This study aims to assess the potential effects and the potential targets of possible pronatalistic family policies in Japan. Proportional hazards and binomial logit models have been applied to the data from the Eighth National Fertility Survey (with 8,000 samples) contacted by the Institute of Population Problems in 1982. As a dependent variable, the third birth probability (interval) has been emphasized because it is often the target of pronatalistic family policies and because most Japanese couples have had at least two children anyway. The excess of the ideal over the expected family size is also emphasized because those couples with an excess are assumed to be those in need of policy supports to attain the goals and because the government can legitimize the intervention targeted toward those needy couples. We have focused on couple's income, wife's employment, and housing situation as policy-related independent variables after controlling for marriage cohort, wife's age at first marriage, mate selection method, postnuptial residence, wife's education, husband's occupation, urban-rural residence, and region.
The following results have been obtained regarding these policy-related variables. The couple's income does not have any significant effects on the cumulative probability of the third birth in the total, but the upper middle income has a significant and negative effect in a more recent marriage cohort. The upper middle income also has a positive effect on an excess of the ideal over, the expected family size, which is also observed in a more recent marriage cohort. These results seem to suggest that financial support (the increase in child allowance or tax break for the third child) targeted toward the couples in the upper middle income bracket may have pronatalist effects.
The wife's full-time employment has a significant and negative effect on the cumulative probability of the third birth and a significant and positive effect on an excess of the ideal over the expected number of children. The wife's part-time employment also has a negative effect on the third birth. These results suggest that the measures to help working women (and their spouse) to harmonize work and family life, including the improved supply of child-care services, may have pronatalist effects.
Living in a parents' housing or on a parents' land has a significant and positive effect on the cumulative probability of the third birth, which means that all the other types of home ownership (non-ownership) has a negative effect on the third birth. Living in an owned condominium also has a negative effect on the third birth. These results imply that the improved provision of public housing and/or the introduction of housing allowance (tax break for rent) can be possible pronatalist measures if the reduction of inheritance or gift tax is not desirable for other policy goals.