How Green Credit Policy Affects Commercial Banks' Credit Risk?: Evidence and Federated Learning-Based Modeling From 26 Listed Commercial Banks in China

Tongyue Feng, Jiexiang Xu, Zehan Zhou, Yilang Luo

Source Title: Journal of Cases on Information Technology (JCIT)26(1)

ISSN: 1548-7717|EISSN: 1548-7725|EISBN13: 9798369323878|DOI: 10.4018/JCIT.333858

MLA

Feng, Tongyue, et al. "How Green Credit Policy Affects Commercial Banks' Credit Risk?: Evidence and Federated Learning-Based Modeling From 26 Listed Commercial Banks in China." JCIT vol.26, no.1 2024: pp.1-21. http://doi.org/10.4018/JCIT.333858

APA

Feng, T., Xu, J., Zhou, Z., & Luo, Y. (2024). How Green Credit Policy Affects Commercial Banks' Credit Risk?: Evidence and Federated Learning-Based Modeling From 26 Listed Commercial Banks in China. Journal of Cases on Information Technology (JCIT), 26(1), 1-21. http://doi.org/10.4018/JCIT.333858

Chicago

Feng, Tongyue, et al. "How Green Credit Policy Affects Commercial Banks' Credit Risk?: Evidence and Federated Learning-Based Modeling From 26 Listed Commercial Banks in China," Journal of Cases on Information Technology (JCIT) 26, no.1: 1-21. http://doi.org/10.4018/JCIT.333858

Export Reference

Favorite Full-Issue Download

View Full Text HTML

View Full Text PDF

Abstract

The green credit policy has significantly influenced the growth of green industries in China. This study evaluates its impact on reducing bank credit risk using data from 26 Chinese banks from 2015 to 2021. The authors discovered that the policy's primary effect is linked to banks' financial leverage. Notably, green credit's influence on insolvency risk is most evident in leverage risk. However, despite governmental support for green credit collaboration, prevalent information gaps between banks and green enterprises lead to misjudgments and subsequent credit losses. To address the balance between credit risk mitigation and privacy, the authors investigated vertical joint learning for a precise risk control model grounded in commercial banks' practices. Experiments revealed that this joint model outperforms the sole “bank internal model” in presenting green credit data, underscoring the potential of machine learning to refine green credit systems and bolster banks' credit risk management.