Does child labor always decrease with income?

"This paper investigates the relationship of household income with child labor. The analysis uses a rich dataset obtained in the context of a conditional cash transfer program in a poor region of Nicaragua in 2005 and 2006. The program has a strong productive emphasis and seeks to diversify the work portfolio of beneficiaries while imposing conditionalities on the household. The author develops a simple model that relates child labor to household income, preferences, and production technology. It turns out that child labor does not always decrease with income; the relationship is complex and exhibits an inverted-U shape. Applying the data to the model confirms that the relationship is concave when all children (8-15 years of age) are included in the sample. Expanding the analysis by stratifying the sample by age and gender shows that the relationship holds only for older children, both genders. The author investigates the effect of the conditional cash transfer program on child labor. The results show that the program has a decreasing effect on total hours of work for the full sample of children. Disentangling labor into two types - physically demanding labor and non-physical labor - reveals that the program has opposite effects on each type; it decreases physically demanding labor while increasing participation in non-physical (more intellectually oriented) tasks for children. "--World Bank web site.