This report presents an examination of the impact of PROGRESA on household expenditures. In particular, we focus on how participation in PROGRESA affects the acquisition of food by poor households. This analysis has been based largely on the ENCEL 98O, ENCEL99J and ENCEL99N surveys, although we also draw on the Seguimento surveys, administrative data and ENCEL 98M. Two themes underlie this work. One is the importance of understanding how PROGRESA actually operated on the ground in the period March 1998 to November 1999. The second is the need to understand behavioral parameters, in particular households’ marginal propensity to purchase calories out of additional income.