EX ANTE IMPACT ASSESSMENT OF CONDITIONAL CASH TRANSFER PROGRAMS USING AN AGENT-BASED MODEL

A model for ex ante impact evaluation of Conditional Cash Transfer Programs (CCT) is described. Assessment of the impacts of such programs is critical for the design of public policies. We developed an agent-based model, aimed at predicting the decisions that families would make, given their characteristics and environment but without any assumption of rationality. Thus, expected behaviors and effects can be predicted, based on the individuals' traits, by computing the odds of making different choices concerning children´s education. The model we developed has the capacity to simulate the decision making process, including adaptive behaviors and learning strategies to realistically represent actual situations in more detail. The model was applied to Brazil´s Bolça Familia program in 2011 and results compared to a previous study. Our model indicates that this Conditional Cash Transfer program has been successful and that, as a consequence, school attendance has increased during the period between the evaluations.