Demand versus Returns? Pro-Poor Targeting of Business Grants and Vocational Skills Training

Macours, Karen ; Premand, Patrick ; Vakis, Renos

World Bank, Washington, DC

关键词: ACCESS TO MARKETS;    ACCOUNTING;    AGE GROUPS;    AGRICULTURAL ACTIVITIES;    AGRICULTURAL ACTIVITY;   

DOI  :  10.1596/1813-9450-6389
RP-ID  :  WPS6389

学科分类：社会科学、人文和艺术（综合）

来源: World Bank Open Knowledge Repository

【 摘 要 】

Interventions aimed at increasing theincome generating capacity of the poor, such as vocationaltraining, micro-finance or business grants, are widespreadin the developing world. How to target such interventions isan open question. Many programs are self-targeted, but ifperceived returns differ from actual returns, thoseself-selecting to participate may not be those for whom theprogram is the most effective. The authors analyze anunusual experiment with very high take-up of business grantsand vocational skills training, randomly assigned amongnearly all households in selected poor rural communities inNicaragua. On average, the interventions resulted inincreased participation in non-agricultural employment andhigher income from related activities. The paperinvestigates whether targeting could have resulted in higherreturns by analyzing heterogeneity in impacts by statedbaseline demand, prior participation in non-agriculturalactivities, and a wide range of complementary assetendowments. The results reveal little heterogeneity alongobserved baseline characteristics. However, the pooresthouseholds are more likely to enter and have higher profitsin non-agricultural self-employment, while less poorhouseholds assigned to the training have highernon-agricultural wages. This heterogeneity appears relatedto unobserved characteristics that are not revealed bystated baseline demand, and more difficult to target. Inthis context, self-targeting may reduce thepoverty-reduction potential of income generatinginterventions, possibly because low aspirations limit thepoor's ex-ante demand for productive interventionswhile the interventions have the potential to increase thoseaspirations. Overall, targeting productive interventions topoor households would not have come at the cost of reducingtheir effectiveness. By contrast, self-targeting would havelimited poverty reduction by excluding the poorest.