This paper explores the inequality of numeracy and education by studying school years and numeracy of the rich and poor, as well as of tall and short individuals. To estimate numeracy, the age-heaping method is used for the 18th to early 20th centuries. Testing the hypothesis that globalization might have increased the inequality of education, we find evidence that 19th century globalization actually increased inequality in Latin America, but 20th century globalization had positive effects by reducing educational inequality in a broader sample of developing countries. Moreover, we find strong evidence for Kuznets’s inverted U hypothesis, that is, rising educational inequality with GDP per capita in the period until 1913 and the opposite after 1945.