On Value Efficiency and Data Envelopment Analysis

This paper builds on earlier efforts to integrate preferences into a DEA model. Rather than including an explicit utility function as a constraint to a DEA problem as in Färe et al. (2002), here we use the notion of dominance to identify a ’¡Ærevealed’¡Ç indifference curve, which is then included in the DEA model. This allows us to identify what Halme et al. (1999) call the most preferred solution (MPS) as part of the solution to our modified DEA problem. This in turn allows us to decompose what they would call value efficiency into a technical and preference related component.