Is child work necessary?

This paper investigates why children work by studying the wage elasticity of child labour supply. Incorporating subsistence constraints in to a model of labour supply, we show that a negative wage elasticity favours the hypothesis that poverty compels work whereas a positive wage elasticity would favour the alternative view that children work because the relative returns to school are low. Distinguishing between these alternatives is important for policy. Existing studies have concentrated on the income elasticity, but this tells us nothing other than that leisure (or education) is a normal good. Using a large household survey for rural Pakistan, we estimate structural labour supply models for boys and girls in wage work, conditioning on full income and a range of demographic variables. Our estimates describe a forward falling labour supply curve for boys, consistent with the view that boys work on account of the compulsions of poverty. This is less clear in the case of girls. Therefore raising the return to schooling for girls may draw them out of work, but eliminating boys' wage work requires alleviation of the poverty of their households. Trade sanctions or bans on child labour may have deleterious consequences for these households unless they are compensated for the loss in income.