Files
Abstract
We evaluate the size of the welfare losses from using alternative “imperfect”
welfare indicators as substitutes for the conventionally preferred consumption indicator.
We find that whereas the undercoverage and leakage welfare indices always suggest
substantial losses, and the poverty indices suggest substantial losses for the worst
performing indices, our preferred welfare index based on standard welfare theory
suggests much smaller welfare losses. We also find that we cannot reject the hypothesis
that the welfare losses associated with using the better performing alternative indicators
are zero. In the case of our preferred welfare index, this reflects the fact that most of the
targeting errors, i.e., exclusion and inclusion errors, are highly concentrated around the
poverty line so that the differences in welfare weights between those receiving and not
receiving the transfers are insufficient to make a difference to the overall welfare impact.
Our results appear to be robust to the aversion to inequality assumed, as well as across
the various welfare indices.