Parents' education and child body weight in France: The trajectory of the gradient in the early years

- We examine the gradient in parental education and reported child body weight across childhood.
- Maternal high education is associated with a 7.20 percentage points decrease in child overweight.
- Our specification allows the evolution of the gradient to be non-monotonic with child age.
- The gradient follows an inverted U-shape with age: maternal high education is correlated with 5.30, 9.62, and 1.25 percentage points decreases in child overweight at ages 2, 8, and 17, respectively.
- The results suggest an equalization in body weight or a change in reporting style in youth.

This paper explores the relationship between parental education and offspring body weight in France. Using two large datasets spanning the 1991-2010 period, we examine the existence of inequalities in maternal and paternal education and reported child body weight measures, as well as their evolution across childhood. Our empirical specification is flexible and allows this evolution to be non-monotonic. Significant inequalities are observed for both parents' education - maternal (respectively paternal) high education is associated with a 7.20 (resp. 7.10) percentage points decrease in the probability that the child is reported to be overweight or obese, on average for children of all ages. The gradient with respect to parents' education follows an inverted U-shape across childhood, meaning that the association between parental education and child body weight widens from birth to age 8, and narrows afterward. Specifically, maternal high education is correlated with a 5.30 percentage points decrease in the probability that the child is reported to be overweight or obese at age 2, but a 9.62 percentage points decrease at age 8, and a 1.25 percentage point decrease at age 17. The figures for paternal high education are respectively 5.87, 9.11, and 4.52. This pattern seems robust, since it is found in the two datasets, when alternative variables for parental education and reported child body weight are employed, and when controls for potential confounding factors are included. The findings for the trajectory of the income gradient corroborate those of the education gradient. The results may be explained by an equalization in actual body weight across socioeconomic groups during youth, or by changes in reporting styles of height and weight.