Abstract
We examine the effect of fertility on marital stability in the UK using twin births and sex composition to tackle the endogeneity of family size. For women with children younger than age 12, we find little evidence that family size affects the probability of marital dissolution. For women with children at least 12 years or older, we do find evidence of a causal effect of children, but that this effect depends on how family size is increased.
Keywords:
JEL Classification:
Notes
1 Throughout the analysis, we focus on women whose first marriages began when they were 17–30 years of age. This includes 97% of all first marriages.
2 Probit estimates produce marginal effects that are similar to those produced by the linear probability model.
3 In these models, a boy first indicator was added as an additional control. The coefficient (SE) on the variable boy first is 0.001 (0.006) in the sample of young children and 0.004 (0.016) in the sample of older children.