Economic inefficiency measurement of input spending when decision-making units face different input prices

• Published in: The Journal of the Operational Research Society Vol. 55; no. 10; pp. 1102 - 1110
• Main Authors: Fukuyama, H, Weber, W L
• Format: Journal Article
• Language: English
• Published: Basingstoke Taylor & Francis 01-10-2004; Palgrave Macmillan Press; Palgrave; Taylor & Francis Ltd
• Subjects: Allocative efficiency; application areas (BA: banking); Applied sciences; Consumer spending; Cost allocation; Cost efficiency; Cost estimates; Data envelopment analysis; Decision making; Decomposition; Distance functions; Efficiency; Efficiency metrics; Exact sciences and technology; Input prices; Mathematical programming; Mathematical vectors; Operational research and scientific management; Operational research. Management science; Portfolio theory; Prices; Spending; techniques/approaches (DE: data envelopment analysis); Theoretical Papers; Measurement; Decision support system; Efficiency; Decision making; Banking; techniques/approaches (DE: data envelopment analysis); application areas (BA: Banking); Data envelopment analysis
• ISSN: 0160-5682; 1476-9360
• DOI: 10.1057/palgrave.jors.2601750

In a recent paper, Kaoru Tone (J Opl Res Soc (2002) 2: 429-444) showed that when the Farrell measure of cost efficiency is estimated for two firms that have different input prices, a firm with higher costs can be deemed more efficient than a firm with lower costs. As an alternative approach, Tone proposed a radial cost efficiency measure that is estimated using levels of spending on each input, rather than input quantities. Thus, firms with higher costs are less efficient than firms with lower costs. In this paper, we extend Tone's approach by allowing for non-radial changes in spending. Our approach builds on earlier work by Luenberger (J Math Econ (1992) 21: 461-481) and Chambers et al (J Econ Theo (1996) 70: 407-419) who use directional distance functions to measure inefficiency. We provide an example and illustration of our approach using Japanese bank data.