基于Copula的财险公司应收保费信用风险度量
|
摘要
2008年由美国次贷危机引发的全球金融危机后,信用风险成为全球关注的焦点。各金融机构包括保险公司都在积极研究如何有效地防范和化解信用风险。信用风险具有非对称、非线性及尾部相关等特征,度量该风险主要是度量违约相关性,而常见的线性相关系数已不再适合描述信用风险的相关信息。Copula函数可以捕捉较为复杂的违约相关性,特别是在刻画尾部相关性方面。
本文对财险公司的信用风险进行研究,将应收保费信用风险作为研究重点。财险公司代理人的违约是应收保费信用风险的主要原因,是国内财险公司主要的信用风险。为了更好的度量应收保费信用中的各种相关性,本文采用两种不同的Copula函数,对财险公司的整体以及划分不同业务线后的信用风险进行度量。
测算的结果证明：Copula函数在信用风险度量时能够很好地捕捉变量及业务线间的相关性；引入Copula后,测算出的财险公司所需的资本金变大。与Gaussian Copula相比,t-copula能更好地度量尾部相关性。数据显示应收保费信用风险会给财险公司带来一定的损失,并且不同的业务线因为应收保费信用风险造成的损失程度各不相同。在各业务线中,企财险的损失大大超出了其他几条业务线的损失之和,而业务量占比最多的车险带来的损失则非常少。财险公司要针对不同业务线中的风险大小对产品结构进行调整。
本文最后针对财险公司应收保费存在的问题,提出了几点管理上的建议,希望对财险公司的合理管控有所帮助。
In 2008, the credit risk attracted worldwide attention due to the global financial crisis triggered by the subprime mortgage crisis in the United States. The governments and the financial institutions including insurance corporations in the world are actively considering how to effectively measure and prevent credit risk. Credit risks display non-linear, non-symmetric and tail dependence.Measuring default correlation is one of the important elements in credit risk management.But, the linear correlation coefficient has been no longer suit to describe the default correlation of credit risk. Copula function can be used to capture the complex part of default correlation, particularly in the tail dependence.
This paper choose premium receivable credit risk as analysis focus and consider the default of the insurance agents is the first reason of the premiums receivable, which is one of the most important credit risks in domestic insurance companies.To better measure the correlation among the premium receivable credit risk, two types of Copula fuctions are adopt to calculate capital of the credit risk in different business lines, as well as the one for whole property insurance company.
The research results reveal that Copula functions capture the correlation among the factors and different business lines,the capital of the company turn to a large number. Compared to Gaussian Copula, t-copula has advantage in capturing the tail dependence. The data give us the information that the property insurance company suffered from many losses caused by the premiums receivable credit risk, but the losses of different business lines were different. The loss of business property insurance is far beyond the summation of all the other business lines. Even occupying the majority of total premiums, auto insurance needs very small capital.So the property insurance company should adjust business structure based on the risk among different business lines.
The last part of this paper gives the property insurance company some proposals on how to control the premiums receivable credit risk, on the purpose of a better management of the credit risk.
本文对财险公司的信用风险进行研究,将应收保费信用风险作为研究重点。财险公司代理人的违约是应收保费信用风险的主要原因,是国内财险公司主要的信用风险。为了更好的度量应收保费信用中的各种相关性,本文采用两种不同的Copula函数,对财险公司的整体以及划分不同业务线后的信用风险进行度量。
测算的结果证明：Copula函数在信用风险度量时能够很好地捕捉变量及业务线间的相关性；引入Copula后,测算出的财险公司所需的资本金变大。与Gaussian Copula相比,t-copula能更好地度量尾部相关性。数据显示应收保费信用风险会给财险公司带来一定的损失,并且不同的业务线因为应收保费信用风险造成的损失程度各不相同。在各业务线中,企财险的损失大大超出了其他几条业务线的损失之和,而业务量占比最多的车险带来的损失则非常少。财险公司要针对不同业务线中的风险大小对产品结构进行调整。
本文最后针对财险公司应收保费存在的问题,提出了几点管理上的建议,希望对财险公司的合理管控有所帮助。
In 2008, the credit risk attracted worldwide attention due to the global financial crisis triggered by the subprime mortgage crisis in the United States. The governments and the financial institutions including insurance corporations in the world are actively considering how to effectively measure and prevent credit risk. Credit risks display non-linear, non-symmetric and tail dependence.Measuring default correlation is one of the important elements in credit risk management.But, the linear correlation coefficient has been no longer suit to describe the default correlation of credit risk. Copula function can be used to capture the complex part of default correlation, particularly in the tail dependence.
This paper choose premium receivable credit risk as analysis focus and consider the default of the insurance agents is the first reason of the premiums receivable, which is one of the most important credit risks in domestic insurance companies.To better measure the correlation among the premium receivable credit risk, two types of Copula fuctions are adopt to calculate capital of the credit risk in different business lines, as well as the one for whole property insurance company.
The research results reveal that Copula functions capture the correlation among the factors and different business lines,the capital of the company turn to a large number. Compared to Gaussian Copula, t-copula has advantage in capturing the tail dependence. The data give us the information that the property insurance company suffered from many losses caused by the premiums receivable credit risk, but the losses of different business lines were different. The loss of business property insurance is far beyond the summation of all the other business lines. Even occupying the majority of total premiums, auto insurance needs very small capital.So the property insurance company should adjust business structure based on the risk among different business lines.
The last part of this paper gives the property insurance company some proposals on how to control the premiums receivable credit risk, on the purpose of a better management of the credit risk.
引文
[1]Basel Committee on Banking Supervision.International Convergence of Capital Measurement and Capital Standards [EB].www.bis.org/publ/bcbs128.pdf,2005-11
[2]International Actuarial Association.A Global Framework for Insurer Solvency Assessment[EB]. www.actuaries.org/library/papers/Global_Framework_Insurer_Solvenc y_Assessment-public.pdf,2004
[3]European Commission. Study into the methodologies to assess the overall financial position of an insurance undertaking from the perspective of prudential supervision [EB].http://www.pinggu.org/UploadFile_20082009/2008-6/2008629117381395.pdf, 2002-05
[4]保监会.保险公司风险管理指引[EB]. http://www.circ.gov.cn/tabid/106/InfoID/45352/ frtid/3871/Default.aspx.2007
[5]保监会.保险公司偿付能力额度及监管指标管理规定[EB]. http://www.circ.gov.cn/ tabid/106/InfoID/57554/frtid/3871/Default.aspx.2007
[6]Merton, Robert C, Andre A Perold. Theory of Risk Capital in Financial Firms[J]. Journal of Applied Corporate Finance,1993,1 (1):11-15
[7]New York:Morgan Guaranty Turst ComPany. J. P Morgn:risk Metrics-Technical[EB] www.document.down.cenet.org.cn/view.asp?id=23192,1996
[8]KMV Corporation. Modeling Default Risk[EB]. http://222.edu.pinggu.com/attachments/ uploadfile_20082009/2006-8/20068510372594067.pdf
[9]Credit Suisse Financial Products. CreditRisk+:A credit risk management framework [EB], http://www.cstb.com/institutional/research/assets/creditrisk.pdf,1997
[10]McKinsey and Company. Credit Portfolio View TM Approach Documentation and User's Documentation. Zurich:McKinsey and Company[R],1998
[11]Basle Committee on Banking Supervision. Credit risk modeling:current practices and applications [EB]. www.bis.org/publ/bcbs49.htm.1999
[12]Altman Edward I, Kishore,V. M. Defaults and Returns on High Yield Bonds:Analysis through 1996[J]. Special Report, New York University Salomon Center,N. Y.1997
[13]Altman, Resti,Sironi. The Link between Default and Recovery Rates:Theory, Empirical Evidence, and Implications[J]. Journal of Business,2005,78(6):2203-2227
[14]Report of Solvency Working Party [EB]. http://www.actuaries.org/CTTEES_INSREG/ Documents/Solvency_Report_EN.pdf,2002
[15]王春梅.保险公司经济资本管理应用研究[D].浙江工商大学硕士论文.2008：17-18
[16]IAA Solvency Assessment Working Party. A Global Framework for Insurer Solvency Assessment[EB]. International Actuarial Association. http://www.actuaries.org/ LIBRARY/papers/global_framework_insurer_solvency_assessment-public.pdf,2004
[17]Sklar A. Fonctions de repartition an dimensions et leurs marges [M]. Publication de insititut de Statistique de UniVersite de Paris,1959,229-231
[18]Nelson R B. An Introduction to Copulas[M]. New York:Springer,1998.
[19]Matteis, Fitting Copulas to Data. Diploma thesis,Zudch[R]. http://www.math.unizh.ch/ fileadmin/math/nser/Roberto/da/do/Diplomathesis-Copula.pdf,2001.
[20]Marting R. Credit portfolio modeling handbook. Suisse First Boston:The Quantitative Credit Strategis[R],2005
[21]Di Clemente,A. Romano,C. Measuring Portfolio Value-at-Risk bu a Copula-EVT Based Approach[J]. Studi Economoco,2005:29-57
[22]Kole et al. Selecting Copulas for risk management[J]. Journal of Banking and Finance,2007:2405-2423
[23]Jun Yan, Enjoy the Joy of Copulas:With a Package Copula[J]. Journal of Statistical Software 2007, (4):10,1-21.
[24]Zongwu Cai, Yanqin Fan, Xian Wang. Selection of Copulas with Applications in Finance[R], working paper.2008, (5).
[25]吴冲锋,程鹏.上市公司信用状况分析新方法[J].系统工程理论方法应用,2002,(11)：23-28
[26]梁琪.宏观经济环境与信用风险度量和管理研究[J].南开经济研究,1999,27(3)：62-68
[27]周昭雄.基于我国上市公司的KMV模型研究[J].工业技术经济.2006(07)：143-148
[28]张海燕.CreditRisk+模型与中国商业银行信用风险管理[J].吉首大学学报,2007,28(3)：119-123
[29]陈迪红,盛文文,林晓亮.财产保险公司代理人信用风险的度量[J].统计与决策,2009(03)：119-120
[30]刘莉,彭建坤.保险公司信用风险研究[J].特区经济,2005(6)：316—317.
[31]滕帆.中国保险公司经济资本估算[J].统计与信息论坛,2005,20(6)：33-36.
[32]滕帆.中国保险公司经济资本估算[J].统计与信息论坛.2005(6)：33—36.
[33]黄素,林晓亮.财险公司经济资本配置模型的构建与实证分析[J].浙江金融,2008(04)：48-49
[34]陈迪红,林晓亮.我国财险公司产品业务线经济资本配置的实证分析[J].财经理论与实践,2008(06)
[35]侯旭华.应收保费的风险防范与收帐对策[J].中国保险管理干部学院学报.1996(6)：29-31
[36]张清华.浅谈产险公司应收保费的管控[J].上海保险.2002(1)：15-17
[37]黄永波.论财产保险公司应收保费的成因和管理[J].保险职业学院学报.2005(06)：32-34
[38]刘玉焕.防范我国保费虚增的对策——改进保险会计新准则[J].吉林财税高等专科学校学报.2007(1)：48-50
[39]刘玉焕,张维.根除应收保费的建议——基于“见费出单”的思考[J].浙江金融2009(2)：47-48
[40]张云高垒.关于实施见费出单制度的几点思考[J].上海保险.2009(05)：29-30
[41]侯恩波.论财产保险公司的应收保费管理[J].内蒙古科技与经济.2009(16)：47-48
[42]朱克桐,张瑜.我国财产保险公司应收保费管控问题探讨——基于“见费出单”视角[J].保险职业学院学报.2010(1)：47-49
[43]张尧庭.我们应该选用什么样的相关性指标[J].统计研究,2002,(9)：41-44.
[44]张尧庭.连接函数(Copula)技术与金融风险分析[J].统计研究,2002,(4)：48-51.
[45]张明恒,多金融资产风险价值的Copula计量方法研究[J].数量经济技术经济研究,2004,(4),67-70.
[46]吴振翔,叶五一,缪柏其.基于Copula的外汇组合风险分析[J].中国管理科学,2004,12(4)： 1-5
[47]朱世武.基于Copula函数度量违约相关性[J].统计研究,2005,(4)：61-64.
[48]李健伦,方兆本,鲁炜,李红星.Copula方法与相依违约研究[J].运筹与管理,2005,14(3)：111-116.
[49]朱世武.基于Copula的度量与事后检验[J].数理统计与管理,2007(06)：984-991
[50]苏静,杜子平.Copula在商业银行组合信用风险度量中的应用[J].金融理论与实践,2008(05)：6-8
[51]朱世武.Copula函数度量我国商业银行资产组合信用风险的实证研究[J].金融研究,2009(04)：129-141
[52]赵先信.银行内部模型和监管模型——风险计量与资本分配[M].上海：上海人民出版社,2004
[53]杨军.银行信用风险--理论、模型和实证风险[M].北京：中央财政经济出版社,2004
[2]International Actuarial Association.A Global Framework for Insurer Solvency Assessment[EB]. www.actuaries.org/library/papers/Global_Framework_Insurer_Solvenc y_Assessment-public.pdf,2004
[3]European Commission. Study into the methodologies to assess the overall financial position of an insurance undertaking from the perspective of prudential supervision [EB].http://www.pinggu.org/UploadFile_20082009/2008-6/2008629117381395.pdf, 2002-05
[4]保监会.保险公司风险管理指引[EB]. http://www.circ.gov.cn/tabid/106/InfoID/45352/ frtid/3871/Default.aspx.2007
[5]保监会.保险公司偿付能力额度及监管指标管理规定[EB]. http://www.circ.gov.cn/ tabid/106/InfoID/57554/frtid/3871/Default.aspx.2007
[6]Merton, Robert C, Andre A Perold. Theory of Risk Capital in Financial Firms[J]. Journal of Applied Corporate Finance,1993,1 (1):11-15
[7]New York:Morgan Guaranty Turst ComPany. J. P Morgn:risk Metrics-Technical[EB] www.document.down.cenet.org.cn/view.asp?id=23192,1996
[8]KMV Corporation. Modeling Default Risk[EB]. http://222.edu.pinggu.com/attachments/ uploadfile_20082009/2006-8/20068510372594067.pdf
[9]Credit Suisse Financial Products. CreditRisk+:A credit risk management framework [EB], http://www.cstb.com/institutional/research/assets/creditrisk.pdf,1997
[10]McKinsey and Company. Credit Portfolio View TM Approach Documentation and User's Documentation. Zurich:McKinsey and Company[R],1998
[11]Basle Committee on Banking Supervision. Credit risk modeling:current practices and applications [EB]. www.bis.org/publ/bcbs49.htm.1999
[12]Altman Edward I, Kishore,V. M. Defaults and Returns on High Yield Bonds:Analysis through 1996[J]. Special Report, New York University Salomon Center,N. Y.1997
[13]Altman, Resti,Sironi. The Link between Default and Recovery Rates:Theory, Empirical Evidence, and Implications[J]. Journal of Business,2005,78(6):2203-2227
[14]Report of Solvency Working Party [EB]. http://www.actuaries.org/CTTEES_INSREG/ Documents/Solvency_Report_EN.pdf,2002
[15]王春梅.保险公司经济资本管理应用研究[D].浙江工商大学硕士论文.2008：17-18
[16]IAA Solvency Assessment Working Party. A Global Framework for Insurer Solvency Assessment[EB]. International Actuarial Association. http://www.actuaries.org/ LIBRARY/papers/global_framework_insurer_solvency_assessment-public.pdf,2004
[17]Sklar A. Fonctions de repartition an dimensions et leurs marges [M]. Publication de insititut de Statistique de UniVersite de Paris,1959,229-231
[18]Nelson R B. An Introduction to Copulas[M]. New York:Springer,1998.
[19]Matteis, Fitting Copulas to Data. Diploma thesis,Zudch[R]. http://www.math.unizh.ch/ fileadmin/math/nser/Roberto/da/do/Diplomathesis-Copula.pdf,2001.
[20]Marting R. Credit portfolio modeling handbook. Suisse First Boston:The Quantitative Credit Strategis[R],2005
[21]Di Clemente,A. Romano,C. Measuring Portfolio Value-at-Risk bu a Copula-EVT Based Approach[J]. Studi Economoco,2005:29-57
[22]Kole et al. Selecting Copulas for risk management[J]. Journal of Banking and Finance,2007:2405-2423
[23]Jun Yan, Enjoy the Joy of Copulas:With a Package Copula[J]. Journal of Statistical Software 2007, (4):10,1-21.
[24]Zongwu Cai, Yanqin Fan, Xian Wang. Selection of Copulas with Applications in Finance[R], working paper.2008, (5).
[25]吴冲锋,程鹏.上市公司信用状况分析新方法[J].系统工程理论方法应用,2002,(11)：23-28
[26]梁琪.宏观经济环境与信用风险度量和管理研究[J].南开经济研究,1999,27(3)：62-68
[27]周昭雄.基于我国上市公司的KMV模型研究[J].工业技术经济.2006(07)：143-148
[28]张海燕.CreditRisk+模型与中国商业银行信用风险管理[J].吉首大学学报,2007,28(3)：119-123
[29]陈迪红,盛文文,林晓亮.财产保险公司代理人信用风险的度量[J].统计与决策,2009(03)：119-120
[30]刘莉,彭建坤.保险公司信用风险研究[J].特区经济,2005(6)：316—317.
[31]滕帆.中国保险公司经济资本估算[J].统计与信息论坛,2005,20(6)：33-36.
[32]滕帆.中国保险公司经济资本估算[J].统计与信息论坛.2005(6)：33—36.
[33]黄素,林晓亮.财险公司经济资本配置模型的构建与实证分析[J].浙江金融,2008(04)：48-49
[34]陈迪红,林晓亮.我国财险公司产品业务线经济资本配置的实证分析[J].财经理论与实践,2008(06)
[35]侯旭华.应收保费的风险防范与收帐对策[J].中国保险管理干部学院学报.1996(6)：29-31
[36]张清华.浅谈产险公司应收保费的管控[J].上海保险.2002(1)：15-17
[37]黄永波.论财产保险公司应收保费的成因和管理[J].保险职业学院学报.2005(06)：32-34
[38]刘玉焕.防范我国保费虚增的对策——改进保险会计新准则[J].吉林财税高等专科学校学报.2007(1)：48-50
[39]刘玉焕,张维.根除应收保费的建议——基于“见费出单”的思考[J].浙江金融2009(2)：47-48
[40]张云高垒.关于实施见费出单制度的几点思考[J].上海保险.2009(05)：29-30
[41]侯恩波.论财产保险公司的应收保费管理[J].内蒙古科技与经济.2009(16)：47-48
[42]朱克桐,张瑜.我国财产保险公司应收保费管控问题探讨——基于“见费出单”视角[J].保险职业学院学报.2010(1)：47-49
[43]张尧庭.我们应该选用什么样的相关性指标[J].统计研究,2002,(9)：41-44.
[44]张尧庭.连接函数(Copula)技术与金融风险分析[J].统计研究,2002,(4)：48-51.
[45]张明恒,多金融资产风险价值的Copula计量方法研究[J].数量经济技术经济研究,2004,(4),67-70.
[46]吴振翔,叶五一,缪柏其.基于Copula的外汇组合风险分析[J].中国管理科学,2004,12(4)： 1-5
[47]朱世武.基于Copula函数度量违约相关性[J].统计研究,2005,(4)：61-64.
[48]李健伦,方兆本,鲁炜,李红星.Copula方法与相依违约研究[J].运筹与管理,2005,14(3)：111-116.
[49]朱世武.基于Copula的度量与事后检验[J].数理统计与管理,2007(06)：984-991
[50]苏静,杜子平.Copula在商业银行组合信用风险度量中的应用[J].金融理论与实践,2008(05)：6-8
[51]朱世武.Copula函数度量我国商业银行资产组合信用风险的实证研究[J].金融研究,2009(04)：129-141
[52]赵先信.银行内部模型和监管模型——风险计量与资本分配[M].上海：上海人民出版社,2004
[53]杨军.银行信用风险--理论、模型和实证风险[M].北京：中央财政经济出版社,2004