The Food and Agriculture Organization (FAO) and the World Bank suggest that agricultural growth is the most effective way to fight (child) malnutrition. Whether this is indeed the case is still very much debated, partly because there is little direct evidence on this topic. Using a dataset of 50 MDG1 countries observed between 1991 and 2009, this paper estimates and compares the impact of agricultural growth on child stunting against those of industrial growth and services growth. We find that, to achieve a 1 percentage point reduction in child stunting, a 11.1% increase in agricultural GDP per capita is necessary, against a 9.7% increase in industrial GDP per capita and a 7.8% increase in services GDP per capita. In other words, contrary to the policy narrative developed by many international development organizations, the services sector is the most effective engine towards reducing child stunting. Finally, we find weak evidence of the impacts of food prices on child stunting and no evidence of the impacts of food prices volatility for the period observed.