As exemplified by the Millennium Declaration of the United Nations, the reduction of poverty and hunger are now seen as central objectives of international development. Yet the modalities for attaining these goals are contested. Further, while it might be assumed that interventions that alleviate poverty will automatically reduce hunger, a number of studies of the relationship between income and the acquisition of food suggest that this assumption may be incorrect. This paper contributes to this debate through an analysis of a Mexican antipoverty program called PROGRESA (the Programa de Educacion, Salud y Alimentacion). PROGRESA provides cash transfers linked to children’s enrollment and regular school attendance and to clinic attendance. By 2000, it reached approximately 2.6 million families, about 40 percent of all rural families and about one-ninth of all families in Mexico. We use a longitudinal sample of approximately 24,000 households from 506 communities. A distinguishing characteristic of this sample was that communities were randomly selected for participation in PROGRESA, while the rest were introduced into the program at later phases. Exploiting this feature in our analysis, we find that households receiving PROGRESA benefits increased caloric acquisition compared to comparable households not receiving these benefits. By November 1999, beneficiary households in treatment localities obtained 7.1 percent more calories than did comparable households in control localities. Perhaps more significantly, we find that the impact is greatest on dietary quality as measured by the acquisition of calories from vegetable and animal products—a finding consistent with the view of respondents themselves that PROGRESA was enabling them to “eat better.”