Economic and Sociological Analysis (London: Routledge, 2008) edited by John Benson and Ying Zhu.
Posted at Japan Focus on November 3, 2008.
 Cole, R. (1971) Japanese Blue Collar: The Changing Tradition, Berkeley: University of California Press.Dore, R. (1990) British Factory – Japanese Factory, Berkeley, University of California Press; Galenson, W. and Odaka, K. (1976) ‘The Japanese labour market’, in H. Patrick and H. Rosovsky (eds) Asia’s New Giant: How the Japanese Economy Works, Washington: Brookings Institution; Kawanishi, H. (1992) Enterprise Unionism in Japan, London: Kegan Paul International; Galenson, W. and Odaka, K. (1976) ‘The Japanese labour market’, in H. Patrick and H. Rosovsky (eds) Asia’s New Giant: How the Japanese Economy Works, Washington: Brookings Institution.
 Shirai, T. and Shimada, H. (1978) ‘Japan’, in J. Dunlop and W. Galenson (eds) Labour in the Twentieth Century, New York: Academic Press.
 Hazama, H. (1997) The History of Labour Management in Japan, Basingstoke: MacMillan; Lockwood, W. (1968) The Economic Development of Japan, Princeton: Princeton University Press; Nakamura, T. (1993) A History of Showa Japan: 1926-1989, Tokyo: University of Tokyo Press.
 Weinstein, D. (1994) ‘United we stand: firms and enterprise unions in Japan’, Journal of the Japanese and International Economies, 8: 53-71.
 Sakurabayashi, M. (1985) The Organization and Functions of the Industrial Association for Serving the Nation, Tokyo: Ocha-no-mizu-shobo.
 Hart, R. and Kawasaki, S. (1999) Work and Pay in Japan, Cambridge: Cambridge University Press.
 Farley, M. (1950) Aspects of Japan’s Labor Problems, New York: John Day Company.
 Tokunaga, S. (1983) ‘A Marxist interpretation of Japanese industrial relations, with special reference to large private enterprises’, in T. Shirai (ed.) Contemporary Industrial Relations in Japan, Wisconsin: University of Wisconsin Press; Gordon, A. (1985) The Evolution of Labour Relations in Japan, Cambridge, Mass.: Harvard University Press; Matsuzaki, H. (1992) ‘Japanese business unionism: the historical development of a unique labour movement’, Studies in Human Resource Management and Industrial Relations in Asia, No. 1, Sydney: University of New South Wales; Marsland, S. (1989) The Birth of the Japanese Labor Movement, Honolulu: University of Hawaii Press.
 Japan Institute for Labour, (2000) Japanese Working Life Profile 2000: Labour Statistics, Tokyo: Japan Institute of Labour; Kawanishi, 1992, op cit.
 Inagami, T. (1988) Japanese Workplace Industrial Relations, Tokyo: Japan Institute of Labour.
 Taishiro, T. (2000) Japanese Industrial Relations, Tokyo: Japan Institute of Labour.
 Matsuzaki, H. (1992), op cit.
 Koike, K. (1988) Understanding Industrial Relations in Modern Japan, London: MacMillan.
 Shirai, T. (1983) ‘A theory of enterprise unionism’, in T. Shirai (ed.) Contemporary Industrial Relations in Japan, Wisconsin: University of Wisconsin Press.
 Matsuda, Y. (1993) ‘Japan’, in S. Deery and R.Mitchell (eds) Labour Law and Industrial Relations in Asia, Melbourne: Longman Cheshire.
 Shinoda, T. (1987) ‘Rengo and policy participation: Japanese-style neo-corporatism’, in M. Sako and H. Sato (eds) Japanese Labour and Management in Transition, London: Routledge, 187-214.
 Japan Institute for Labour Policy and Training (2008) Japanese Working Life Profile 2007/2008: Labour Statistics, Tokyo: Japan Institute for Labour Policy and Training; Fujimura, H. (1997) ‘New unionism: beyond enterprise unionism?’, in M. Sato and H. Sato (eds) Japanese Labour and Management in Transition, London: Routledge; Kuwahara, Y. (1993) ‘Industrial relations in Japan’, in G. Bamber and R. Lansbury (eds) International and Comparative Industrial Relations, Sydney: Allen and Unwin.
 Sako, M. (2006) Shifting Boundaries of the Firm: Japanese Company – Japanese Labour, Oxford: Oxford University Press; Sako, M. and Sato, H. ‘Union networks in the extended enterprise in Japan: evidence from the automobile and electrical machinery industries’, paper presented at the International Industrial Relations Association Congress, Tokyo, May-June 2000.
 Japan Institute of Labour (2002) Formation of Labor Unions and Responses to Management Crises – Industrial Relations in the Late 1990s (Summary), Tokyo: Japan Institute of Labour.
 European Foundation for the Improvement of Living and Working Conditions (2005) Industrial Relations in the EU, Japan and the USA, 2003-2004, Dublin: European Foundation for the Improvement of Living and Working Conditions.
 Japan Institute for Labour Policy and Training (2005) Labor Situation in Japan and Analysis: Detailed Exposition 2005/2006: Labour Statistics, Tokyo: Japan Institute for Labour Policy and Training.
 Benson, J. and Debroux, P. (2000) ‘Japanese trade unions at the crossroads: dilemmas and opportunities created by globalization’, Asia Pacific Business Review, 6(3/4): 114-32.
 Sako, M. (1997) ‘Shunto: the role of employer and union coordination at the industry and inter-sectoral levels’, in M. Sako and H. Sato (eds) Japanese Labour and Management in Transition: Diversity, Flexibility and Participation, London: Routledge; Suzuki, F. (1997) ‘Labour relations, trade union organizations, collective bargaining, and labour management consultation’, in DGB-JTUC Future of Work, Future of Social Welfare State, Future of Trade Unions, Tokyo: Rengo.
 Nimura, K. (1994) ‘Post-second world war labour relations in Japan’, in J. Hagan and A. Wells (eds) Industrial Relations in Australia and Japan, St. Leonards, New South Wales: Allen and Unwin.
 Nikkeren (1999) The Current Labour Economy in Japan, Tokyo: Japan Federation of Employer Associations; Tachibanaki, T. and Noda, T. (2000) The Economic Effects of Trade Unions in Japan, London: MacmMillan.
 Suzuki, F. (1997), op cit.
 Rengo (2002) Rengo Hakusho [Rengo White Book], Tokyo: Japanese Trade Union Confederation.
 Nakamura, K. (2007) ‘Decline or revival?: Japanese labor unions’, Japan Labor Review, 4(1): 7-22.
 Weathers, C. (2003) ‘The decentralization of Japan’s wage setting system in comparative perspective’, Industrial Relations Journal, 34(2): 119-34.
 Japan Institute for Labour (2000), op cit.
 Yamakawa, R. (1998) ‘Overhaul after 50 years: the amendment of the labour standards law’, Japan Labour Bulletin, 37(11): 5-12.
 European Foundation for the Improvement of Living and Working Conditions (2005), op cit.
 European Foundation for the Improvement of Living and Working Conditions (2005), op cit.
 Japan Institute for Labour Policy and Training (2007), op cit.
 Whitley, R. (1992) Business Systems in East Asia: Firms, Markets and Societies, London: Sage.
We welcome your comments on this and all other articles. More are available on our homepage. Please consider subscribing to our email newsletter or RSS feed, or following us via Twitter or Facebook.