In the middle of the nineties the rural population in Burkina Faso was seriously hit by rising food prices. Whereas cotton farmers were able to cope with this shock given the simultaneous boom in the cotton sector, food crop farmers had to withdraw children from school and to let them work more intensively. Using the exogenous character of the income variation as an instrument allows to disentangle the pure effect of parental income from effects related to parental education, family background and other unobservables. A set of simple policy simulations illustrates the potential of unconditional cash transfers to raise schooling levels and to protect investment in children's education against transitory income shocks. Although the involved effects are not negligible and much higher as simulations based on the pure OLS effect would suggest, they also show that making transfers conditional on attendance might largely increase the efficiency of such transfers.
Child Labor Education Income Elasticity of Education Agricultural Shocks Cotton Production Burkina Faso