|YAMAGUCHI Kazuo (Visiting Fellow, RIETI)
|October 2011 11-J-069
|Download / Links
This paper employs a macro-data analysis of GDP per hour among the Organisation for Economic Co-operation and Development (OECD) countries and a micro-data analysis of the performance of Japanese firms, and examines how Gender Empowerment Measure (GEM) is associated with GDP per hour, and how the work-life balance (WLB) practices of firms affect their performance, measured by the gross margin per employee or per hour for the employees' total hours of work. Since gross margin can take on negative values, the analysis that uses the logarithm of gross margin per employee or per hour as the dependent variable employs the Tobit regression model. The results of the analyses show the following. While GEM has no unique association with GDP per capita, controlling for the Human Development Index (HDI), it shows a significant positive association with GDP per hour, controlling for HDI, and thereby suggests the association of the utilization of women's human resources with productivity per hour. Among Japanese firms, those that have concrete WLB practices and consider "encouraging employees to fulfill their potential regardless of gender" as very important in their management practices more so than other firms perform better, but they are still scarce in Japan. Unlike the results for male regular employees, firms' performance does not depend on the educational composition of female regular employees, thereby suggesting that Japanese firms on average fail to utilize the human resources of highly educated women. However, controlling for the gender ratio of regular employees, firms with a higher ratio of women in their managerial/administrative positions demonstrate better performance. In addition, as the ratio of women in managerial/administrative positions increases, a positive effect of the average educational attainment of female employees on firms' performance emerges. However, the average ratio of women in managerial/administrative positions is still very low among Japanese firms. The policy implications of those findings are also discussed.