Measuring technical efficiency

Abstract The theoretical consideration of technical efficiency has existed since Koopmans (Koopmans, T.C., 1951. An analysis of production as an efficient combination of activities. In: Koopmans, T.C. (Ed.), Activity Analysis of Production and Allocation. Cowles Commission for Research in Economics, Monograph No. 13, Wiley, New York) defined it as production possibilities for which it is not possible to increase any output without simultaneously increasing any input, ceteris paribus. The non-parametric approach to efficiency calculation known as Data Envelopment Analysis is not based on this definition but rather on the radial index of Farrell (Farrell, M.J., 1957. The measurement of productive efficiency. Journal of the Royal Statistic Society Ser. A 120, 253–281). Even after Farrell efficiency is achieved there may exist additional slack in individual inputs and/or outputs, implying that Farrell's index does not necessarily measure Koopmans inefficiency. Various measures have been proposed to solve this problem. This paper discusses shortcomings of these alternative measures and introduces a new relative measure of technical efficiency. Using simulated data, it is shown that the new measure outperforms existing measures in the majority of cases.