Age at Parents' Separation and Achievement: Evidence from France Using a Sibling Approach

This paper investigates the link between parental separation and children's achievement in adulthood. Using a French dataset on "Education-Training-Employment", I first estimate a random effects model and then examine the differences in age at divorce for children within the same family, to control for divorced family selection. Three outcomes are analysed: number of years of schooling, earnings-weighted education and social position. Using a random effects model, parental separation is linked to poorer educational attainment for their children, from 32% to 12% of a standard deviation lower where the number of years of education is concerned, and from 30% to 8% of a standard deviation lower where the earnings-weighted education is concerned. This effect varies with age: least affected are the 16 to 18-year-olds, and most affected are the youngest. Where social position is concerned, effects are weaker, but remain negative. Accounting for the family fixed effect yields somewhat weaker estimated effects for the youngest, but results remain similar. Parental separation is more detrimental to boys' education under both models, but conducting a F-test, we only reject the nul hypothesis for earnings-weighted education where family fixed effect is accounted for. In results from both models, teenagers who experience a parental separation are less affected if born after 1970, but differences are not statistically different from zero where the family fixed effect is accounted for.