We examined whether Brazil’s educational spending on public primary schools resulted in better quality of education, 2003-2009. Our hypothesis was municipalities that received higher government spending on primary education had higher students’ tests scores. We used a panel analysis with fixed effects, including a simulated instrumental variable to control for endogeneity. Even though the magnitude of all significant estimated coefficients increased, when controlling for endogeneity, they were still very small. Findings proved that the relationship between quality of education and educational spending was not optimal, weakly contributing to the increase in test scores.