Is Cohabitation Bad for Children? Assessing the Causal Impact of Legal Marriage on Child Outcomes Using a Siblings Analysis

Donna K. Ginther, University of Kansas
Marianne Sundström, Swedish Institute for Social Research

This paper examines children who live with both biological parents and analyzes whether parental marriage confers educational advantages to children relative to cohabitation. Cohabitation is more common in Sweden, and few studies have analyzed whether outcomes differ by parents' marital status. We control for the self-selection of marriage by estimating family fixed effects. Since siblings differ in age and over half of Swedish children are born to cohabiting parents who may or may not eventually marry, siblings will differ in the fraction of childhood lived with cohabiting and married parents. We use Swedish register data on over 100,000 full siblings born 1972 – 1980 and measure children’s educational outcomes by grade point averages (GPA) at age 16. Preliminary results indicate that children born to cohabiting unions have lower educational outcomes without controlling for family fixed effects. Future empirical results will determine whether legal marriage confers educational advantage.