A quantile regression approach to bank efficiency measurement

Despite the plethora of efficiency studies in the banking literature, there is no consensus on the preferred approach for the empirical estimation of the frontier (production, cost, profit etc.) of fully efficient firms. The present paper aims to provide an overview of this promising alternative approach, that of quantile regression analysis, along with an empirical application in a large international dataset, including 1,520 commercial banks operating in 73 countries, between 2000 and 2006. Apparently, with such a wide coverage our sample is quite heterogeneous both in terms of the countries’ development as well as in terms of the banks’ characteristics. Given the increasing number of cross-country studies, our approach provides an ideal setting for the application of quantile regression that can be particular useful in samples with large bank heterogeneity. The next section discusses the methodological framework of quantile regression. Then we present the empirical results. The concluding marks are discussed in the last section.