基于空间关联的区域房价泡沫化传染风险及长效调控政策研究

Currently, the risks caused by bubble become an important issue to be solved in the real estate market. In order to solve this problem, it is necessary to find out the internal correlation and external contagion mechanism of housing price bubble. The proposed research project starts from the measurement and spatial correlation of regional housing price bubble, presents the definitions, existing way, test method of cross-section dependence. It aims at investigating the risk contagion mechanism within and between real estate market, exploring the interaction of direct transmission channels and space transmission channels, so as to systematically revealing the basic rule of contagion within real estate market, financial system, and macroeconomic. Successively, we will build a single equation cross-section dependence dynamic panel data (CS-DPD) model and multi-equations globe vector auto-regression (GVAR) model. Based on the rule of risk contagion, we using two-regime spatial durbin model (TRSDM) measure the effect of regulatory policies in limiting the spread of house price bubbles. We will also propose some differentiated regulatory policies and targeted prevention measures. The work is an innovative exploration to promote stable and healthy development of China's real estate market which incorporate the contagion problem of housing price bubble within the real estate market, financial system, and the macroeconomic into the same framework. Thus it also provides theoretical support to the stable economic growth policies designing.

当前，警惕和化解“泡沫化”引起的各类风险，已成为房地产市场亟需解决的重要问题，“去泡沫化”也将成为下一阶段政策调控的重要目标。为有效解决这一问题，需要探明房价泡沫化的内部关联特征与外部传染机制。本项目拟从区域房价泡沫化的度量与空间关联特征出发，探讨区域房价泡沫化传染风险在市场内和市场间的形成机制，剖析直接传染途径和空间传染途径及其交互过程；构建区域房价泡沫的截面相关动态面板数据模型和全局向量自回归模型，进而系统地揭示区域房价泡沫化在房地产市场、金融市场、宏观经济中的传染规律；在此基础上，运用两区制空间杜宾模型，测度各类调控政策在限制房价泡沫传染方面的效果，并提出差异化调控政策和针对性防范措施。本课题是在充分考虑空间关联下，将区域房价泡沫化在房地产市场、金融市场、宏观经济的传染问题纳入同一框架研究的创新性探索，可为促进我国房地产市场平稳健康发展和宏观经济稳定有序运行的政策设计提供理论支持。

进入新世纪以来，特别是次贷危机爆发以后，房价的持续上涨引发了对房价泡沫化问题的广泛担忧。与此同时，由于房地产市场间的密切联动及房地产业的关联特性，区域性房价泡沫会迅速蔓延至全国范围，甚至诱发区域性和系统性金融风险。对此，国家领导人与政府工作报告中多次提出“房住不炒”和“促进房地产市场平稳健康发展”，严控房价过快上涨。本项目基于空间关联特征视角，从房地产属性与房价泡沫测度出发，系统地探讨了区域房价的异质性空间传染机制、房价持续上涨的金融传导机制和房价泡沫的空间传染机制，揭示了房价及房价泡沫在房地产市场和金融市场的传染渠道与方式；创新性地提出了能体现非对称空间关联方式的扩展空间杜宾模型（ESDM）与估计方法，并基于行政边界给出了两个具体应用场景；针对性地分别运用扩展空间杜宾模型、社会网络分析法(SNA)和全局向量自回归模型(GVAR)检验了区域房价在地理空间上的异质性传染方式、城市房价泡沫的空间网络结构特征与房价泡沫在金融市场的风险传染过程。研究结果表明，空间传染途径是房价泡沫在房地产市场与金融市场传染的重要渠道，房价泡沫出现后，行政边界无法发挥其屏蔽效应，忽视房价泡沫的空间传染将会低估其区域性与系统性风险。本项目的研究在理论上丰富了空间计量领域的理论模型和房价泡沫领域的研究方法，具有重要的学术价值；在应用上可为促进我国房地产市场平稳健康发展和宏观经济稳定有序运行的政策设计提供理论支持，具有较强的现实意义。