Modeling the efficiency of Tunisian and Moroccan banks using the SFA approach

Purpose - The purpose of this paper is to evaluate and compare bank efficiency between the two Maghreb countries, Tunisia and Morocco, over the period 2005–2014. Design/methodology/approach - The authors follow the stochastic frontier analysis, where the preferred cost model is determined via various hypothesis tests based on the maximum likelihood estimation. Then, the first and the second derivates of the cost function are employed to determine scale elasticities, scale inefficiencies and technological progress. Findings - Specification tests indicate that the Fourier Flexible form provides better fit to the data set. Further, the estimated model shows that Tunisian and Moroccan banks’ efficiency is positively affected by banking service quality, but negatively influenced by both bank capitalization and GDP growth. Overall, Moroccan banks are found to be the most efficient despite the decrease of efficiency levels in both countries. Additionally, foreign banks have a higher scale inefficiency and, therefore, a lower cost efficiency. Equally, the technical progress raises banking costs in both countries, providing a decrease in efficiency scores. Practical implications - The findings of this study provide novel insights to Tunisian and Moroccan policy makers on the relevance of the smaller banks’ consolidation to improve bank efficiency by achieving unrealized economies of scale. Also, more reforms should be implemented in Tunisia to reduce non-performing loans. Originality/value - To the best of the authors’ knowledge, this study is the first which offers a comparison between Tunisian and Moroccan banks to clarify the sources of inefficiency and to make strategic decisions.