Education conditional cash transfer programs may increase school attendance in part due to the information they transmit to parents about their child's attendance. This paper presents experimental evidence that the information content of an education conditional cash transfer program, when given to parents independently of any transfer, can have a substantial effect on school attendance. The effect is as large as 75 percent of the effect of a conditional cash transfer incentivizing parents, and not significantly different from it. In contrast, a conditional transfer program incentivizing children instead of parents is nearly twice as effective as an "information only" treatment providing the same information to parents about their child's attendance. Taken together, these results suggest that children have substantial agency in their schooling decisions. The paper replicates the findings from most evaluations of conditional cash transfers that gains in attendance achieved by incentivizing parents financially do not translate into gains in test scores. But it finds that both the information only treatment and the alternative intervention incentivizing children substantially improve math test scores.