Labour market and business practices differ between the United States and Japan in a variety of
ways. Writing in the latest issue of the Economic Journal, Professor Hodaka Morita explores the
key differences:
l First, Japanese firms conduct continuous process improvement (or 'kaizen') more than US
firms do.
l Second, the labour turnover rate - the degree to which employees move firms - is higher in
the United States than in Japan.
l Third, Japanese firms train their employees more than US firms do.
l Finally, training in Japanese firms is more 'firm-specific' and less general than in US firms.
These differences are important, Morita argues, since investment in training and manufacturing
processes are fundamental factors in determining the strength of firms and industries. For
example, the MIT Commission on Industrial Productivity, headed by Michael Dertouzos, Richard
Lester and Robert Solow, concluded in their influential book Made in America that lack of
attention to manufacturing processes and on-the-job training were crucial reasons for the weakness
of US industrial performance in the 1980s. While the 1990s have seen a different story, these
issues may well become important once more as the US economy slows and the high-tech
industries shake out.
Morita's analysis provides a theoretical explanation for the differences between US and Japanese
practices. The logic goes as follows. In general, continuous process improvement involves a
number of small changes and modifications, which are mostly unobservable from outside the firm.
Therefore, if a firm conducts continuous process improvement, a degree of specificity is
introduced into the firm's technology. An improved technology yields higher productivity only if it
is operated by an employee who has been trained in the technology and who is therefore familiar
with its firm-specificity. Hence, the higher the number of employees who remain with a firm and
the longer they stay, the greater the benefit of continuous process improvement.
When other firms also conduct continuous process improvement, each firm's turnover rate
becomes lower because its employees are less productive in other firms. Hence, the benefit of
continuous process improvement increases when other firms also conduct continuous process
improvement. If the net profit from continuous process improvement is positive only if other firms
also conduct it, there are two possible outcomes: either all firms conduct continuous process
improvement or none do.
Morita interprets the former as the Japanese outcome and the latter as the US outcome. Since
continuous process improvement leads to a degree of specificity in a firm's technology, in the
Japanese outcome, training provided by a firm is less effective in other firms. Hence, training in
the Japanese outcome is partly firm-specific, whereas it is general in the US outcome. This lowers
the turnover rate in the Japanese outcome, which in turn implies that Japanese firms have more
incentive to train their employees.
Why have different outcomes emerged in the United States and Japan? Historical events during
the Second World War provide an answer to this question. During the war, government labour
regulations were different between the two countries in many aspects. In particular, the US
government promoted industry-wide training programmes (e.g., the Training Within Industry
programme) and standardisation of skills in order to maximise production efficiency, whereas the
Japanese government encouraged corporate training programmes and the development of firmspecific
skills. These differences led the two countries to select different outcomes in the post-war
period.
One key innovation of Morita's analysis is that it addresses the connection between labour market
practices and the nature of technology. By doing so, it captures two US-Japanese differences not
previously understood: why Japanese firms conduct continuous process improvement more than
US firms do; and why human capital accumulation is more firm-specific in Japanese firms.
A natural direction of future research is to apply this framework to analyse the recent changes in
labour market practices in Japan. For example, if the recent advances in information technology
reduce the firm-specificity of technology associated with continuous process improvement, then
the framework predicts that the Japanese outcome will converge on the US one.
ENDS
Note for Editors: 'Choice of Technology and Labour Market Consequences: An Explanation of
US-Japanese Differences' by Hodaka Morita is published in the January 2001 issue of the
Economic Journal. Morita is at the University of New South Wales.
For Further Information: contact Hodaka Morita on 00-61-2-9385-3341 (email:
H.Morita@unsw.edu.au); RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or
07768-661095 (email: romesh@compuserve.com); or RES Media Assistant Niall Flynn on 020-
7878-2919 (email: nflynn@cepr.org).