LAUSR

Abstract Although disputed, population is often used to measure shrinking cities. Facing by this challenge, this paper proposes an alternative indicator, average human capital (HC), to investigate city vitality. The improved measurement relaxes the traditional assumption that HC has a linear relation with schooling, and combine inputs with outputs of HC by considering the heterogeneity of returns to schooling of each city. Empirical analysis was performed by constructing a comprehensive dataset and calculating average HC level for each city in the year of 2000, 2005 and 2010, our results suggest that city shrinkage might not be so serious as emphasized in previous literature based on population decline. But average HC across Chinese cities has been diverging since 2005. We also validate that our measurement is substantially better than population for explaining economic vitality in cities. The article sheds a new insight on the understanding of urban growth process.