互联网金融、存款竞争与银行风险承担

本文通过构建纳入互联网金融的银行环形城市模型,推演了“互联网金融→存款结构/付息成本→银行风险承担”的传导机制。在此基础上,以2003-2016年我国83家商业银行为样本,建立多重中介效应模型进行实证检验。研究结果表明:(1)互联网金融发展经由恶化存款结构和抬高付息成本两种渠道显著加重了银行风险承担水平,其中,恶化存款结构效应的相对贡献为50%左右,抬高付息成本效应的相对贡献为35%左右;(2)相较于互联网渠道构筑业态,互联网支付结算、互联网资源配置和互联网财富管理业态对银行存款结构和付息成本的不利影响更为强烈;(3)相对于国有、大规模、低流动性和低资本充足率商业银行,面对互联网金融的冲击,非国有、小规模、高流动性和高资本充足率商业银行的客户存款流失更快,平均付息成本上涨更多。

Internet Finance,Deposit Competition,and Bank Risk-taking

Amid advances in information technology and financial innovation, internet finance practices are changing continuously, and new internet finance models are constantly emerging. Since 2013, the development of internet finance has created significant prosperity, leading to rapid changes in how people conduct financial activities and fundamentally altering the financial ecosystem. Commercial banks, the mainstay of the traditional financial sector, are facing increasingly fierce deposit competition. The 19^th National Congress of the Communist Party of China in October 2017 proposed “perfecting the financial supervision system and holding the bottom line of systemic financial risks.” The March 2018 meeting of the Central Committee of Finance and Economics emphasized that “preventing and defusing financial risks is related to national security, overall development, and people's property security.” These statements, coming against the backdrop of the rise of internet finance, identified the regulation of bank risk-taking as a central requirement. This raises several important questions: how does internet finance affect bank risk-taking through the channel of intensifying deposit competition？ Do the various models of internet finance have heterogeneous impacts on commercial banks？ Do different types of commercial banks have diverse responses to internet finance？ These are important practical and theoretical issues that need to be solved to prevent bank risks and deepen financial system reform.
By constructing a bank circular city model that incorporates internet finance, this paper deduces the following transmission mechanism: internet finance → deposit structure/interest-paying costs → bank risk-taking. Then, based on an internet finance index (constructed via text mining) as the core explanatory variable and using a sample of 83 commercial banks in China from 2003 to 2016, a multiple mediation model is built for empirical testing. Furthermore, this paper uses system generalized moment estimation (SYSGMM), iterative seemingly uncorrelated estimation (SUR-i), and instrument variable (IV) technology to verify the existence and dual heterogeneity of the above effects.
The main conclusions of this paper are as follows: 1) The development of internet finance has significantly increased the risk-taking level of banks through deposit structure deterioration and interest-paying cost increases; the relative contribution of deposit structure deterioration is about 50%, and that of interest-paying cost increases is about 35%. 2) Relative to internet channel construction, the paper finds that internet payment settlement, internet resource allocation, and internet wealth management have stronger negative impacts on banks' deposit structure and interest-paying costs. 3) Compared with state-owned, large-scale, low-liquidity, and low-capital-adequacy commercial banks, non-state-owned, small-scale, high-liquidity, and high-capital-adequacy commercial banks suffer more rapid losses of customer deposits and larger increases in average interest-paying costs when confronted with challenges from internet finance.
This paper makes several contributions to the literature. First, based on the unique economic and financial environment of China, this paper constructs a bank circular city model that incorporates internet finance and deduces the internal transmission mechanism of internet finance → deposit structure/interest-paying costs → bank risk-taking. Second, by establishing a multiple mediation model, this paper for the first time verifies the mechanism of internet finance development's impact on bank risk-taking; it also evaluates the relative contribution of two aspects of this impact, deposit structure deterioration and interest-paying cost increases. Lastly, while earlier studies pay more attention to the overall impact of internet finance on commercial banks, this paper divides internet finance into four typical models according to financial function to explain the heterogeneous impacts of various models on commercial banks. In addition, using four dimensions of banks (property-rights type, asset scale, liquidity level, and capital strength), this paper examines the diverse responses of different types of commercial banks to internet finance.