网贷借款者违约风险识别——基于借款描述语言特征的视角

当代财经 ›› 2020, Vol. 0 ›› Issue (12): 63-74.

网贷借款者违约风险识别——基于借款描述语言特征的视角

王三兴¹, 岳磊磊¹, 文雁兵², 熊格³

1.安徽大学经济学院,安徽合肥 230601;
2.浙江财经大学浙江经济研究院,浙江杭州 310018;
3.成都量子数能科技有限公司,四川成都 610041

收稿日期:2020-05-13 修回日期:2020-10-30 出版日期:2020-12-15 发布日期:2021-01-07
通讯作者: 岳磊磊,安徽大学博士研究生,主要从事金融市场效率、网络金融研究,联系方式yuellkd@126.com
作者简介:文雁兵,浙江财经大学副教授,博士,主要从事新制度经济学、浙江经济研究;熊格,成都量子数能科技有限公司总经理,硕士,主要从事互联网金融风险管理与控制研究。王三兴,安徽大学教授,博士生导师,博士,主要从事国际金融、金融市场研究
基金资助:
国家社会科学基金项目“纠正人力资本错配促进经济高质量增长研究”(18BJL025); 国家社会科学基金项目“利率放开后金融市场间利率传导阻滞的成因、效应与对策研究”（19BJY241）

Identification of Default Risks of Online Loan Borrowers:From the Perspective of Language Features in Loan Description

WANG San-xing¹, YUE Lei-lei¹, WEN Yan-bing², XIONG Ge³

1. Anhui University, Hefei 230601;
2. Zhejiang University of Finance and Economics, Hangzhou 310018;
3. Chengdu Quantum Data Intelligence Technology Co., LTD., Chengdu 610041, China

Received:2020-05-13 Revised:2020-10-30 Online:2020-12-15 Published:2021-01-07

摘要/Abstract

摘要： 基于借款描述语言特征视角,使用我国A网贷平台公司——年借贷数据,通过研究如何有效识别网贷借款者违约风险,发现借款描述语言难读程度、语言情感积极程度及语言长度与网贷借款者违约风险呈正相关,而借款描述语言具体程度与网贷借款者违约风险呈负相关;借款描述语言难读程度越高、语言情感积极程度越高、语言长度越长及语言具体程度越低的网贷借款者,违约风险越高。网贷平台可以根据借款描述语言特征与网贷借款者违约风险之间的关系,优化风险评估体系、完善风险控制模型、丰富风险控制策略,有效地识别并降低网贷借款者违约风险。

关键词: 网贷, 违约, 风险识别, 借款描述, 语言特征

Abstract: From the perspective of language features in loan description, this paper makes use of the loan data of online loan platform Company A in China from 2010 to 2018 to study how to effectively identify the default risks of online loan borrowers. The results show that such features as the degree of difficulty in reading the loan description language, the degree of positive language emotion and the length of the loan description language are positively correlated with the default risks of online loan borrowers, while the specific degree of the loan description language is negatively correlated with the default risks of online loan borrowers. The more difficult to read the loan description language is, the more positive the language emotion is, the longer the language length is, and the less specific the language is, the higher the default risks will be. According to the relationship between the characteristics of the loan description language and the default risks of the online loan borrowers, the online loan platforms can effectively identify and reduce the default risks of the online loan borrowers by optimizing the risk-assessment system, improving the risk-control model, and enriching the risk-control strategies.

Key words: online loan, default, risk identification, loan description, language characteristics

中图分类号:

F830.49

王三兴, 岳磊磊, 文雁兵, 熊格. 网贷借款者违约风险识别——基于借款描述语言特征的视角[J]. 当代财经, 2020, 0(12): 63-74.

WANG San-xing, YUE Lei-lei, WEN Yan-bing, XIONG Ge. Identification of Default Risks of Online Loan Borrowers:From the Perspective of Language Features in Loan Description[J]. Contemporary Finance & Economics, 2020, 0(12): 63-74.

参考文献

[1] 向虹宇,王正位,江静琳,廖理. 网贷平台的利率究竟代表了什么?[J]. 经济研究,2019,(5):47-62.
[2] 王晶晶,高凯丽. P2P网络贷款对小微企业融资的影响分析[J]. 对外经贸,2019,(11):85-87.
[3] 王立勇,石颖. 互联网金融的风险机理与风险度量研究:以P2P网贷为例[J]. 东南大学学报(哲学社会科学版),2016,(3):103-112.
[4] 李霖魁,张成虎. P2P 网络借贷中的借款人社会资本、风险甄别与市场均衡实现[J]. 当代财经,2017,(10):46-57.
[5] 巴曙松,侯鑫彧,张帅. 基于生存模型的 P2P 平台生存规律与政策模拟研究[J]. 当代财经,2018,(1):44-56.
[6] 廖理,李梦然,王正位. 聪明的投资者:非完全市场化利率与风险识别——来自P2P网络借贷的证据[J]. 经济研究,2014a,(7):125-137.
[7] 廖理,李梦然,王正位. 中国互联网金融的地域歧视研究[J]. 数量经济技术经济研究,2014b,(5):54-70.
[8] 廖理,王会娟. 中国P2P网络借贷平台信用认证机制研究——来自人人贷的经验证据[J]. 中国工业经济,2014,(4):136-147.
[9] 孙小丽,彭龙. KMV模型在中国互联网金融中的信用风险测算研究[J].北京邮电大学学报(社会科学版),2013,(12):75-87.
[10] 刘浩华,陈春花. 借款者婚姻状态与贷款违约——基于网络借贷平台数据的实证研究[J]. 江西财经大学学报,2019,(4):34-42.
[11] 廖理,吉霖,张伟强.语言可信吗?借贷市场上语言的作用——来自P2P平台的证据[J]. 清华大学学报(自然科学版),2015a,(4):413-421.
[12] 廖理,吉霖,张伟强. 借贷市场能准确识别学历的价值吗?来自P2P平台的证据[J]. 金融研究,2015b,(3):146-159.
[13] Pennebaker J.W., King L. A. Linguistic Styles: Language Use as an Individual Difference[J]. Journal of Personality and Social Psychology, 1999, 77(6): 1296-1312.
[14] Tan K.W. P. , Swee D., Lim C., Detenber B. H., Alsagoff L. The Impact of Language Variety and Expertise on Perceptions of Online Political Discussions[J]. Journal of Computer-Mediated Communication, 2007, 13(1): 76-99.
[15] Herzenstein M., Sonenshein S., Dholakia U.M. Tell Me a Good Story and I May Lend You Money: The Role of Narratives in Peer-to-Peer Lending Decisions[J]. Journal of Marketing Research, 2011, 48(SPL): S138-S149.
[16] 李焰,高弋君,李珍妮,才子豪,王冰婷,杨宇轩. 借款人描述性信息对投资人决策的影响:基于P2P网络借贷平台的分析[J]. 经济研究,2014,(S1):143-155.
[17] 王会娟,何琳. 借款描述对P2P网络借贷行为影响的实证研究[J]. 金融经济学研究,2015,(1):77-85.
[18] Larrimore L., Jiang L., Larrimore J., Markowitz D., Gorski S.Peer to Peer Lending: The Relationship between Language Features, Trustworthiness, and Persuasion Success[J]. Journal of Applied Communication Research, 2011, 39(1):19-37.
[19] Gao Q., Lin M.Linguistic Features and Peer-to-Peer Loan Quality: A Machine Learning Approach[J/OL]. Social Science Electronic Publishing, 2014.
[20] Flanagin A.J. Commercial Markets as Communication Markets: Uncertainty Reduction through Mediated Information Exchange in Online Auctions[J]. New Media & Society, 2007, 9(3): 401-423.
[21] Berger C.R., Calabrese R. J. Some Exploration in Initial Interaction and Beyond: Toward a Developmental Theory of Communication[J]. Human Communication Research, 1975, 1(2): 99-112.
[22] Schwanenflugel P.J., Stowe R. W. Context Availability and the Processing of Abstract and Concrete Words in Sentences[J]. Reading Research Quarterly, 1989, 24(1): 114-126.
[23] Seifert L.S. Activating Representations in Permanent Memory: Different Benefits for Pictures and Words[J]. Journal of Experimental Psychology. Learning, Memory & Cognition, 1997, 23(5): 1106-1121.
[24] Paivio A.Mental Representations: A Dual Coding Approach[M]. Oxford: Oxford University Press, 1988.
[25] Paivio A.Dual Coding Theory: Retrospect and Current Status[J]. Canadian Journal of Psychology, 1991, 45(3):255-287.
[26] 李绍山. 易读性研究概述[J]. 解放军外国语学院学报,2000,(4):1-5.
[27] Betts E.A. Readability: Its Application to the Elementary School[J]. The Journal of Educational Research, 1949, 42(6): 438-459.
[28] Li F. Annual Report Readability, Current Earnings,Earnings Persistence[J]. Journal of Accounting & Economics, 2008, 45(2-3): 221-247.
[29] Rennekamp K.M. Processing Fluency and Investors’Reactions to Disclosure Readability[J]. Journal of Accounting Research, 2012, 50(5): 1319-1354.
[30] 秦雪征,庄晨,杨汝岱. 计划生育对子女教育水平的影响——来自中国的微观证据[J]. 经济学(季刊),2018,(4):897-922.
[31] Gunning R.The Technique of Clear Writing[M]. New York: McGraw Hill, 1952: 36-37.