This study aims to examine the relationship between cost efficiency and risk in Islamic and conventional banks of the Gulf Cooperative Council countries over the period 2006-2015. The sample is composed of 99 commercial banks divided into 51 conventional banks and 48 Islamic banks. To achieve this, we have used stochastic frontier in the first stage and meta-frontier analysis in the second stage to calculate efficiency scores. The GMM model is used to examine the relationship between cost efficiency and credit risk.
The specific as well as common frontiers obtained by the stochastic frontier method and meta-frontier analysis show the superiority of conventional banks compared to Islamic banks in terms of cost efficiency. Moreover, descriptive statistics show that Islamic banks are more liquid, more exposed to credit risk and more capitalized than conventional banks. The regression estimation further indicated a positive and significant effect of the credit risk on cost efficiency obtained by the stochastic frontier of Islamic banks, and a negative one in the case of conventional banks. The common frontier reveals no significant effect. However, this study proves that credit risk has a positive and significant relationship with the cost efficiency of Islamic and conventional banks obtained by the meta-frontier analysis. This study shows that the choice of the evaluation method of the efficiency scores can influence the results obtained. As for the impact of the regulatory capital ratio, it is found to be positively correlated with the cost efficiency of both categories of banks.