Capital structure, bank size and profitability of commercial banks: Dynamic panel evidence from Bangladesh

Abstract

This paper examines how capital structure and bank size impact the profitability of commercial banks in Bangladesh. A balanced panel of 12 commercial banks was studied, covering 2013 to 2022; there were a total of 120 bank-year observations. Return on Equity (ROE) and Net Interest Margin (NIM) were utilized as metrics of bank profitability. The Debt-to-Total-Assets Ratio (DTA), Debt-to-Equity Ratio (DTE), and Long-Term Debt-to-Capitalization Ratio (LTDCR) capture aspects of bank capital structures, while bank size was estimated using the natural log of total bank assets. Initially, fixed-effects and random-effects models, followed by a one-step System GMM, were utilized to address profitability persistence, potential endogeneity, and reverse causality. DTA was positively correlated with both ROE and NIM and was statistically significant at the 1% level in all of the dynamic models. Conversely, DTE had a negative correlation with both ROE and NIM and was also statistically significant at the 5% level in all of the dynamic models. LTDCR was not statistically significant. Bank size was directly correlated with bank profitability, supporting the economies-of-scale argument. These findings suggest that overall asset leverage may increase bank profitability; however, excessive debt compared with equity will decrease performance due to increased financial risk and funding costs. Additionally, this paper enhances the existing literature regarding commercial banking in South Asia by isolating leverage into its component parts and utilizing a dynamic panel approach to estimate those impacts.