首页|大数据对银行信贷行为的影响——来自数字社会信用平台的证据

大数据对银行信贷行为的影响——来自数字社会信用平台的证据

扫码查看
随着中国数字经济的深入推进,城市数字社会信用平台集成了多维度的大数据,其经济效应卓有成效.本文将数据要素纳入银企信息不对称的分析框架中,基于银行逐笔信贷数据,利用城市数字社会信用平台建立的准实验,实证考察了大数据对银行信贷行为的影响.研究发现,数字社会信用平台建设显著降低了信贷违约风险.机制分析证实,数字信用平台建设降低了借款企业的逆向选择和道德风险,提高了银行的违约预测能力,通过改变银企匹配结构,降低了信贷违约风险.进一步的研究发现,数字社会信用平台对税款逾期的借款企业实施了有效的联合惩戒,降低了信用贷款的概率,提高了担保物要求.此外,数字社会信用平台对贷款金额和利率的影响与借款企业规模及银行的数据处理能力有关:企业规模越小,越难从信用平台中获益;银行的数据处理能力越强,越容易借助信用平台差异化利率定价.本文研究表明,大数据的运用能显著加强银行对违约风险的识别和监测;不同的企业规模、差异化的数据分析能力会导致贷款企业福利损失.上述发现为更好地发挥数据要素在信贷市场中的作用提供了事实依据.
Effect of Big Data on Bank Lending:Evidence from Digital Social Credit Platform
The impact of digitalization on economic growth,with data as its core production factor,has become increas-ingly evident.In recent years,cities in China have progressively undertaken digital transformation,building upon the original social credit system.This transformation involves the establishment of database systems and service platforms,facilitating the sharing of administrative big data across government departments,and building a nationwide digital social credit system.As a country mainly financed by bank credit,China needs to focus on the quality of bank credit and the risk of default to prevent and resolve financial risks.This paper incorporates data elements into the analytical framework of information asymmetry between banks and firms and examines how improved data availability affects banks'lending behavior.This paper divides the information asymmetry into two levels.The first level is the asymmetry in the availability of the data themselves.This is reflected in the fact that the digital social credit system increases the availability of data to banks,helping them to identify credit de-fault risks and establish a trustworthy environment.The second level is the asymmetry at the level of the ability to ana-lyze the data.On the one hand,if the gap in data processing capabilities between banks is too large,banks with greater ca-pabilities will enjoy greater monopoly power,which may hinder competition in the credit market and worsen the welfare of all borrowers in the credit market.On the other hand,different allocation of data elements may affect the structure of the credit market with significant welfare effects.Using a total of 5,370,263 corporate loan tracking records as a sample and the establishment of a digital social credit platform as a quasi-natural experiment,this paper applies the difference-in-differences(DID)model to test the effect of improving bank data availability on banks'monitoring of credit default risk and the mechanisms behind it.This paper finds that:(1)The improvement of bank data availability leads to a significant decrease of 1.60%in the loan default rate,and the effect is more significant in borrowers with rich information footprints and banks with stronger data processing capabilities;(2)Mechanism test shows that the digitalized social credit platform significantly mitigates the adverse selec-tion and moral hazards between banks and borrowers,enhances the information screening capabilities of banks,and change the matching structure of banks and borrowers,thus effectively reducing the risk of credit default;(3)Further analysis shows that by integrating the information of the tax authorities through the digital social credit platform,banks carry out effective punishment for borrowers with overdue taxes,reduce the probability of granting credit loans to them,and increase the requirement of collaterals;(4)The impact of the digital social credit platform on the amount of loans and interest rates varies with the size of the borrowers and the data processing capacity of the lending bank.The smaller the size of the borrowers,the more difficult for them to benefit from the credit platform;and the greater the data processing capacity of banks,the higher the interest rate after the establishment of the digital credit platform.This paper has the following contributions.Firstly,it examines the impact of the digital social credit platform on bank credit behavior,providing empirical evidence supporting multi-departmental collaborative supervision.Secondly,in the era of big data,discussing data accessibility and data processing capabilities becomes equally important,as lacking ei-ther capability may lead to new social inequalities.Unlike previous research frameworks that do not directly distinguish whether information asymmetry stems from insufficient data accessibility or inadequate data processing capabilities,this paper categorizes information asymmetry between banks and enterprises into data accessibility and data processing capa-bilities,addressing the heterogeneous effects of improved bank data accessibility and differences in bank data processing capabilities on borrower welfare.Thirdly,it discusses for the first time the impact of digital social credit platform con-struction based on government big data on information asymmetry between banks and enterprises in credit activities.This enriches research on methods for reducing information asymmetry between banks and enterprises and holds practical sig-nificance for the high-quality development of both digital and real economies.

Digital EconomyData ElementsSocial Credit PlatformsBank Credit

佘楷文、申宇、赵绍阳

展开 >

西南财经大学金融学院,邮政编码:611130

四川大学经济学院,邮政编码:610065

数字经济 数据要素 社会信用平台 银行信贷

国家自然科学基金面上项目

71872150

2024

经济研究
中国社会科学院经济研究所

经济研究

CSTPCDCSSCICHSSCD北大核心
影响因子:4.821
ISSN:0577-9154
年,卷(期):2024.59(3)
  • 57