Sibling correlations and intergenerational mobility in Latin America

In this article, we use sibling correlations in schooling to measure differences in intergenerational mobility for 16 Latin American countries. The results show that there are substantial differences in mobility within Latin America. Social mobility increases with mean schooling and income per capita but is only mildly associated with public expenditures on education. We argue in this article that by comparing sibling correlations of schooling, we can learn about the differences in the degrees of social mobility among countries (e.g., we can learn about the extent to which family background determines socioeconomic success in different countries). Our analysis is limited for obvious reasons. First, schooling is an imperfect measure of child outcomes. School quality, for example, is conspicuously absent from our analysis, as are differences in parental investments. Second, schooling does not capture all possible channels through which family background affects socioeconomic success. Family connections, for example, can make all the difference when children enter the labor force. Parental wealth also can make a big difference later in life. Both factors, however, have been left out of our analysis. The above-mentioned problems notwithstanding, we believe that, especially for developing countries, schooling provides an early glimpse of what is to come, and hence it can be used to gauge differences in social mobility. Our results are noncontroversial in that they reiterate a piece of conventional wisdom: education is perhaps the most expeditious way to enhance equality of opportunity. We find, in particular, that access to education (measured, e.g., by average schooling gaps) is a powerful predictor of the importance of family background in socioeconomic performance. We also find that in Latin America, social mobility is only loosely related to income per capita and inequality is strongly associated with sorting in marriage markets.