Understanding Clusters in China's Real Estate Market

The paper studies the convergence of regional house prices in China. Using a non-linear time-varying factor model and monthly prices for various property types in China from January 2004 to June 2021, we investigate the evolution of housing prices in China and test for price convergence across 30 major cities. We find that the price convergence patterns vary a lot across different housing types. The commercial property market is more homogeneous, with 29 out of 30 cities converging over time. At the same time, there is substantial heterogeneity in residential property prices. The formation of convergence clubs varies over time and across types of properties. The market for newly-constructed housing becomes more integrated over time, with a smaller number of regional clusters identified in the latter period, while the second-hand housing market becomes more fragmented. Using regression analysis, we find that rent, land supply, air quality, and economic development of cities play an important role in explaining house price convergence across cities.