This preview shows pages 13–15. Sign up to view the full content.
between age at first job and health reflects a (child) work effect, an education effect or both. If some kids combine work and school, the independent variation in the two factors allows both to be included in the analysis. Any remaining influence of age at entry to the labour force must reflect a direct effect of child labour on health. Kassouf et al find evidence of such an effect but only for males 28-47 years and females 18-27 and 38-47. The dilution of the effect suggests that either the initial correlation between child work activity and health is largely spurious, reflecting the influence of 11Both studies control for the potential endogeneity of work (and school) time either through the use of instruments (Akabayashi and Psacharopoulos, 1999) or by the inclusion of a measure of innate intellectual ability in the regressions for test scores (Heady, 2000).
has intentionally blurred sections.
Sign up to view the full version.