Population Mobility Structural Analysis and Population Estimation Using a Quantitative Spatial Model

This article proposes a population forecasting using endogenous population model, which incorporates a spatial model considering the spatially heterogeneous feature of agents and the economy by employing quantitative spatial models. In this endogenous population model, agents migrate to maximize their utility. This model is estimated using the twostage estimation approach, which is extensively used in quantitative spatial literature. Estimated parameters are significant and almost consistent with the economic and demographic stylized facts. Using the parameters concerning migration and local utilities, we conduct projection analyses for 2015-2125 across all prefectures in Japan, which is now experiencing regionally asymmetric population decline. In the baseline projections, the population in less populated prefectures is mitigated slightly by introducing the young generation's migration behavior. Counterfactual analyses are then conducted to break down the factors of population decline in Japan. Among several factors, birth abandonments due to some constraints and slow productivity growth after 1995 turned out to have severely impacted demographics. The development of networks resulted in having negative impacts on demographics though they had positive impacts on the welfare of economic agents in many aspects.