Using multidimensional poverty measures in impact evaluation : Emergency housing and the “declustering” of disadvantage

Abstract

Abstract: During the past two decades, impact evaluation and multidimensional poverty measurement have gained increasing relevance in development practice and research. The objective of this paper is to propose empirical strategies for using the multidimensional poverty measures proposed by Alkire and Foster (2011) in impact evaluation. The principal argument for taking this approach is that it provides a means for assessing the effects of social programmes on the simultaneous occurrence or joint frequency of deprivations, what Wolff and de-Shalit (2007) call the “clustering” of disadvantage. These strategies are applied to the evaluation of the NGO TECHO’s emergency housing programme in the informal settlements of Buenos Aires, Argentina. The results show that the programme produces a large reduction in the simultaneous occurrence of disadvantages.