Starting from the existing output‐oriented plant capacity measure, this paper proposes a new input‐oriented plant capacity measure. Furthermore, we empirically illustrate both input‐oriented and output‐oriented decompositions of technical efficiency integrating these technical concepts of capacity utilization. In particular, we pay attention to the impact of convexity of the technology on both input‐oriented and output‐oriented plant capacity measures.

No abstract

Starting from the existing input-and output-oriented plant capacity measures, this contribution proposes new long-run input-and output-oriented plant capacity measures. While the former leave fixed inputs unchanged, the latter allow for changes in all input dimensions to gauge either a maximal plant capacity output or a minimal input combination at which non-zero production starts. We also establish a formal relation between the existing short-run and the new long-run plant capacity measures. Furthermore, for a standard nonparametric frontier technology, all linear programs as well as their variations are specified to compute all efficiency measures defining these short-and long-run plant capacity concepts. Furthermore, it is shown how the new long run plant capacity measures are identical to existing models of a variable returns to scale technology without inputs or without outputs (see Lovell and Pastor (1999)): thus, we offer an interesting production economic justification for these models. Finally, we numerically illustrate this basic relationship between these short-run and long-run technical concepts of capacity utilisation.

The purpose of this contribution is to empirically implement and supplement the proposals made by Podinovski (2004b) to explore the nature of both global and local returns to scale in nonconvex nonparametric technologies. In particular, we both propose a simplified method to compute the global returns to scale and employ some secondary data sets to investigate the frequency of the special case of global sub-constant returns to scale. Furthermore, when determining global returns to scale using both convex and nonconvex technologies, we verify how often the resulting information is concordant or conflicting. Finally, besides comparing the FDH and DEA evolution of ray-average productivity for some typical individual observations, we introduce in the literature two original methods for the determination of local returns to scale in non-convex technologies. * We acknowledge comments made by participants in the North American Productivity Workshop VIII (Ottawa) and in the 10th Asia-Pacific Productivity Conference (Brisbane). The constructive comments of three referees are gratefully acknowledged.

In a multiple input–output setting, we offer an example of a full‐blown efficiency analysis based on the free disposal hull assumption of a non‐convex technology. On a cross section of 92 units, for the time period 2006–2007, we measure technical and allocative efficiency, cost efficiency, as well as returns to scale and technical progress. Furthermore, by introducing a new cost‐efficiency measure that is corrected for differences in the scale of operations (average cost efficiency), we are able to obtain an estimate of cost reductions ensuing from network restructuring based on optimal scale sizes. A detailed regional analysis of results is presented. Differences in efficiency scores, ascertained across space and time, are assessed by means of non‐parametric inference testing. The input ratio is found not to be statistically correlated with cost‐efficiency scores. Local provision of public services at a provincial level results in a considerable amount of average cost inefficiency.