摘要
近年来,金融科技发展给银行业带来了深刻变革。人们普遍关注金融科技能够在多大程度上为商业银行赋能,又会给市场格局带来哪些变化。文章使用Python网络爬虫技术,构建了商业银行的金融科技运用程度指标,并利用2010-2018年261家国内银行数据,考察了金融科技的运用对不同类型商业银行风险的异质性影响。研究发现,金融科技的运用显著降低了商业银行的风险水平,改善了其风险承受能力,但这种作用对中小银行来说较弱,而且大银行运用金融科技的行为刺激了中小银行风险水平的上升。为了探寻这一现象产生的原因,文章分析了2011-2017年国内银行对小微企业的贷款数据,发现金融科技的运用降低了银企之间的信息不对称程度,缩小了大银行与中小银行在获取软信息方面的能力差距,从而增加了大银行对小微企业的贷款。在此过程中,大银行因其资金成本上的优势,抢占了中小银行的一些优质低风险客户,对中小银行产生了挤出效应。
In recent years, the development of FinTech has brought profound changes to the banking industry. In the field of inclusive finance, the use of the latest IT technologies such as big data, cloud computing,AI and blockchain has given new solutions to the financial constraint problems of SMEs. For this reason, as representatives of traditional financial institutions, commercial banks have also begun to apply financial technology actively and have started strategic transformation. At the same time, people are generally concerned about the extent to which FinTech can empower commercial banks, and what changes will be brought to the market structure.In the process of transformation, large banks represented by state-owned banks and national joint-stock banks often invest hugely and comprehensively, relying on their advantages in capital, technology and talents.However, small and medium banks have comparative advantages in shorter decision-making chains and more flexible responsiveness. Therefore, to which side will the use of FinTech bring more help is a question worth discussing. Using Python tools to crawl web pages, this paper constructs the FinTech application index of commercial banks, and uses data from 261 Chinese banks in 2010-2018 to conduct research. The results show that, large commercial banks have significantly reduced their own risk levels and improved risk tolerance ability after applying FinTech, and the improvement is better than that of small and medium banks.In addition to the impact on individual banks, FinTech will also change the structure of the market.FinTech application has reduced the information asymmetry between banks and enterprises, and narrowed the gap between large banks and small and medium banks in their ability to acquire soft information. In the meanwhile, by virtue of advantages in lower capital cost, large banks are therefore able to identify and more easily attract SMEs with relatively low risks, thus forming a "dimensional reduction" on small and medium banks in the field of inclusive finance, which has brought negative impacts on small and medium banks. The results show that by the application of FinTech, large banks’ loans to SMEs have increased significantly, but the risks have not changed significantly. In contrast, small and medium banks’ loans to SMEs have not increased significantly, but the risks have increased significantly. This phenomenon reflects the loss of high-quality customers in small and medium banks and the existence of market crowding-out effect.The main contributions of this paper are as follows: Firstly, based on the existing literature, we use Python tools to improve the "text mining method", constructing the FinTech application index that is detailed to the individual level of commercial banks;Secondly, for the first time we quantitatively examine the heterogeneous impact of FinTech on traditional banks and the interaction between different banks from the perspective of commercial banks applying FinTech;Thirdly, for the first time we quantitatively examine the crowding-out effect between large banks and small and medium banks by applying FinTech in the field of inclusive finance.This paper has innovations in the index construction and mechanism testing, and is meaningful to the commercial banks that use FinTech for digital transformation, as well as competition and structural changes in banking industry caused by FinTech development.
作者
金洪飞
李弘基
刘音露
Jin Hongfei;Li Hongji;Liu Yinlu(School of Finance,Shanghai University of Finance and Economics,Shanghai 200433,China)
出处
《财经研究》
CSSCI
北大核心
2020年第5期52-65,共14页
Journal of Finance and Economics
基金
国家社会科学基金重点项目“科技型中小企业融资征信特质研究”(15AZD059)。
