28

The study examines the impact of poverty on child labor. First, the “luxury axiom” seems relatively robust. Econometric analysis shows that the proportion of economically active children aged 5-14 is definitely higher in “poor” households than in “non-poor” households. Second, child labor may reflect household strategy aimed at minimizing the risk of interruption in the income stream. Such results should be interpreted in the context of two phenomena: (1) the increase in total poverty between 1998 and 2003, which involved a fall in lasting poverty and a rise in transient poverty; (2) a slowdown in the redistribution process conducted through remittances from Côte d’Ivoire.