Changing patterns of child labor around the world since 1950 : the roles of income growth, parental literacy, and agriculture

Using country-level data, this report lays out the broad stylized facts regarding the relationship between child labor and per capita GDP, adult literacy, and the share of agriculture in the economy. The relationship between child labor force participation and per capita income is convex and stable over time. The implication is that as a country develops, child labor will decrease, but at a decreasing rate. At some point, further reductions in child labor may require more than just increasing per capita income. Child labor also is affected by the perceived return to child time in the labor market relative to child time in school. The strength of demand for child labor is highly correlated with the share of agriculture in the economy. Parental perception of the importance of education is highly correlated with the parents'own education. A 10 percent decrease in agriculture's share of GDP, decreases child labor by about 20 percent. A 10 percent decrease in adult illiteracy also decreases child labor by 20 percent. In Latin America, all three of these factors have contributed to decreases in child labor since 1950. Increases in per capita income have lowered the child labor participation rate by 2.9 percentage points. The reduction in adult illiteracy was responsible for a 4.2 percentage point reduction in child labor participation and reductions in agriculture's share of production, lowered child labor by an additional 1.2 percentage points. It is possible that income redistribution may lead to lower incidence of child labor, even if increases in average income will not. However, child labor is still found even at the upper end of the income distribution in Latin America. Consequently, income transfer programs alone will not eliminate child labor.