# Brazilian conditional cash transfer programme’s impact on youth human capital outcomes: the 2004 Pelotas Birth Cohort

## TL;DR

This study examines how Brazil's Bolsa Família Programme affected youth outcomes at age 18, finding limited impact on most measures.

## Contribution

The study evaluates the long-term human capital outcomes of a conditional cash transfer programme using a longitudinal birth cohort.

## Key findings

- BFP participation showed no significant effects on most outcomes like schooling, work, or crime at age 18.
- Weak evidence suggested possible reductions in drug use and non-violent crime.
- Longer exposure to the programme may be needed for measurable impacts.

## Abstract

Brazil’s Bolsa Família Programme (BFP) is one of the largest conditional cash transfer (CCT) programmes globally. This study evaluated the impact of BFP during childhood on human capital outcomes at age 18.

We analysed data from 2743 participants in the 2004 Pelotas Birth Cohort. BFP participation was assessed at ages 6 and 11, and outcomes at age 18 included grade repetition, being neither in school nor working, criminal behaviour, tobacco use, binge drinking and drug use. To address selection bias, we used Propensity Score Matching to estimate the average treatment effect on the treated (ATT).

No significant effects of BFP participation at ages 6 or 11 were observed for most outcomes: not studying or working (ATT 0.01; 95% CI −0.04 to 0.06), tobacco use (ATT −0.03; 95% CI −0.10 to 0.05), binge drinking (ATT −0.05; 95% CI −0.15 to 0.05), violent crime (ATT 0.00; 95% CI −0.08 to 0.07) or any crime (ATT −0.06; 95% CI −0.14 to 0.03). Weak evidence suggested reductions in non-violent crime (ATT −0.04; 95% CI −0.09 to 0.01; p=0.072) and drug use (ATT −0.07; 95% CI −0.17 to 0.02; p=0.075), but a possible increase in grade repetition (ATT 0.07; 95% CI −0.02 to 0.16; p=0.097). Analyses of BFP receipt at only one time point showed no effects, indicating that longer exposure may be necessary for impact.

Participation in the BFP during childhood was not associated with significant improvements in most human capital outcomes at age 18. Nonetheless, potential reductions in drug use and non-violent crime, and the complex relationship with school performance, warrant further investigation into the long-term and multidimensional effects of CCTs. These results are specific to the Pelotas cohort and may not be generalisable to different geographical settings within Brazil.