Measuring efficiency has become the core objective of a structured research agenda in production economics and management sciences since at least the early 1980s. The methods used, however, might have a significant impact on results and need to be adapted to the data at hand. Using a synthetic general model to compare the various approaches and relying on key contributions of the literature, we show that each class of model implies specific assumptions on the nature of the data, and that in some cases, the models are inconsistent. Most studies meet the basic requirements proposed by Cowing and Stevenson in 1983, as they rely on the solid theoretical foundations of production economics. Yet, many methods were nevertheless developed in the fields of statistics, operations research, or accounting. These methods often fail to include all relevant theoretical considerations. For example, authors relying on economic theory have applied empirical methods with stochastic error terms that are sometimes at odds with certain properties of their models.
Efficiency Measurement: A Methodological Review and Synthesis Available to Purchase
N.B.: Members of the Productivity Analysis Research Network (PARN) provided highly valuable input at the start of this project, following a call to all. In particular, V. Valdmanis, M.D. Rosko, and B. Hollingsworth forwarded many of their own contributions and we would like to thank them for their help. R. Banker and two referees also provided very useful comments and helped better define the purpose of this paper. We would also like to thank the Centre for Productivity and Prosperity of the École des HEC Montréal for its financial support.
Ouellette P, Petit P (2018), "Efficiency Measurement: A Methodological Review and Synthesis". Data Envelopment Analysis Journal, Vol. 4 No. 1 pp. 67–107, doi: https://doi.org/10.1561/103.00000024
Download citation file:
