We employ a regression model with spillover effects to show that the impact of peer quality on wages is quite large. We estimate that a 10 percent increase in peer quality implies a 2.1 percent increase in an individual's wage. In addition, we estimate the external returns to education using a novel identification strategy, which is strictly based on the peer effect channel, netting out the role of homophily and labor market sorting. We show that a one-year increase in the co-workers' education leads to a 0.58 percent increase in wages. We also show that both effects fade smoothly over time.