EXPLORING INTERRELATIONS AMONG THE DETERMINANTS OF BANK FINANCIAL PERFORMANCE: PANEL EVIDENCE FROM INDIAN SCHEDULED COMMERCIAL BANKS
DOI: https://doi.org/10.5281/zenodo.23278801
Dr. RAMU MUTHE, Dr. CHANDRA REDDY KURRI, Dr. M. NAGALAKSHMI, G. KIRAN
Views: 64 | Downloads: 37
Keywords:
Bank profitability performance, Return on assets (ROA), Non-performing assets of the Banks, Capital adequacy, Panel data, Indian banks.
Abstract:
Over the past twenty years, Indian banks have experienced rapid growth, a bad-loan crisis and a recovery that came sooner than many expected. The researcher wanted to know which ratios really move a bank's return on assets over such a cycle, and how those ratios lean on one another. Design – The study draws on the Reserve Bank of India's Select Ratios of Scheduled Commercial Banks for 53 public and private sector banks, 2005–2025, giving 894 bank-year observations. Once the data were cleaned and winsorised, we compared the two ownership groups, examined the correlation structure and checked for multicollinearity. We then estimated pooled, fixed-effects, random-effects and two-way fixed-effects panel models. Specification tests decided between them, and the final estimates carry Driscoll–Kraay standard errors. Findings – Bad loans outweigh everything else. Each one-point increase in the net NPA ratio takes roughly 0.23 points off ROA, and in standardised terms the effect is more than double that of the next factor. Stronger capital and more fee income lift returns; higher operating costs drag them down. The CASA share, cost of funds and priority-sector lending, by contrast, show no independent effect once bank and year effects are in the model. Interestingly, a point of NPA reduces private banks' ROA far more than it reduces public banks'. Originality – Using a single, consistent regulatory source, the paper follows Indian banks through the whole pre-crisis, asset-quality review and recapitalisation period. It also shows which apparent drivers of profit hold up under proper controls, and which turn out to reflect little more than differences between banks.