Coffee Production Effects on Child Labor and Schooling in Rural Brazil

I use variation in coffee production to measure changes in local economic conditions, which proxy for the value of children's time. I test how this variation affects child labor and schooling in Brazilian coffee regions using seven rounds of household surveys. Increases in county-level value of coffee production led to more work and less schooling among boys, and lower schooling among girls. Girls without childcare responsibilities were more likely to work. Poorer boys and girls were withdrawn from school, while richer children were more likely to attend. Thus, during periods of economic growth, education of the poor is adversely affected