This study evaluates the treatment effect of adoption of biofortified iron beans in Rwanda on household yields, bean consumption from own-production and purchases, bean sales, and the probability of being a net seller of beans. Because the adoption decision could be endogenous to these outcomes, we use an instrumental variable approach to quantify the impacts of adoption. Some of the iron bean varieties provide yield gains (measured as multiplication ratio, i.e. quantity harvested/quantity planted) of 19-25% over traditional varieties. Iron bean adoption also increases household consumption of own-produced beans by 14%. It reduces the probability of purchasing beans to meet consumption needs by 13% and reduces total purchases by 3.9 kg per adult equivalent on average. Finally, adoption increases the probability that a household sells beans by 16% and increases the probability that it is a net seller of beans by 14%. These findings are promising for the continued adoption of iron beans in Rwanda and elsewhere and provide evidence that biofortified crops are a good investment for nutrition, food security, and poverty reduction. Acknowledgement :