The Japanese firm

The ideas introduced in this chapter have recently been widely discussed in the context of the contrast between hierarchical structures in Japan and those in the United States and other Western countries. We cannot offer a survey of what is now a very large literature.33 However, even a cursory sketch reveals how closely the debate follows the theoretical issues. The Japanese system is characterised by three important features:

  1. Japanese workers expect to remain with their employer for longer than do workers in the United States. The name given to this system of ‘lifetime employment’ is something of a misnomer, since Japanese workers do move when young and face mandatory retirement at a rel­atively early age.34 However, Hashimoto and Raisian (1985) found that a Japanese male worker would typically hold 4.9 jobs before retir­ing compared with eleven jobs in the case of a typical male worker in the United States. Continuous employment is fostered also by sub­stantial separation payments which vary according to whether the separation is a result of ‘private’ or ‘company’ reasons. At the time of mandatory retirement, the separation payment represents, on average, seventy per cent of the present value of retirement benefits provided by companies.35
  2. Earnings rise with tenure in Japan as in other countries, but the rela­tionship is stronger in Japan than in the United States. Comparing the effect of general labour market experience and firm-specific experience, Hashimoto and Raisian (1985) found that firm-specific experience was more important in Japan as a determinant of earnings, whereas the reverse was true of the United States.
  3. Japanese unions are not the industry-wide, general or craft-based unions, common in other countries. They are mainly enterprise unions.36

The general significance of these bald observations for our purposes here is simply that they relate closely to the characteristics of internal labour markets discussed earlier. They suggest that Japanese governance arrange­ments put greater emphasis on the continuity of relationships than is the case in the United States. There is a suggestion that either because of greater trust or because of greater investment in specific human capital the relationship between tenure and wages is more pronounced. There is also an implication that the enterprise union plays an important part in moni­toring the very specific circumstances of the firm; an undertaking that ‘outside’ institutions are less well equipped to perform. The role of the union is to mediate on the distribution of rents in an environment where returns to specific human capital are very significant and where the quit threat is worthless.37

Differences in institutional structures do not in themselves lead to straightforward inferences about relative efficiency. Not only is the infor­mation about institutions continually changing, but varying possible interpretations can be put upon it. Clark and Ogawa (1992), for example, have recently cast doubt on the conventional wisdom about the wage-job tenure relationship in Japan. They show that considerable changes to the wage-tenure relationship have occurred in recent decades as the popula­tion has aged, the retirement age has risen, and productivity growth has declined. ‘By 1986, the value of an additional year of general experience exceeds the gain from a year of job tenure’ (p.344). Further, as Aoki (1990, p. 12) argues, how far seniority payments reflect learning (human capital) and how far ‘bonding’ for diligence, is still not empirically settled.

Another aspect of the Japanese system, which has interested students of business structure, relates not to the role of the firm as a monitoring system but as a means of generating new information. Aoki (1986) contrasts the system by which jobs are defined by function and controlled by a vertical hierarchy according to fixed rules, with a system in which people are expected to accumulate knowledge of a wide range of jobs and take respon­sibility for coping with emergent events.38 The former situation (which he admits is overdrawn) represents the American or ‘A’ firm. The latter is the Japanese or ‘J’ firm. The American system encourages specialisation and the development of outside markets in these specialist types of labour. The Japanese system encourages wide knowledge of the operations of a partic­ular firm, a great emphasis on learning on the job, and the willingness to act flexibly and cooperatively.39 Employees see themselves as members of a team rather than of a trade.

This argument of Aoki that the J firm is characterised by horizontal coordination while the A firm represents a system of rational ‘hierarchi­cal’ control relates closely to the encouragement of entrepreneurship within the firm. The local knowledge accumulated by people in the J firm along with the ability to make suitable minor adjustments to production plans means that it is an excellent vehicle for the small-scale Cassonian entrepreneur and makes good use of ‘tacit knowledge’. Information is gen­erated and disseminated to those who will find it useful and continual flexibility at this level is achieved. The rents generated by this type of knowledge are really entrepreneurial in nature. Seniority payments and promotion as a reward for good cooperative performance may then be seen as ways of distributing these entrepreneurial rents between the various members of the team. Small-scale adjustments to evolving conditions are well handled by the J firm according to Aoki. It is not so clear whether the J firm has a similar advantage in the realm of larger scale changes of the Schumpeterian variety. Where old bonds have to be broken, a system which is designed to nurture and protect them may confront awkward choices. This is, however, an issue that is more conveniently discussed in Chapter 7.

Source: Ricketts Martin (2002), The Economics of Business Enterprise: An Introduction to Economic Organisation and the Theory of the Firm, Edward Elgar Pub; 3rd edition.

One thought on “The Japanese firm

Leave a Reply

Your email address will not be published.