On the Effects of Employers' Work-Life Balance Policy and Restricted Regular Employment Policy on Gender Wage Gap

This paper examines whether certain diversity management policies promoting empowerment of women in Japanese firms increase women's wages and thereby decrease the gender wage gap. In particular, the study focuses on the influence of (1) employers' personnel policies that "encourage employees to fulfill their potential regardless of gender," (hereafter, the "GEO"-gender equality of opportunity-policy), along with (2) employers' systematic work-life balance (WLB) policies and (3) regular employment systems for employees with restricted location (hereafter, the "restricted regular employment" policy). The data employed for analysis are the linked Japanese employer and employee survey data from the 2009 International Comparative Survey on Work-Life Balance conducted by the Research Institute of Economy, Trade and Industry. There exist selection biases in employers' policies and measures since employees are not exposed to them randomly. This paper tries to eliminate the bias of observed employers' and employees' characteristics on policies and wages using a semiparametric propensity-score weighing that does not assume a parametric outcome equation for the determinants of individual wages. Furthermore, the paper makes a causal interpretation on the potential effect of unobserved employer characteristics assuming that such characteristics affect the mean wages of employees but not the gender wage gap within each firm. The analytical results are as follows:

(1) Under the GEO policy, mean female wage increases and the gender wage gap decreases.

(2) The effects of both the systematic WLB policy and the restricted regular employment policy depend on the presence of the GEO policy. If the GEO policy is present, both the WLB policy as well as the employment policy increase mean female wage and decrease the gender wage gap beyond the effect of the GEO policy.

(3) Without the GEO policy, the presence of a systematic WLB policy instead increases the gender wage gap, while the restricted regular employment policy has no significant effect on the wage gap.

(4) Hence, the WLB policy should be considered a double-edged sword, where its effect on the gender-wage gap reverses based on the presence of the GEO policy. On the other hand, the restricted regular employment policy has no such risk; it increases women's economic empowerment when combined with the GEO policy, and shows no detrimental effect without.