Estimating the effects of pronatal policies on residential choice and fertility

In this paper, we estimate the impacts of local-government-sponsored pronatal policies on fertility by exploiting the geographical variation in policies across municipalities in Japan. We develop an empirical model that accommodates both the location and fertility choices of households to take into account their self-selected migration across municipalities. We estimate the model using microdata on households in metropolitan areas. The results suggest that self-selection may generate substantial upward bias in the estimated impacts of pronatal policies on fertility. We also find that some types of noncash benefit pronatal policies significantly increase the probability of births occurring in metropolitan households.