Explaining cross-country differences in policy response to child labour

We develop a model of child labour where poverty and inequality combine to determine policy response to child labour. If there are strategic complementarities between parents' decisions to educate their children and firms' technology choice, multiple school-enrolment equilibria arise. Only rich countries and those that are not `too' poor and have a low wealth inequality benefit from adopting child labour laws. This is because such laws commit an economy with either of those initial conditions to the full school-enrolment equilibrium which Pareto-dominates all other equilibria. Moreover, wealth redistribution is not sufficient to eliminate child labour.