Offre de travail des m res fran aises : l'effet d'une variation exog ne du nombre d'enfants

Between 1962 and 2005, whereas the activity rate of French men decreased, the activity rate of French women increased from 45,8% to 63,8%. However, women's activity rate remains correlated with the number of children: women with the lowest number of children are also the ones with the highest participation rate in the labour market. To what extent having an additional child reduces the mother's probability to participate in the labour market? The link between fertility and mothers' participation decisions is complex because they have joint determinants, and influence each other. Hard then to say a priori if the choice of working or not is the cause or the consequence of the decision of having a certain number of children. As Angrist and Evans (1998), we use a source of exogenous and random variation of fertility to measure the causal effect of fertility on French mothers' labour supply. As in the United States, we find that the probability of having a third child is higher among parents with same sex siblings, and that in this case, mothers' participation in the labour market is reduced. Because sex mix is randomly assigned and because it has an effect on participation only through its impact on the probability of having a third child, we produce instrumental variable estimates of the effect of having more than two children on mothers' participation in the labour market. We find that having more than two children reduces significantly the mothers' participation proobability and the hours worked per week. These results are confirmed by the use of twin second birth as the exogenous fertility shock. Also, our results indicate that having more than two children especially affects the labour supply of less graduated mothers but has no effect on fathers' labour supply