This study examines the effect of child labour on children's learning outcomes in Ghana, with a focus on the mediating role of cognitive skills development. Using nationally representative data from the Multiple Indicator Cluster Surveys (MICS) for children aged 5–14 years, the study employs an instrumental variable probit (IV-probit) model to address the endogeneity of child labour participation. The results demonstrate that child labour significantly reduces both literacy and numeracy outcomes. Importantly, child labour also impairs executive function readiness, while executive function itself exerts a strong positive effect on learning outcomes. The inclusion of cognitive skills reduces the magnitude of the child labour coefficient, providing evidence of partial mediation. Additional findings highlight the importance of parental education, parental support, school quality, and household size in shaping children's learning trajectories. The study concludes that child labour affects educational outcomes through two pathways: time constraints and impaired cognitive development. These findings underscore the need for integrated policy interventions targeting both child labour reduction and cognitive skill enhancement to improve human capital formation in Ghana.