Policy failure or success? Detecting market failure in China's housing market

• Published in: Economic modelling 2016-08, Vol.56, p.109-121
• Main Author: Chen, W.D.
• Format: Article
• Language: English
• Subjects: Bootstrap method; Economic models; Housing; Housing market; Housing prices; Market failure; Mean reversion; Persistency; Public policy; Studies; Transaction cost; Vacancies
• ISSN: 0264-9993; 1873-6122
• DOI: 10.1016/j.econmod.2016.03.024

Due to a high vacancy rate of residential homes, housing prices remain sticky in most urban areas of China, which causes higher searching and bargaining costs. With an inefficient outcome, deadweight loss and market failure arises. To assess the Chinese government's housing policies in 2010, we develop a dynamic equilibrium model, in which we demonstrate how the sticky price results in market failure. We apply a multiple-factor panel data model to show that a high degree of market failure is associated with a high ratio of persistent components in the gap between price and equilibrium. As the persistent components will cause the market's instability, we can use the ratio between persistent and mean reverting components as an indicator to supervise the status of the housing market. We investigate the new and second-hand markets in 19 major cities, including 4 municipalities and 15 vice-provincial cities. Through our multiple-factor model, we explore the situation for each city. The results indicate these policies did improve the housing market's efficiency. It is therefore useful for the Chinese government to extend these policies to other areas to include not only big cities, but entire provinces which can improve its economic system efficiency and fairness even when its economic growth is slowing down. •We apply our model on the new and second-hand house markets in China.•We assess the Chinese government's housing policies in 2010.•We find these policies did improve the house markets' efficiency.•We demonstrate how the sticky price results in increased searching and bargaining costs.•A multiple-factor panel data model is developed.