Topics in banks’ profitability, risk and efficiency

Investigating the factors which determine bank profitability has been a hot topic for decades, however, there is still no consensus among researchers, practitioners, or policymakers regarding the set of factors that affect bank profitability due to the differences in social, economic, legal, and technical contexts across banking systems all over the world which give rise to the existence of many econometric specifications and samples design. In this paper, we survey 65 studies on bank profitability which covers the banking systems in three major blocs; the US, the EU, and the rest of the world (RoW). We followed two approaches; the first is to classify the surveyed studies into subgroups based on the used measure of profitability (ROA or ROE), sample size (small or large), and the time frame (the period) that data covers in each study (short or long). We then trace the impact of each of the factors of profitability (defined earlier) on the measure of profitability within the aforementioned subgroups across the different geographical areas. We call this approach “between blocks”. The second approach was to trace the impact of each factor on ROA or ROE within each geographical area; namely the US, EU, and RoW. We call this approach the “within blocks”. Factors are said to have an impact on ROA or ROE based on a set of rules we design within the paper. Based on “between blocks”, we find three factors with consistent impact on ROA in all subgroups (CAP, CE, and GGDP) which are all positively related to ROA; One factor (CR) was found to have a negative impact on ROA in three subgroups. Comparing the impact of each factor on ROE across the three subgroups, we find only one factor (CE) that is positively related to ROE in three subgroups. Based on “within blocks”, and within the US, factors that are found to be positively related to ROA are CAP, CE, GGDP, and INF, while CR is negatively related to ROA. On the other hand, factors that are found to be positively related to ROE are CE and NII, while CR and LIQ are negatively related to ROE. In the EU, factors that are found to be positively related to ROA are CAP, CE, and GGDP, while CR, LIQ, BSD, and INF are found to be negatively related to ROA. On the other hand, factors that are found to be positively related to ROE are CE and GGDP, while CR is the only factor with a negative impact on ROE. Finally, factors that are found to have a positive impact on ROA within RoW are CAP and CE, while CR and LIQ are negatively related to ROA. When it comes to ROE, CE and GGDP have a positive impact while LIQ is negatively related to ROE. To check for the robustness of our findings, we decided to account for the “quality” of each of the surveyed papers to assure that they are equally weighted in terms of each paper’s quality. Hence, we have to consider only the papers which come from top journals, or in some cases, bad journals but the papers are of high quality. To perform this task, we build a “quality index” which rules out all the papers that are of scholarly low quality by defining some rules. As a result of applying this index, our surveyed studies were downsized from 65 to only 45 studies. Our findings confirm the results obtained before the index was applied which confirms that the inclusion of such index represents a special case of the approaches we developed in tracing the impacts of determinants of bank profitability internationally and, therefore, our results are robust.