The collaborative paper titled "A Good Sketch is Better than A Long Speech: Evaluate Delinquency Risk through Real-Time Video Analysis" by Professor Jing Shi(corresponding author), Professor Xiangyu Chang from Xi'an Jiaotong University, Associate Professor Lili Dai from the University of New South Wales, Associate Professor Lingbing Feng from Jiangxi University of Finance and Economics, Dr. Jianlei Han from Macquarie University, and Professor Bohui Zhang from The Chinese University of Hong Kong, Shenzhen, has been published in Review of Finance.

Review of Finance is the official journal of the European Finance Association, primarily publishing high-quality theoretical and empirical academic papers in the field of finance. It is recognized as a top-tier international journal in finance and is classified as an A* journal by the Academic Journal Classification of Zhongnan University of Economics and Law (2023).

Abstact：This paper proposes an innovative method to assess borrowers’ creditworthiness in consumer credit markets by conducting machine-learning-based analyses on real-time video information that records borrowers’ behavior during the loan application process. We find that the extent of borrowers’ micro facial expressions of happiness is negatively associated with loan delinquency likelihood, while the degree of fear expressions is positively associated with delinquency risk. These results are consistent with two economic channels relating to the adequacy and uncertainty of borrowers’ future income, drawn from the extant psychology and economics literature. Our study provides important practical implications for fintech lenders and policymakers.

Link：https://doi.org/10.1093/rof/rfae044

Paper Introduction

The authors address countermeasures for borrowers defaulting on loan repayments across three stages: pre-loan, loan-in-progress, and post-loan. They point out that lending institutions can utilize digital footprints to create personal profiles during the pre-loan stage to assess borrowers' default risks. During the loan-in-progress stage, big data can be leveraged to monitor borrowers' financial status and loan usage. Post-loan, digital information of borrowers can be employed to enhance their willingness and likelihood of repayment. The authors then analyze the differences between ordinary facial expressions and micro-expressions, highlighting the advantages of using micro-expressions. Unlike ordinary facial expressions, micro-expressions are involuntary and inadvertently leaked, thus better reflecting borrowers' true intentions and can be extracted to enrich their profiles. Consequently, the authors propose a hypothesis linking micro-expressions to the likelihood of borrowers defaulting on loan repayments. Drawing on a case study of a leading microfinance company, the authors conclude that happy micro-expressions are significantly negatively correlated with borrowers defaulting on loan repayments, while fearful micro-expressions are significantly positively correlated. This conclusion is grounded in psychological and economic theories and remains valid after undergoing various robustness tests.

Author profile

Jing Shi is currently a Professor of Finance at Zhongnan University of Economics and Law and the Chief Expert of the Innovation and Talent Base for Digital Technology and Finance. He has been selected into the national high-level talent program and holds a PhD in Finance from the Australian National University. Currently, he is a Fellow of the Accounting and Finance Association of Australia and New Zealand, a Council Member of the Asian Finance Association, and the Deputy Editor of the SSCI-indexed journal Accounting and Finance, as well as the Area Editor of Journal of Accounting Literature and the Associate Editor of Research in International Business and Finance. His research areas include financial technology, empirical corporate finance, corporate governance, bond ratings, and corporate business strategies, with a primary focus on the Chinese economy and financial markets. He has published over 70 academic papers in domestic and international journals such as American Economic Review, Journal of Political Economy, Administrative Science Quarterly, Strategic Management Journal, Management Science, Review of Finance, and Economic Research.