Intergovernmental Competition, Labor Mobility And Housing Price Differentials Across Cities In China

The intergovernmental competition behavior of China’s local governments has a tremendous impact on housing price differences by influencing a variety of behaviors and factors, including the residential land supply constrain and the demand factors such as wages, population and quality of life, just to name a few. Although a lot of urban economic literatures have focused on the relationship between demand factors and residential price differences, studies from the aspect of land supply are limited. Given the importance of supply factors in explaining the urban price differences, China’s special land market structure provides an excellent chance for us to further identify the extent to which the land supply behavior can be critical in determining the the residential land price differences across cities.The issue this paper aims to address is how intergovernmental competition behavior of China’s local governments has affected housing price differentials across Chinese cities. This paper first constructs a general equilibrium model in which, local governments under regional economic growth competition, on the one hand, supply industrial land as a production input to facilitate the urban economic growth and, on the other hand, use residential land sales as a financing tool of public infrastructure investment to increase the level of amenities in the city to attract mobile factors.The model shows that:(1) In equilibrium, local governments’productive land suppy takes a constant part of the total urban land supply. And the more government’s over-emphasis on economic growth, government’s land supply will be more industry-preferred.(2) There are two type of effect within the intergovernmental competition.The first is "competition effects" which makes the housing price higher. The other is "polarization effects" which makes the housing price differentials across cities larger.To test these hypotheses, we apply the most recently developed panel unit root, a panel cointegration and panel based error correction model on the panel data of35large and medium-sized cities for the period1999-2010. The empirical results confirm a long-run cointegrated relationship between the series of the land supply structure and government’s growth-enhancing preference as well as between the series of the residential land price and the level of amenities. The empirical results suggest variation in land supply structure plays a considerable role in explaining the surge of residential land price. The policy implication from this study is significant. To correct the distortions in both housing market and land market, government’s manipulation of land resource allocation should be avoided and under-pricing of land input in the industrial production should be gradually eliminated.