| Abstract: | Despite every policymaker's recognition, enrollment rates of the low‐income countries remain low. A simple framework of understanding educational outcomes is presented using a unitary model with an altruistic parent and a child. The traditional interventions, so‐called supply‐side policies, and recent innovation of relaxing constraints faced by households, the conditional transfer programs or so‐called demand‐side policies, are reviewed. In addition, recent trends on estimation technique are discussed. It has been argued that randomization is clearly the best for inference, however, one may still want to choose the optimal combination of randomized experiments and observational data, as the former requires more resources and time. This is particularly true for economics than other hard sciences, partly because of our inability to fine‐tune the control, and partly because of our lack of solid microfoundation than other sciences when an experiment shows unexpected results. |
