Determinants of bank efficiency: evidence from a semi-parametric methodology

Purpose -This paper aims to analyze bank efficiency into a number of bank-specific, industryspecific and macroeconomic determinants. Design/methodology/approach - The authors follow a semi-parametric two-stage methodology, where productive efficiency is derived via a non-parametric technique in the first stage and then the scores obtained are linked to a series of determinants of bank efficiency, using a double bootstrapping procedure. Findings - Overall, it is found that the banking sectors of almost all the sample countries show a gradual improvement in their efficiency levels. The model used shows that a number of determinants like bank size, industry concentration and the investment environment have a positive impact on bank efficiency, which is not the case when standard Tobit models are employed. Research limitations/implications - The findings have important implications for the relevance of well-known hypotheses that refer to the performance of the banking sectors, like the structureconduct- performance and the efficient structure hypotheses. These implications are not necessarily verified when past conventional econometric methodologies are used. Practical implications - The paper offers new insights to policy makers, bank managers and practitioners on the relevance of a number of driving factors of bank efficiency that might help them to improve the performance of the banking system and enhance the quality of services provided. Originality/value - This is the first paper in the bank efficiency literature that employs a semiparametric two-stage model, which relaxes several deficiencies of previous two-stage empirical approaches thus, offering a solution to the many problematic features of standard censored regressions.