Endogenous Weights and Multidimensional Poverty: A Cautionary Tale

Highlights

•A large and growing body of work uses endogenous (data driven) weights to compute multidimensional poverty.

•Broad classes of endogenous weights violates key properties of poverty indices such as monotonicity and subgroup consistency.

•Using data from Ecuador and Uganda we show that these violations are widespread.

•Poverty evaluation under these circumstances are seriously compromised.

•Our results can be extended to other composite welfare measures such as the widely used asset indices.

Abstract
Multidimensional poverty measures have become a standard feature in poverty assessments. A large and growing body of work uses endogenous (data driven) weights to compute multidimensional poverty. We demonstrate that broad classes of endogenous weights violates key properties of poverty indices such as monotonicity and subgroup consistency, without which poverty evaluation and policy targeting are seriously compromised. Using data from Ecuador and Uganda we show that these violations are widespread. Our results can be extended to other composite welfare measures such as the widely used asset indices.