2004 UC Proceedings Abstract
Mapping Honduras Poverty: Indicators and Perceptions Track: Sustainable Development and Humanitarian Affairs Author(s): Grégoire Leclerc This work is about bringing together different perceptions of poverty to improve the design of pro-poor policies. I describe how various poverty indicators, classic ones such as unsatisfied basic needs (UBN) or novel ones based on local perceptions, can be obtained from census data. I find that classic survey-based poverty maps can be very sensitive to the choice of indicator and to the level of aggregation, which raises questions about the potential dangers of using a "canned" poverty map without questioning its validity. I therefore introduce simple methods to cross-check and tweak various indicators and maps to test their suitability for decision making. As an example I examine the extent of under-coverage and leakage of targeted policies that would rely on either a classic poverty measure similar to the ones used in many development projects in Honduras, or an alternative measure based on local perceptions. It appears that the latter measure, which is obtained from a combination of participatory appraisals and census data, is not only more accurate and robust than classic UBNs, but also more appropriate from the standpoint of either the poor or the policy-maker. This implies that most poverty targeting is ineffective as it is currently done, and that serious work is needed to 1) address the problem of error and bias in poverty estimates, 2) bring to decision makers and analysts the richness of raw census data with powerful tools to process and display them, and 3) embed the viewpoint of the poor in policy design. Grégoire Leclerc Centre de coopération Internationale en Recherche Agronomique pour le Développement (CIRAD) Laboratoire d'Enseignement et de Recherches en Géomatique, Ecole Supérieure Polytechnique Université Cheikh Anta Diop Dakar-Fann BP 25275 Sénégal E-mail: gregoire.leclerc@cirad.fr |