In this paper, we apply a convex hull approach to counterfactual analysis of trade openness and growth. The experiments we choose evaluate the importance of trade openness for growth across African countries. Specifically, we ask the question what would happen if African countries were more open?. The evidence indicates that several countries don´t fall within the convex hull of the observed data and therefore counterfactual inferences are risky. This conclusion is at odds with the literature arguing that greater trade openness would unequivocally lead to higher growth in Africa.