ABSTRACT
This article discusses a way of solving the ‘Greene Problem’ that has been termed for a technical difficulty in estimating a profit system with technical and allocative inefficiencies. As the ‘Greene Problem’ is due to the existence of unobserved technical inefficiency interactive in a nonlinear form, the article proposes a use of homogeneity in technology to untangle the nonlinear form, enabling the estimation of a profit system as well as the identification of the source of inefficiencies.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Neither Tobit regression in the second stage nor maximum likelihood estimation (MLE)-based truncated normal distribution in the first stage corrects this bias unless a scaling factor (or a correcting factor) is introduced. For details, see Simar and Wilson (Citation2008, 91–95).
2 This problem is often termed as ‘Greene Problem’ because Greene (Citation1980) noticed this problem and irrelevantly imposed an assumption of independence between allocative inefficiency and cost to avoid this problem.
3 The proofs of the Proposition 1 is provided in the Appendix.
4 The proofs of the Proposition 2 is provided in the Appendix.
5 Kumbhakar and Lovell (Citation2000, 57–60 provide a complete theoretical explanation of the decomposition of profit efficiency.