Measuring economic efficiency with incomplete price information: With an application to European commercial banks

Measuring economic efficiency requires complete price information, while resorting to technical efficiency exclusively does not utilise any price information. In most studies, at least some information on the prices is available from theory or practical knowledge of the industry under evaluation. In this paper we derive upper and lower bounds for Farrell's overall economic efficiency, assuming incomplete price information. The bounds typically give a better approximation for economic efficiency than technical efficiency measures that use no price data whatsoever. Standard linear programming techniques can obtain nonparametric estimates for these bounds from empirical data. From operational point of view, the approach developed in this paper both extends the scope of the weight restricted Data Envelopment Analysis (DEA) models to more general technologies, and explicitly links these techniques to the classic theoretical framework of efficiency measurement. The approach is illustrated with an empirical application to large European Union commercial banks.