1. Introduction

Koopmans (1951) provided a useful framework for the measurement of efficiency by defining technical efficiency as feasible input–output combinations where it is not possible to increase output (decrease inputs) without simultaneously increasing inputs (decreasing output). In a seminal paper, Farrell (1957) showed how efficiency can be measured relative to a given isoquant as the maximum radial reduction in observed inputs holding output constant. Farrell further provided the decomposition of overall efficiency into technical and allocative parts. Farrell’s paper serves as the foundation for the nonparametric and parametric models of technical efficiency estimation. While Farrell provided a useful foundation for the measurement of technical efficiency, the model allowed only one output and assumed constant returns to scale. The assumption was relaxed to allow increasing returns to scale in Farrell and Fieldhouse (1962).