Equivalence and implementation of alternative methods for determining returns to scale in data envelopment analysis

This paper discusses alternative methods for determining returns to scale in DEA. The methods for estimating returns to scale in DEA, as developed by Banker and coworkers, are proved to be conceptually equivalent to the two-stage methods of Färe, Grosskopf and Lovell when their assumptions apply. Here the emphasis is on the CCR model of DEA and very simple methods are introduced for determining returns to scale locally with this model by reference to Banker's concept of Most Productive Scale Size.