in: Roger White (ed.): Measuring multidimensional poverty and deprivation: incidence and determinants in developed countries, Palgrave Macmillan, 31-68
This chapter deals with the two concepts of efficiency and distributive justice in multidimensional poverty measures. These two concepts, though clearly related, are not the same, yet they are usually equated in one way or the other in multidimensional poverty measurement. In other words, multidimensional poverty measures either focus on the spread of simultaneous deprivations across the population (distribution-sensitivity) or the relationship between different poverty dimensions (association-sensitivity or efficiency). This chapter seeks to bring these two concepts together by measuring multidimensional poverty more holistically as the association-sensitive spread of simultaneous deprivations across the population. Focussing on the context of ordinal poverty measures, it provides axiomatic justification for distribution- and association-sensitive multidimensional poverty measures by using a new Sensitivity to Inequality Increasing Switch (SIIS) axiom. The resulting class of ordinal poverty measures is unique in the sense that it is association- and distribution-sensitive as well as fully decomposable. An empirical application for India illustrates the distinctive characteristics of these poverty measures.