摘要
银行业顺周期问题是巴塞尔资本协议Ⅲ中提出的宏观审慎两大监管问题之一,由于行业的特异性,顺周期问题在不同行业的表现显著不同。本文以穆迪公司的分行业违约概率数据为对象,基于线性和非线性因果检验,考察了违约概率的顺周期问题。结果表明,违约概率和工业增加值之间不存在明显的线性因果关系,但双向非线性因果关系十分显著;违约概率和宏观景气指数之间仅存在前者到后者的非线性因果关系;9个重点行业的违约概率和工业增加值之间存在前者到后者或双向的非线性因果关系。违约概率引领经济指标体现了其顺周期性,一定程度上对宏观经济产生影响。正确合理地评估内部模型的顺周期性,从而降低银行部门的顺周期效应,对于监管部门具有重要意义。
Procyclicality of banking industry is one of the two macroprudential questions proposed in Basel Ⅲ. Because of specificity of industry, procyclicality shows significant differences in different industries. Based on Moody＇s probability of default data of different industries, we empirically test the procyclicality of the probabilityof default using linear and nonlinear Granger causality test. The result shows that there is no linear Granger causality between the probability of default and the added value of industry, yet two-way nonlinear Granger causality between the two; there is only nonlinear Granger causality from the probability of default to the prosperity index; there are one-way（the former to the latter） or two-way nonlinear Granger causality between the probability of default of 9 important industries and the added value of industry. Probability of default causing economic indexes reflects its procyclicality and exerts an effect onthe macro economy. Therefore evaluating the procyclicality of internal model rightly and properly in order to reduce the procyclicality of banking sector will be of great importance for regulators.
作者
李楠
陈暮紫
LI Nan;CHEN Mu-zi(School of Management Science and Engineering,Shandong Normal University,Shandong Jinan 250014,China;School of Management Science and Engineering,Central University of Finance and Economies,Beijing 100081,China)
出处
《数理统计与管理》
CSSCI
北大核心
2018年第5期815-827,共13页
Journal of Applied Statistics and Management
基金
国家自然科学基金(71673315,71203247,71704098)
山东省自然科学基金(ZR2016GQ03,ZK2016GB03)
北京市社科基金(16YJB036)
山东省软科学项目(2017RKB01138)
广西石化资源加工及过程强化技术重点实验室开放课题基金的资助
关键词
巴塞尔资本协议Ⅲ
违约概率
顺周期
工业增加值
非线性因果检验
Basel Ⅲ
probability of default
procyclicality
added value of industry
nonlinear Granger causality test
