Using baseline data from a randomized experiment, this article extends and tests in the context of health, the feasibility of a recently proposed reduced form approach to ex ante evaluations of social programs with an application to a conditional cash transfer program in Nicaragua. It uses a behavioural model to estimate the impact on preventive care utilization outcomes for children younger than 3 years. It validates the model with the results of the experiment and then simulates two alternate policy scenarios. The model performs well in predicting the health related outcomes and shows different results for the two sets of policy scenarios. In addition, simulations are also carried out for the school component of the cash transfer program. Copyright © 2012 John Wiley & Sons, Ltd.