Understanding Intercity Mobility Patterns in Rapidly Urbanizing China, 2015-2019: Evidence from Longitudinal Poisson Gravity Modeling

Whereas much academic effort has been devoted to the physics and geographies of daily intraregion individual movements using new big data on human locations, systematic econometric modeling of the spatiotemporal logic of periodic interregional mobility has received limited attention. Using a multiyear, location-based service data set of daily intercity mobility from the Internet company Tencent, this study systematically examines China's intercity mobility patterns between 2015 and 2019 for the first time. Following a conceptual framework, a Poisson pseudo-maximum likelihood estimation (PPML) gravity approach is applied. It reveals a stable "diamond" pattern of high-value mobility flows among the four vertexes of Beijing, Shanghai, Guangzhou/Shenzhen, and Chengdu, embedding radiation patterns of flows connecting some large cities to their neighbors. The econometric results indicate the influence of gravity factors, short-term trips, long-term mobility tendencies, and transportation facilities. Factors of origin and destination exert the same effects on mobility, implying a circulation character. Results of subsample heterogeneity analysis (urban agglomerations vs. nonurban agglomerations, larger cities vs. smaller cities) and the moderating effects of time, distance, and economy are discussed. Our findings reveal differences between intercity mobility and traditional migration under the hukou system and propose implications for urban governance in the postepidemic era.