The Global Trade Analysis Project (GTAP) model has been used to claim that genetically modified (GM) crops have substantial positive welfare potential for sub-Saharan Africa. The purpose of this paper is to critically examine these claims with a view to seeing if this type of modelling provides any meaningful basis for agricultural policy recommendations.Design/methodology/approach – One particular study is examined to show that because any possible negative effects of GM foods are ignored, optimistic assumptions are made about increases inproductivity and trade, and selective scenarios are used, the modelling must, perforce, result in positive welfare effects.Findings – It is found that, because of the assumptions of the study and the restricted scenarios investigated, all the modelling can do is estimate the size of the supposedly positive benefits.A negative outcome is not possible.Practical implications – It is argued that the GTAP model is inappropriate for sub-Saharan Africa and that the sweeping conclusions from this type of stylised modelling trivialize the complex poverty and socio-economic problems of sub-Saharan Africa.Originality/value – It is demonstrated that, in this case, the GTAP model is not only redundant, but also can yield risky policy recommendations for sub-Saharan Africa as it affects not only the livelihoods of millions of poor Africans but the ecological balance, biodiversity and economic independence of these nations.