In contrast to the popular policy claim that sport might serve as vehicle to meet the Millennium Development Goals, empirical evidence based on large-scale survey data is largely missing. We use panel data based on a cohort of children and employ propensity score matching to identify the effects of sports participation on child development in Peru. Our findings suggest that participation in a sports group has positive impacts on subjective health and a measure of social capital. However, and in contrast to developed countries, we find no statistically significant effects on well-being and human capital formation.