How cash transfers made to women are used has important implications for models of household behavior and for the design of social programs. In this paper, the authors use the randomized introduction of an unconditional cash transfer to poor women in rural Ecuador to analyze the effect of transfers on the food Engel curve. There are two main findings. First, the authors show that households randomly assigned to receive Bono de Desarrollo Humano (BDH) transfers have a significantly higher food share in expenditures than those that were randomly assigned to the control group. Second, they show that the rising food share among BDH beneficiaries is found among households that have both adult males and females, but not among households that only have adult females. Bargaining power between men and women is likely to be important in mixed-adult households, but not among female-only households, where there are no men to bargain with. Finally, the authors show that within mixed-adult households, program effects are only significant in households in which the initial bargaining capacity of women was likely to be weak. This pattern of results is consistent with an increase in the bargaining power of women in households that received BDH transfers.