Performance-pay and the gender wage gap in Japan

• We examine the gender wage gap by performance-pay group across all distribution.
• A glass ceiling effect is observed for workers without performance pay wage.
• The raw gender wage gap among the performance-pay group is greater.
• Promotion difference explains the gender wage gap at the top distribution.

Using the Japanese Survey of Living Preferences and Satisfaction, we examine the gender wage gap by performance-pay group across the whole earnings distribution in Japan. The main finding is that a sharp acceleration of the gap at the top is not observed in Japan when we use a sample of all workers. However, a glass ceiling effect is observed for white collar workers who do not receive performance-based pay. On the other hand, the feature of maintaining a raw gender wage gap at about the same level above the 60 percentile is observed for white collar workers who receive performance pay. In addition, we find that the raw gender wage gap among the performance-pay group is about 5–18 points greater across the wage distribution than that among the non-performance-pay group. A second finding is that after performing the counterfactual decomposition analysis introduced by Machado and Mata (2005), differences in promotion between women and men explain the gender wage gap at the top of the distribution for non-performance pay workers. On the other hand, the gender difference for those working in large companies explained the gender wage gap at the top distribution for performance pay workers.