A Nonparametric Analysis of Production Models with Multidimensional Quality

This article proposes a methodology to measure the characteristics of intermediate products when quality is multidimensional. It uses a general representation of the multioutput technology via directional distance functions and constructs quality indicators based on differences rather than ratios. The quality indicators may be used to evaluate firms' output, taking into account the whole set of quality attributes. The paper explores how aggregate quality may vary with the production level and finds a trade-off between quantity and aggregate quality in wine grapes.