ILO Convention No. 138, Art. 7(b) stipulates that light work may be permitted as of the age of 12 or 13 provided it does not “prejudice attendance at school” nor “the capacity to benefit from the instruction received”. This raises the issue of the impact of child labour on schooling for children in these age groups. Notwithstanding a large and rapidly expanding literature on child labour, there is not much empirical evidence on this issue since much of this literature has concentrated on analysing the causes of child labour rather than studying its consequences, especially for the child’s learning. The cost of child labour for human capital accumulation has simply been assumed rather than formally investigated. The limited evidence that does exist on this issue makes little or no attempt to control for the endogeneity of child labour hours in the estimation. Such endogeneity can arise because of the reverse causation of child labour by learning disadvantage and lack of necessary intrinsic skills. The present study seeks to fill this significant gap in the literature on child labour. The exercise is conducted on the data sets involving 12-14 year old children from 7 countries collected under the ILO’s “Statistical Information and Monitoring Programme on Child Labour” (SIMPOC). The chosen countries span a wide spectrum geographically, culturally and in the range of economic development, namely, from the developed European country context of Portugal to poor Afro-Asian countries such as Namibia and Philippines. The list also includes Sri Lanka which, notwithstanding its status as a developing country, has school enrolment rates that approach those in developed countries.