Labor supply response and preferences specification: Estimates for prime-age males in Japan

In this paper, we estimate labor responses of Japanese prime-age males by taking into consideration the Japanese income tax system and utilizing a large micro-data set. We employ three maximum likelihood methods: (i) a modified version of (Hausman, 1979) and (Hausman, 1981), which assumes a linear labor supply function; (ii) that given by Zabalza (1983), which assumes CES preferences; and (iii) that given by Van Soest (1995), which employs the translog utility function and assumes discrete labor hour choice. While the estimates based on the Hausman and Zabalza methods fare poorly, those based on the Van Soest method result in more plausible labor responses. However, these responses are larger than those of the North American and European counterparts.