Regime-switching model estimates the impact of bank liquidity on bank performance across G20 countries: a moderate role for solvency, total loans, and total debt

This study employs a Panel Smooth Transition Autoregressive (PSTAR) model to investigate the impact of bank liquidity on bank performance, using a sample of 113 banks across 14 G20 countries from 2000 to 2022. The empirical findings reveal a nonlinear relationship characterized by two LDR thresholds at 51.558 and 54.022. In the first regime, bank liquidity exerts an adverse effect on performance, reflecting the costs of excessive idle reserves. In the second regime, the impact of liquidity turns positive, albeit moderate, indicating that banks begin to deploy their liquid resources more efficiently. In the third regime, the positive effect intensifies, with a stronger coefficient, demonstrating that optimal liquidity levels can significantly enhance profitability. Robustness checks using the system GMM approach confirm this nonlinear, inverted-U relationship, with a positive effect of 0.067 and a negative squared term of 0.64e−3, highlighting diminishing marginal returns to liquidity at higher levels. Furthermore, the analysis uncovers significant, positive interaction effects: liquidity combined with solvency strengthens bank performance; liquidity deployed through loans amplifies profitability; and the interaction between liquidity and debt ratios also positively affects performance. These findings indicate that regulators and central banks should adopt flexible liquidity policies that encourage banks to deploy excess funds productively while maintaining adequate buffers, with substantial capital and prudent leverage frameworks enhancing financial stability and sustainable profitability across G20 banking systems.