Estimating Multidimensional Poverty: A New Methodological Approach

Abstract

ABSTRACT This paper proposes a new welfare-based multidimensional poverty index (WMPI). The index introduces extent and severity parameters into the multidimensional poverty estimation. The method also introduces a new individual and indicator-specific weighting scheme. This approach is simple to apply and accurately measures poverty with less susceptibility to measurement error. The index satisfies all multidimensional poverty axioms, including transferability and monotonicity. The newly proposed method has been applied to Ethiopian data and compared with similar methods. The estimation result reveals that the multidimensional poverty rate for Ethiopia with a tight poverty policy is 38.5 percent, while with a loose poverty policy, it amounts to 33.1 percent. On the contrary, with the Alkire-Foster method, the multidimensional poverty rate is 43.1 percent.