P2P网络借贷平台风险防控机制和预警体系——基于信用中介化与流动性风险视角
|
摘要
基于信用中介化与流动性风险恶性循环机制的理论分析,对网络借贷出借行为结构模型进行优化,采用普通最小二乘法(OLS)模型、逻辑回归(Logit)模型、Probit模型实证检验借款信息变量的借贷策略效应,通过大数据决策树算法对所选取的变量进行预警过程的仿真模拟,结果表明:P2P网络借贷平台的信用中介化融资风险市场系统化效应显著,第三方大数据风险监测与预警有利于规制P2P网络借贷平台信用中介化下的主体融资风险,P2P网络借贷平台应该定位为信息中介。为了回归这一定位,增强平台性质"变异"过程中第三方风控预警能力,必须落实平台信息中介政策,严惩信用中介化网络借贷平台;实行行业数据共享,完善征信体系建设,构建信贷市场有序化发展的良性契约基础;加强金融风险教育,建立政府—平台—主体"三位一体"的风险规制体系,完善市场准入机制和平台动态监管预警机制,建立良好、有序的网络信贷市场。
The authors mainly discuss the risk prevention and control mechanism and early-warning system of P2P network lending platform from the perspective of credit intermediation and liquidity risk. Firstly,the authors theoretically analyze the optimization of the structural model of the network lending behavior when considering the factors of credit intermediation and liquidity risk deteriorating circle mechanism. Secondly,they empirically analyze the effect of the borrowing strategy caused by the borrowing information variables with the OLS model,Logit model and Probit model,and simulate the risk alerting process through the variables selected by the big data decision tree algorithm. The empirical results show a significant relationship between the systemic effect of the financing risk market and the credit intermediary of the P2P network lending platform. The third-party big data risk monitoring and early warning is conducive to regulating the transmission of the main financial risk from the P2P network lending platform of credit intermediary services. Finally,the authors put out some suggestions to enhance the third party's capability for risk control and early warning,we should implement the platform's policy of positioning the information intermediary business,severely punish the online lending platform of credit intermediary services,promote the share of industrial data,perfect the construction of credit reference system,lay the solid foundation for the orderly development of credit market,strengthen the education of financial risk,perfect the mechanism of market access and dynamic supervision and early-warning of platforms,and establish the sound and orderly online credit market.
The authors mainly discuss the risk prevention and control mechanism and early-warning system of P2P network lending platform from the perspective of credit intermediation and liquidity risk. Firstly,the authors theoretically analyze the optimization of the structural model of the network lending behavior when considering the factors of credit intermediation and liquidity risk deteriorating circle mechanism. Secondly,they empirically analyze the effect of the borrowing strategy caused by the borrowing information variables with the OLS model,Logit model and Probit model,and simulate the risk alerting process through the variables selected by the big data decision tree algorithm. The empirical results show a significant relationship between the systemic effect of the financing risk market and the credit intermediary of the P2P network lending platform. The third-party big data risk monitoring and early warning is conducive to regulating the transmission of the main financial risk from the P2P network lending platform of credit intermediary services. Finally,the authors put out some suggestions to enhance the third party's capability for risk control and early warning,we should implement the platform's policy of positioning the information intermediary business,severely punish the online lending platform of credit intermediary services,promote the share of industrial data,perfect the construction of credit reference system,lay the solid foundation for the orderly development of credit market,strengthen the education of financial risk,perfect the mechanism of market access and dynamic supervision and early-warning of platforms,and establish the sound and orderly online credit market.
引文
[1]赵鹞,殷实.P2P网络平台的异化[J].中国金融,2015(15):90-92.
[2]卢馨,李慧敏.P2P网络借贷的运行模式与风险管控[J].改革,2015(2):60-68.
[3]郑扬扬.我国P2P平台的角色定位与治理研究[D].杭州:浙江大学,2016.
[4]冯果,蒋莎莎.论我国P2P网络贷款平台的异化及其监管[J].法商研究,2013,30(5):29-37.
[5]王阳阳,赵峰.网络借贷的流动性风险管理分析[J].金融经济:下半月,2018(3):75-77.
[6]GHATAK M.Group lending,local information and peer selection1[J].Journal of development economics,1999,60(1):27-50.
[7]互联网金融协会.P2P网络借贷平台相关法律法规及案例[M].北京:中国金融出版社,2016:12.
[8]陈冬宇,朱浩,郑海超.风险,信任和出借意愿--基于拍拍贷注册用户的实证研究[J].管理评论,2014,26(1):150.
[9]宋逢明,谭慧.VaR模型中流动性风险的度量[J].数量经济技术经济研究,2004(6):114-123.
[10]张超宇,陈飞.P2P网络借贷平台模式异化及去担保化问题研究[J].南方金融,2018(1):68-74.
[11]李悦雷,郭阳,张维.中国P2P小额贷款市场借贷成功率影响因素分析[J].金融研究,2013(7):126-138.
[12]张玮,杨善林,刘婷婷.基于CART和自适应Boosting算法的移动通信企业客户流失预测模型[J].中国管理科学,2014,22(10):90-96.
[13]雷阳,黄卓.征信体系是P2P回归信息中介的关键[J].征信,2016(4):24-28.
(1)本文将网络借贷平台脱离信息中介本质的过程称为“信用中介化”。
(2)大数据风控可以理解为金融科技,所以本文提出的第二道风控属于传统意义上的金融科技的第二道风控。提出第一道风控和第二道风控概念的原因在于本文关于网络借贷中风险控制的脉络基于这两个层次,即风险防控机制和预警体系。
(3)信用中介化具体类型的定义详见《P2P网络借贷平台相关法律法规及案例》。
(4)拍拍贷互联网金融数据训练营网址:https://www.kesci.com/apps/home/dataset/58c614aab84b2c48165a262d/files。
(5)拍拍贷网址:https://help.ppdai.com/Home/List/7。
(6)C4.5算法步骤和CART算法简述详见http://blog.sina.com.cn/s/blog_7399ad1f01014oic.html。
(7)详见人民网,http://money.people.com.cn/n1/2018/1111/c42877-30393475.html。
(8)详见人民网,http://politics.people.com.cn/n1/2017/0714/c1001-29404027.html。
[2]卢馨,李慧敏.P2P网络借贷的运行模式与风险管控[J].改革,2015(2):60-68.
[3]郑扬扬.我国P2P平台的角色定位与治理研究[D].杭州:浙江大学,2016.
[4]冯果,蒋莎莎.论我国P2P网络贷款平台的异化及其监管[J].法商研究,2013,30(5):29-37.
[5]王阳阳,赵峰.网络借贷的流动性风险管理分析[J].金融经济:下半月,2018(3):75-77.
[6]GHATAK M.Group lending,local information and peer selection1[J].Journal of development economics,1999,60(1):27-50.
[7]互联网金融协会.P2P网络借贷平台相关法律法规及案例[M].北京:中国金融出版社,2016:12.
[8]陈冬宇,朱浩,郑海超.风险,信任和出借意愿--基于拍拍贷注册用户的实证研究[J].管理评论,2014,26(1):150.
[9]宋逢明,谭慧.VaR模型中流动性风险的度量[J].数量经济技术经济研究,2004(6):114-123.
[10]张超宇,陈飞.P2P网络借贷平台模式异化及去担保化问题研究[J].南方金融,2018(1):68-74.
[11]李悦雷,郭阳,张维.中国P2P小额贷款市场借贷成功率影响因素分析[J].金融研究,2013(7):126-138.
[12]张玮,杨善林,刘婷婷.基于CART和自适应Boosting算法的移动通信企业客户流失预测模型[J].中国管理科学,2014,22(10):90-96.
[13]雷阳,黄卓.征信体系是P2P回归信息中介的关键[J].征信,2016(4):24-28.
(1)本文将网络借贷平台脱离信息中介本质的过程称为“信用中介化”。
(2)大数据风控可以理解为金融科技,所以本文提出的第二道风控属于传统意义上的金融科技的第二道风控。提出第一道风控和第二道风控概念的原因在于本文关于网络借贷中风险控制的脉络基于这两个层次,即风险防控机制和预警体系。
(3)信用中介化具体类型的定义详见《P2P网络借贷平台相关法律法规及案例》。
(4)拍拍贷互联网金融数据训练营网址:https://www.kesci.com/apps/home/dataset/58c614aab84b2c48165a262d/files。
(5)拍拍贷网址:https://help.ppdai.com/Home/List/7。
(6)C4.5算法步骤和CART算法简述详见http://blog.sina.com.cn/s/blog_7399ad1f01014oic.html。
(7)详见人民网,http://money.people.com.cn/n1/2018/1111/c42877-30393475.html。
(8)详见人民网,http://politics.people.com.cn/n1/2017/0714/c1001-29404027.html。