The Influence of Digital Transformation on Risk-Taking in Commercial Banks in Indonesia Using Text Mining

Abstract

This research examines the impact of digital transformation on banking risks, specifically focusing on credit risk (NPL), liquidity risk (LDR), and insolvency risk (Z-score). Employing a quantitative method, the study constructs a digital transformation index using text-mining techniques applied to annual reports of Indonesian commercial banks. The analysis utilizes Ordinary Least Squares (OLS), Fixed Effects (FE), and the System Generalized Method of Moments (SYS-GMM) on a dataset comprising 59 commercial banks in Indonesia from 2018 to 2024. The results reveal that digital transformation significantly raises credit risk. In contrast, its effects on liquidity and insolvency risks are statistically insignificant, suggesting potential improvements in credit evaluation through the use of enhanced data and technological tools. Additionally, the study demonstrates the utility of the SYS-GMM model in addressing endogeneity concerns in dynamic panel data. These findings can help regulators understand the strategic role of digital implementation and innovation in enhancing risk management and financial stability within the commercial banking sector.