Abstract
PurposeThis study aims to examine the impact of Brazil's Bolsa Fam & iacute;lia (BF) program on children's educational outcomes, addressing concerns about endogeneity and selection bias in prior research.Design/methodology/approachUsing household survey data and an instrumental variables strategy based on administrative eligibility criteria, this study estimates the causal effect of BF on grade progression, repetition, and dropout, with heterogeneity by age and sex.FindingsThe program did not significantly reduce dropout rates but was associated with increased grade repetition and decreased progression, particularly among younger children. Boys experienced more negative effects than girls. Results suggest that earlier studies may have overstated educational gains due to endogeneity bias.Originality/valueBy leveraging quasi-random administrative variation, this study provides causal estimates of BF's educational impacts, revealing unintended effects on grade progression and repetition that previous observational studies could not identify.