This paper models the short and medium-run impact of aid on migration, considering alternatively the effect of unconditional and conditional cash transfers to financially constrained households. Data from the evaluation of a Mexican development program, Progresa, are used to estimate the effect of the potential grant size on migration. The empirical analysis is consistent with model prediction. It shows that the program is associated with an increase in international migration, which is also a positive function of size of potential transfer. The grant may loosen financial constraints. At the same time, fine-tuned conditional grants targeting prospective migrants (in the form of secondary school subsidies) reduce the short-term migration probability. As regards medium-term migration, secondary school beneficiaries are not more likely to migrate than the control group after they complete the subsidised education cycle.