This paper investigates the extent to which the major bank-specific factors, i.e., loan growth (LG), non-performing loans (NPL), liquidity (LQ), and capital growth (CAG), explain profitability among Cambodian commercial banks using return on assets (ROA). A paned ARDL model is used with the Pooled Mean Group (PMG) and Dynamic Fixed Effects (DFE) estimations to capture long-run equilibrium relationships as well as short-run dynamics among these variables. The substantially negative error-correction terms (ECTs) for the models indicate a stable cointegrating relationship between ROA and the independent variables. The empirical evidence suggests that loan growth is potentially the most significant driver of long-run profitability. The positive and highly significant LG coefficients of both PMG and DFE suggest that those banks that conduct economies of strong credit expansion would ultimately lead to a higher ROA. Since the short-run impact of LG is negative, this could be due to the up-front costs of credit screening, monitoring, and provisioning required before new loans can start earning income. NPL, as a measure of credit risk, has a significant negative effect on profitability in the short run, indicating that increasing bad assets deteriorates interest income and leads to more provisions for loan loss. Although the long-run impact of NPLs is less pronounced among estimators, PMG outcomes suggest there is a weak guiding-down effect. Liquidity has a highly negative long-run effect in the PMG, indicating that holding low-yield liquid assets leads to lower returns, but no short-run effect. Capital appreciation is generally good for profits, but surprisingly the DFE model does not show this, and PMG only finds considerably weaker gains in the long run.
Siphat Lim (2026) studied this question.