Shocks and stability: structural break analysis of non-performing assets and profitability in Indian banking

Abstract

The present paper aims to investigate the relationship between non-performing assets (NPAs) and profitability in Indian commercial banks. Focusing on the major economic disruptions, namely, the global financial crisis, demonetization and the COVID-19 pandemic, it examines their role in structurally reshaping the NPA-profitability nexus. Our dataset comprised a balanced panel of 30 public and private sector commercial banks in India covering the period from 2004 to 2024. We applied Bai–Perron multiple breakpoint tests and Chow tests to identify regime shifts. Static panel regressions with Heteroskedasticity and Autocorrelation Consistent-corrected ordinary least squares and dynamic system generalized method of moments (GMM) estimations are employed to address heteroskedasticity, autocorrelation and endogeneity. The model incorporates bank-specific factors, market concentration indicators, macroeconomic variables and an interaction term between inflation and broad money (BM). The findings reveal multiple statistically significant structural breaks corresponding to major economic shocks. NPAs consistently exert a significant negative impact on profitability across all regimes, while bank size positively influences return on assets, particularly during crisis periods. Market concentration yields mixed and regime-dependent effects. The interaction between inflation and BM significantly moderates profitability during transitional phases, highlighting the role of inflation-adjusted liquidity conditions in shaping bank performance. The findings suggest that bank managers should strengthen credit risk monitoring and adopt proactive asset quality management, especially during periods of inflationary liquidity expansion. The policymakers and regulators can benefit from implementing countercyclical capital buffers and dynamic stress-testing frameworks tailored to structural shocks. However, our results should be interpreted with caution as we excluded foreign banks from our dataset and focused solely on Indian commercial banks, which may limit cross-country generalizability. Future research could extend the framework to comparative emerging-market settings. To the best of our knowledge, this study is the first attempt to integrate structural break analysis with dynamic GMM estimation and to introduce inflation-adjusted liquidity as a moderating mechanism in the NPA-profitability relationship in an emerging economy context.