No abstract

This paper examines the empirical implications of Becker's classical theory of employer discrimination. If the male-female wage differential is due to employer discrimination, then non-discriminatory employers will hire more women and enjoy a higher profit than discriminatory employers. This proposition is tested using Japanese firm-level panel data from the 1990s. The empirical results, based on pooled OLS, indicate that an increase in the proportion of women employed by a firm enhances its operating profit. The female proportion could be endogenous in the profit equation, however, because firms may adjust their female labor in response to positive productivity shocks. To deal with this possible endogeneity, investment or intermediate input amounts are used as proxy variables for productivity shocks, as suggested by Olley and Pakes (1996) and Levinsohn and Petrin (2003). The findings based on these proxy variable estimations confirm the robustness of the finding that hiring more women results in higher profit. This result fails to reject the hypothesis that employer discrimination is a source of the male-female wage gap. However, the size of estimated effect of female proportion on profit is 1/20 of the predicted coefficient calculated based on the assumption that all the observed gender wage gap is due to gender discrimination. This result suggests that the large portion of gender wage gap is due to gender productivity gap. In addition, those firms that hire more women do not necessarily grow faster than firms that hire fewer women. The data do not reject the static implication of Becker's hypothesis, but do reject its dynamic implication.

We investigate the argument that men who are raised by working mothers develop a preference that is favorable toward working women, and, consequently, are more likely to have working wives. We test this hypothesis using the Japanese General Social Surveys 2000-2003. We directly examine the responses to the opinion survey's questions regarding appropriate gender roles. The estimation results indicate that men raised by full-time working mothers are less likely to support traditional gender roles. Those men are also less likely to believe in the negative impact of a mother's working on her children's development. JEL Classification: J12, J16, J22

This paper explains the secular increase of nonstandard workers in Japan, whose share of employment increased from 17 to 34% between 1986 and 2008. Changes in labor force and industrial compositions account for one quarter of the increase of nonstandard workers. Product‐demand fluctuation and the introduction of information and communication technologies increased firms' usage of nonstandard workers, but its quantitative effects are limited. The increase of nonstandard workers was concentrated among new entrants to the labor market, male workers of younger cohorts, and female workers of all cohorts, suggesting that the declining importance of the long‐term employment relationship is a major cause for the increase of nonstandard workers.