This paper evaluates a microsimulation technique by comparing the simulated outcome of a program with its actual effect. The ex ante evaluation is carried out for a conditional cash transfer program, where poor households were given money if the children attended school. A model of occupational choice is used to simulate the expected impact of the program. The results suggest that the transfer would indeed increase school attendance and do more so among girls than boys. While the simulated effect tends to be larger than the actual effect, the latter lies within bootstrapped confidence intervals of the simulation.