This article examines the unexpected impacts of the need-and-merit-based college scholarship program Ser Pilo Paga (SPP) on schools aggregated academic performance and dropout rate. My analysisuses difference in differences methodology to estimate variation in average test scores across cohortsfrom schools that had students selected for SPP. SPP could have had an impact on schools’ resultsthrough different potential mechanisms. SPP could emerge motivational positive incentives to increaseeffort, related with better academic results. This incentive seems to be stronger for last year students. SPP could also cause the reduction of dropout rates caused by the incentive scheme to keep studyinguntil the last year of school and have the opportunity to attend university, producing a change in thecomposition of the distribution of students’ skills. Moreover, schools may react strategically in orderto meet students’, parents’ and public demands to have more selected students. My results suggest thatschools that had SPP students performed better in standardized test for 11th grade, but worse in 9th and5th grade standardized tests. After dealing with some issues related to parallel trend assumption, myestimates for high school dropout rates suggest that SPP reduced such rates supporting the change incomposition hypothesis. I also analyze heterogeneous effects on the proportion of need-eligible studentsand I find that the effect of SPP on outcomes becomes more positive in schools that have a bigger proportion of eligible students.
