Generalized Fisher index or Siegel–Shapley decomposition?

Abstract

It is generally believed that index decomposition analysis (IDA) and input–output structural decomposition analysis (SDA) [Rose, A., Casler, S., Input–output structural decomposition analysis: a critical appraisal, Economic Systems Research 1996; 8; 33–62; Dietzenbacher, E., Los, B., Structural decomposition techniques: sense and sensitivity. Economic Systems Research 1998;10; 307–323] are different approaches in energy studies; see for instance Ang et al. [Ang, B.W., Liu, F.L., Chung, H.S., A generalized Fisher index approach to energy decomposition analysis. Energy Economics 2004; 26; 757–763]. In this paper it is shown that the generalized Fisher approach, introduced in IDA by Ang et al. [Ang, B.W., Liu, F.L., Chung, H.S., A generalized Fisher index approach to energy decomposition analysis. Energy Economics 2004; 26; 757–763] for the decomposition of an aggregate change in a variable in r = 2, 3 or 4 factors is equivalent to SDA. They base their formulae on the very complicated generic formula that Shapley [Shapley, L., A value for n-person games. In: Kuhn H.W., Tucker A.W. (Eds), Contributions to the theory of games, vol. 2. Princeton University: Princeton; 1953. p. 307–317] derived for his value of n-person games, and mention that Siegel [Siegel, I.H., The generalized “ideal” index–number formula. Journal of the American Statistical Association 1945; 40; 520–523] gave their formulae using a different route. In this paper tables are given from which the formulae of the generalized Fisher approach can easily be derived for the cases of r = 2, 3 or 4 factors. It is shown that these tables can easily be extended to cover the cases of r = 5 and r = 6 factors.