Evaluating construction contractors' efficiency in Hong Kong using Data Envelopment Analysis Assurance Region model

Purpose – The purpose of this paper is to investigate whether a few leading Hong Kong construction firms are also efficient contractors. Theoretically, market leaders can exploit the virtue of economy of scale, learning curve and cost leadership to consolidate their competitiveness. However, if market share is key success factor for firms to become and stay competitive, smaller contractors will be further marginalized if the construction market continues to shrink as Hong Kong's economy approaches maturity. Hence the research question is: Have a few leading contractors been taking advantage of their market size to be efficient in Hong Kong? Design/methodology/approach – Based on the Data Envelopment Analysis Assurance Region (DEA_AR) model, the efficiency scores of 17 major Hong Kong contractors are compiled over the last ten years. DEA is a non-parametric approach to examine the relative efficiency among different firms, in terms of output/input ratios. Unlike traditional I-O analysis which relies on a static table of coefficients, DEA does not require any priori and usually subjective weights for inputs or outputs. Findings – The results identify efficient contractors, among which the three leading ones (China State Construction, Shui On Construction and Yau Lee Holdings) have captured the majority of housing construction contract values. Nevertheless, only China State Construction has ranked consistently as most efficient over time. The efficiency scores of Shui On Construction have decreased successively from 2002 to 2009, and it has been the least efficient contractor from 2006 to 2009. Yau Lee Holdings has a V-curve of efficiency scores. Originality/value – This study represents the first attempt to adopt a DEA_AR model to evaluate construction contractors' efficiency, which reduces the number of zeroes and the variation in weights for estimation and enhances the adequacy for individual contractor's efficiency scores, compared to other types of DEA models such as