Under Indonesia's anti-poverty program, IDT, the government provided selected poor villages with grants of the same value, regardless of population size. Exploiting the variation in per household grant value that is caused by this program design, the author estimates the returns to public grants, which are designated for investment loans. Results show that the returns are heterogeneous. Villages with pre-existing market facilities demonstrate increases in male labor supply, per capita income (PCI) and per capita expenditure (PCE). However, villages not accessible by land exhibit few changes in labor supply or PCI and yet an increase in PCE, particularly on festivals. These results suggest that the returns to investment capital are limited without a basic economic infrastructure.