DOES GOVERNANCE MATTER FOR CREDIT RISK? DIRECT AND MODERATING EFFECTS OF COUNTRY GOVERNANCE INDEX ON NON-PERFORMING LOANS IN BANGLADESH

Abstract

Non-performing loans (NPLs) continue to pose a significant challenge to the stability of the banking sector in emerging economies, especially in contexts where institutional weaknesses and macroeconomic fluctuations are prevalent. This study delves into how the Country Governance Index (CGI) directly impacts NPLs and how it moderates the relationship between certain macroeconomic factors and NPLs in Bangladesh. Utilizing a quantitative research design, this study analyzes panel data from 34 banks listed on the DSE over the period from 2013 to 2023. A composite CGI is created through Principal Component Analysis (PCA), drawing on six dimensions of the Worldwide Governance Indicators (WGI): Voice and Accountability, Political Stability, Government Effectiveness, Regulatory Quality, Rule of Law, and Control of Corruption. The hypotheses are evaluated using Pooled Ordinary Least Squares (Pooled OLS), Random Effects Model (REM), and Feasible Generalized Least Squares (FGLS) estimations. The results indicate that CGI does not have a significant direct impact on NPLR. However, governance plays a crucial moderating role in the relationship between GDP growth, inflation, and NPLs. Specifically, the interaction terms CGI*GDP and CGI*INF show significant negative effects on NPLs, suggesting that stronger governance amplifies the positive effects of economic growth while reducing the negative impact of inflation on credit risk. On the other hand, the moderating effects of governance on the relationships between unemployment, lending rates, and NPLs are not statistically significant. Overall, the findings imply that governance acts mainly as a stabilizer that reinforces macroeconomic discipline rather than serving as a direct determinant of NPLs.