中国金融周期与经济周期的非对称波动特征及联动关系研究  

Analysis on the Asymmetric Fluctuation Characteristics and Linkage Relationship between Financial Cycle and Business Cycle in China

作　　者：孙晨童 王相飞 丁新 SUN Chentong;WANG Xiangfei;DING Xin(School of Economics and Management,Tsinghua University,Beijing 100084,China;Research Institute,Fareast Credit Rating Co.,Ltd,Beijing 100007,China;School of Finance,Jilin University of Finance and Economics,Changchun 130117,China;Zhongguancun Smart City Co.,Ltd,Beijing 100080,China;School of Management and Economics,Beijing Institute of Technology,Beijing 100081,China)

机构地区：[1]清华大学经济管理学院,北京100084 [2]远东资信评估有限公司研究院,北京100007 [3]吉林财经大学金融学院,吉林长春130117 [4]中关村科学城城市大脑股份有限公司,北京100080 [5]北京理工大学管理与经济学院,北京100081

出　　处：《统计与信息论坛》2024年第2期43-55,共13页Journal of Statistics and Information

基　　金：国家社会科学基金重大项目“宏观经济稳增长与金融系统防风险动态平衡机制研究”(19ZDA094);国家自然科学基金项目“多重不确定性冲击下金融复杂系统韧性评估与提升研究”(72301059);中国博士后科学基金项目“多重不确定性冲击下‘实体经济-金融系统’双层耦合网络的韧性研究:测度方法及提升路径”(2023M742609)。

摘　　要：区别于传统的经济周期研究,应用马尔科夫区制转换动态双因子(MS-DBF)模型,实现了对中国金融周期与经济周期的测度、波动状态识别以及关联关系的分析。这对于同时监测中国宏观经济与金融部门的周期波动态势以及通过周期视角探究宏观金融关联关系都具有重要的理论价值和现实意义。研究发现:第一,金融周期与经济周期波动具有非对称特征,并且金融周期的波动程度小于经济周期;第二,无论是扩张还是收缩阶段,金融周期对经济周期都具有一定的先行性,但每轮循环的先行期长短不一;第三,两个周期的联合波动具有区制依赖特征,并且不同状态之间的转移概率存在非对称性;第四,由于新冠疫情冲击导致宏观经济数据异常波动,使得传统MS-AR和MS-VAR模型都无法准确地识别经济周期阶段,而MS-DBF模型依托联合转移概率矩阵能够准确地捕捉到经济周期的每一轮波动。2021年下半年以来,中国经济发展面临“三重压力”,经济周期与金融周期正处于“前者收缩、后者扩张”的联合区制状态,经济周期走势深度探底;金融周期虽然处在扩张阶段,但仍位于较低水平。这预示着,现阶段中国系统性金融风险总体可控,宏观调控仍然具有一定的政策调控空间,政府部门可以适度加大政策调控力度以实现经济平稳增长。Accurately grasping the cyclical fluctuations in the financial and economic sectors is a crucial prerequisite for enhancing the precision of macroeconomic regulation and building a sound macroeconomic governance system.Different from traditional business cycle research,the Markov Switching Dynamic Bi-factor(MS-DBF)modelis applied to develop a comprehensive analysis to measure China’s financial and business cycles,identify the volatility,and determine the correlation.It is of great theoretical value and practical significance for simultaneously monitoring the cyclical fluctuation dynamics of macroeconomic and financial sectors in China,as well as exploring macro-financial linkages through a cyclical perspective.The results show that,firstly,the financial and business cycles have asymmetric characteristics,and the degree of volatility in the financial cycle is less than that of the business cycle,the possible reason is that the impact of the pandemic led to significant fluctuations in macroeconomic data after 2020,while financial data is not heavily affected by the pandemic,and the volatility remains within the normal range.Secondly,whether it is expansion or contraction,the financial cycle has a certain advancement in the business cycle,but the advancement period of each cycle is different,with the average leading period during contraction phases slightly exceeding that during expansion phases.Thirdly,the combined fluctuation of the two cycles has the characteristics of regime dependence,and the probability of transition between different states is asymmetric.Finally,due to the impact of the Covid-19 pandemic,macroeconomic data exhibits unusual fluctuations,rendering traditional Markov Switching Autoregressive(MS-AR)and Markov Switching Vector Autoregressive(MS-VAR)models incapable of accurately identifying the phases of the business cycle,while the MS-DBF model can accurately capture each round of fluctuations in the business cycle under the constraint of the joint transfer probability matrix.Since the second h

关 键 词：金融周期 经济周期 周期波动的非对称性 联动关系 稳健性 MS-DBF模型 

分 类 号：F037.1[经济管理—政治经济学]