Reproductive Contributions of Foreign Wives in Taiwan: Similarities and Differences among Major Source Countries

In light of the entrenchment of sub-replacement fertility and the sharp increase in the stock of foreign wives in Taiwan in recent years, this research studies the reproductive contributions of Taiwan’s foreign wives from the top five source countries (China, Vietnam, Indonesia, Thailand, and the Philippines), based mainly on an application of a multinomial logit model to the micro data of the 2003 census of foreign wives. Our main findings are as follows. First, the overall fertility level of the foreign wives was probably somewhat higher than that of the native-born women and definitely lower than the replacement level. Second, among the five nationalities, those from China were much less reproductive than those from the other countries, mainly because the former were more prone to (1) having a rather old marriage age, (2) having a very large spousal age gap, (3) being separated or divorced, (4) having their current marriage being their second marriage, and (5) having a veteran as the husband. Third, among the four Southeast Asian nationalities, those from Indonesia and the Philippines were more reproductive than those from Thailand and Vietnam. This contrast was a muted reflection of the fertility difference in countries of origin. Fourth, for every nationality, marriage duration and marriage age were the most powerful explanatory factors and must be included in the model to avoid getting misleading estimated coefficients of other less powerful explanatory factors, whereas current age was a spurious factor that should not be used in the model. Fifth, in the context of marriage duration and marriage age, the explanatory factors with rather strong explanatory powers for at least one nationality included spousal age gap, marital status, remarriage status, co-residence with parent, and wife’s employment status. Sixth, the expected negative effect of wife’s educational attainment on lifetime fertility turned out to be either non-existent or modest. In particular, it had practically no effect on the probability of being childless. These findings implied that getting better educated foreign wives could increase the quality of their children with little or no reduction in the number of their children and in their probability of being childless