This paper uses experimental panel data for Mexico from 1997 to 2000 in order to test assumptions on the impact of a conditional cash transfer (CCT) program on child labor and school attendance, adding to the literature by emphasizing the differential impact on indigenous households. Using data from the CCT program, PROGRESA (later on known as OPORTUNIDADES), we investigate the interaction between child labor, education and indigenous households. While indigenous children had a greater probability of working before the intervention, this probability is reversed after treatment in the program. Indigenous monolingual children also had lower school attainment compared with Spanish-speaking or indigenous bilingual children. After the program, school attainment among indigenous children increased, reducing the gap. In terms of child labor, the larger reduction is in the group of bilingual children.