Competing Methods for Efficiency Measurement : A Systematic Review of Direct DEA vs SFA/DFA Comparisons

Various authors have advised a wait and see approach in evaluating the relative precision of alternative techniques, such as data envelopment analysis (DEA) and stochastic frontier analysis (SFA), in estimating industry-average and firm-specific inefficiency. Chirikos and Sear (2000), for example, contend that "policy-makers may be well advised to wait until additional research clarifies reasons why DEA and stochastic frontier models yield divergent results" (p. 1389). The main objective of this paper is to highlight the likely trade-offs between competing methods based on direct empirical comparisons using simulated data and to demonstrate the wealth of evidence bearing on a range of real-world applications. Whilst this systematic review indicates that a good deal of evidence is already available, evidence of a different sort may be required to identify a `correct' approach in addressing specific policy problems. In particular, the now routine practice of cross checking should be taken one step further to include realistic simulation studies along-side real-world DEA vs SFA comparisons.