主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2018, Vol. 26 ›› Issue (6): 1-7.doi: 10.16381/j.cnki.issn1003-207x.2018.06.001

• 论文 •    下一篇

基于危机条件概率的系统性风险度量研究

朱晓谦1, 李靖宇1,2, 李建平1, 陈懿冰1, 魏璐1,2   

  1. 1. 中国科学院科技战略咨询研究院, 北京 100190;
    2. 中国科学院大学, 北京 100049
  • 收稿日期:2016-11-22 修回日期:2017-03-22 出版日期:2018-06-20 发布日期:2018-08-22
  • 作者简介:李建平(1976-),男(汉族),浙江建德人,中国科学院科技战略咨询研究院研究员,博士,研究方向:风险管理,E-mail:ljp@casipm.ac.cn.
  • 基金资助:

    国家自然科学基金资助项目(71425002,71601178,71433001);中国科学院青年创新促进会资助项目(2017200,2012137)

An Indicator of Conditional Probability of Crisis for Systemic Risk Measurement

ZHU Xiao-qian1, LI Jing-yu1,2, LI Jian-ping1, CHEN Yi-bin1, WEI Lu1,2   

  1. 1. Institutes of Science and Development, Chinese Academy of Sciences, Beijing 100190, China;
    2. University of Chinese Academy of Sciences, Beijing 100049, China
  • Received:2016-11-22 Revised:2017-03-22 Online:2018-06-20 Published:2018-08-22

摘要: 2007年次贷危机的爆发,令系统性风险的度量受到了广泛关注,但是目前常用的度量方法存在多种问题,不能较好地反映金融业系统性风险的实时变化。本文提出一种新的系统性风险度量方法——危机条件概率(Conditional Probability of Crisis,CPC),将系统性风险定义为单个金融机构发生危机导致整个金融系统也陷入危机的概率,可以利用股票收益的下尾相关性计算得出。该方法概念清晰,较好地体现了系统性风险的内涵,并且可得到实时更新的系统性风险。实证基于中国49家上市金融机构的股票价格数据,得出了2007-2016年我国金融业及金融子行业系统性风险。结果显示:2014年下半年以来,中国整个金融行业的系统性风险呈明显上升的趋势,目前甚至已经显著高于次贷危机时期;证券业系统性风险在样本时间范围内一直呈显著上升趋势;银行业对金融行业的影响最大,证券业和保险业的影响力也在逐步上升。

关键词: 系统性风险, 风险度量, 尾部相关性, 股票收益率, Copula

Abstract: The 2007 subprime mortgage crisis has arouse wide concern about systemic risk measurement, However, the commonly used measures could not reflect the real-time system risk of financial industry well due to their limitations.
In this paper, an indicator of Conditional Probability of Crisis (CPC), which defines systemic risk as the probability of a systemic crisis given the crisis of a single financial institution, is proposed. It can be calculated by the lower tail dependence between stock returns of the financial institutions and the financial system. The efficient-markets hypothesis states that asset's prices fully reflects all valuable information. Firstly, based on the market model, the common part affected by the market factor is removed from the stock returns to derive the heterogeneous parts affected by the firm-specific factor. Then, the Clayton copula is used to estimate the lower tail dependencies between the stock returns of every financial institution and financial system. Lastly, the CPC value of the financial system is calculated by averaging all these tail dependencies. This value reflects the average probability that the crisis of a single financial institution will lead to the crisis of the entire financial system. On one hand, the proposed indicator embodies the connotation of systemic risk very clearly. On the other hand, it can be obtained timely and is comparable across different time points, which is thus useful for monitoring the real-time systemic risk.
In the empirical study, based on the data of 49 Chinese listed financial institutions, including banks, security companies and insurance companies, the systemic risk of the entire Chinese financial industry and its three sub-industries from January 2007 to June 2016 is obtained respectively. The empirical results show that:firstly, the systemic risk of Chinese financial industry has an obvious upward trend since June 2014, and nowadays it's even much higher than that during the subprime mortgage crisis. In addition, the security industry experiences a sustained increase in systemic risk during the sample period. At last, the banking industry always imposes the greatest impact on the systemic risk of the entire financial industry, while the impacts of the securities and the insurances have increased gradually during these years.
The proposed indicator, CPC, could be a competitive alternative measure for systemic risk and the empirical results on Chinese financial industry also provides valuable references to the following researches and the financial regulation in China.

Key words: systemic risk, risk measurement, tail dependence, stock return, copula

中图分类号: