Educational Assortative Mating and Household Income Inequality: Evidence from Brazil, Indonesia, Mexico, and South Africa

Abstract

This paper presents new empirical evidence from four emerging economies on the relationship between educational assortative mating and household income inequality. It also introduces a new methodological approach, first proposed in my earlier work, that overcomes limitations of standard sorting measures used in the literature and allows for the identification of pure sorting trends without imposing restrictive assumptions about the sorting process. I find that people in Brazil, Indonesia, Mexico, and South Africa tend to sort into internally homogeneous marriages based on education level. While educational sorting has a noticeable impact on household income inequality in any given year, changes in the degree of sorting over time barely have any impact on inequality. Further analysis reveals that this counterintuitive result is due to different dynamics within educational groups. The inequality-decreasing impact from reduced sorting among the highly educated is almost entirely offset by the inequality-increasing impact from increased sorting among the least educated. While it is certainly reassuring that concerns about educational assortative mating having a potentially large effect on income disparities between households appear to be unwarranted, these findings suggest another concerning narrative. Marginalization processes are occurring at low levels of educational distribution. The least educated are being left behind, facing limited labor market opportunities and diminished chances of achieving upward socioeconomic mobility through marriage to more educated partners.